& cplSiteName &

Did WorldCom Puff Up the Internet Too?

Light Reading
News Analysis
Light Reading
7/18/2002
50%
50%

The plot thickens. A number of experts are now charging that WorldCom Inc. (Nasdaq: WCOME) didn't just puff up its earnings with questionable accounting -- the company may have helped drive the Internet bubble itself with misleading traffic growth figures.

The new charges are important because the company's technical leaders, including John Sidgmore, the new CEO, were considered experts on the Internet and were frequently consulted on general Internet growth issues during the late 1990s. WorldCom's UUNET division runs what is considered one of the largest, if not the largest, Internet backbones in the world. Sources charge that Sidgmore and his technical team were responsible for inflating bandwidth growth numbers that supported much of the rationale behind the growth in the Internet.

Sources include several academic experts, as well as one former WorldCom employee who worked with both Sidgmore and former WorldCom Chief Scientist Michael O’Dell. The ex-WorldCom employee, speaking to Light Reading under the condition that he not be named, insisted that WorldCom executives including Sidgmore intentionally boosted internet traffic growth numbers to make the industry look more lucrative than it in reality was.

"If you do the math, all the growth they were claiming was physically impossible,” he says, referring to Sidgmore’s claims starting in 1998 that internet traffic was doubling every 3.5 months, or growing at a rate of 1,000 percent a year. “It’s been bullshit from day one... It was all about manipulating the stock market. In reality, what was growing was connectivity,” he says.

WorldCom's Internet statistics were often quoted throughout the industry, and Sidgmore has cited 1,000% annual growth in several public forums and reports. The controversy is important because Sidgmore has distanced himself from WorldCom's ex-CEO Bernie Ebbers and is casting himself as the leader that will clean WorldCom up.

Here's how it worked, according to the former WorldCom employee: WorldCom would hook up new customers with connections capable of handling, say, up to 1.5 Mbit/s of data, knowing that for most of the time the lines would only carry a fraction of this amount. WorldCom would then use the 1.5 Mbit/s figures, not the actual traffic figures, when citing Internet traffic growth statistics.

"There was massive connectivity growth, but UUNET’s business wasn’t growing as much, "says the former employee.

Several studies on Internet traffic growth, including one by Andrew Odlyzko and Kerry Coffman of AT&T Research Labs from 1998, show that Internet traffic was growing 1,000 percent a year over a short period of time between 1995 and 1996. But following that timeframe, studies show that Internet traffic growth remained fairly consistent, doubling every year -- a far cry from Sidgmore's now-famous 1,000% growth rate. A 100 percent growth rate might be a far cry from the huge numbers that WorldCom was touting, but, Coffman insists, it is still quite impressive.

"Doubling every year was still massive," Says Coffman. "People didn’t understand what growing at 10 times a year meant,” he says, talking of all the companies and investors who believed that such enormous growth could continue. “Nobody did sanity checks.”

Odlyzko questions the idea of the Internet's runaway growth as depicted by WorldCom.

"The myth of Internet traffic doubling every 100 days seemed to be based on (i) the fact that such growth rates really did hold during the two-year period 1995-1996, and (ii) WorldCom making misleading claims in subsequent years,” Odlyzko, now with the Digital Technology Center at the University of Minnesota, writes in an e-mail to Light Reading.

Other industry observers, however, say that they’re not so sure that the huge numbers were pure manipulation. “There were some fairly aggressive estimates of growth rates,” says Rick Wilder, principal scientist at Masergy Communications Inc.. “They may have been a bit exaggerated… [but] I think those numbers were factual for UUNet for a couple of years.”

Some experts say that in general, during the bubble years, measurements of Internet growth were subject to widespread abuse. Often, small snapshots of Internet growth were cited, without regard to long-term impact.

"Basically folks were using growth numbers that may have been true for a specific piece of the Internet, e.g., certain links on the NSFNET backbone, for a short period of time, and likely using them to their advantage when it would increase stock price projections (or egos),” Kimberly Claffy of the Cooperative Association for Internet Data Analysis, CAIDA, writes in an e-mail.

Of course Sidgmore and the rest of the WorldCom crew weren’t the only ones pushing the big numbers. “It is hard to hang it all on John Sidgmore,” says one analyst, asking to remain unnamed. “It was pretty widespread. It was definitely questionable, but I feel bad for Sidgmore to get stuck with all the blame. We all did it.”

The former employee, however, says that while much of the industry is guilty of wanting the numbers to be higher than they were, WorldCom was a leader in touting its Internet figures. He says this led to the company building out infrastructure it didn’t need to uphold the impression that traffic was growing as fast as it claimed it was. “They had decided to build x amount of ports each month,” he says, “whether there were customers for them or not.”

Was the Internet growth hype premeditated? It's hard to tell. It may be WorldCom executives thought the traffic was growing as fast as they claimed

One thing is for sure, the controversy points to a central problem in the industry -- that it's tough to get good Internet traffic statistics. Government-sponsored traffic numbers haven’t been released since the NSFNet was was decommissioned and Internet backbone services transitioned to the commercial sector in 1995, according to Claffy. Most carriers don’t even systematically measure their traffic, she says, and those that do use different and often dubious methodology, and are careful to keep the results close to their chest.

Neither Sidgmore nor O’Dell returned numerous calls and requests for comments.

Light Reading is planning a project that would collect actual traffic statistics from hundreds of Internet access lines to enterprise users - giving everybody a much better reading on what's actually happening to business traffic volumes on the Internet. Corporations will be given a free trial of a monitoring service in exchange for participating in the scheme (see Track Your Traffic).

— Eugénie Larson, Reporter, Light Reading
http://www.lightreading.com

(119)  | 
Comment  | 
Print  | 
Threaded  |  Newest First  |  Oldest First        ADD A COMMENT
52442
50%
50%
52442,
User Rank: Light Beer
12/4/2012 | 10:02:09 PM
re: Did WorldCom Puff Up the Internet Too?
Please contact me directly at 212-714-7906.
Steve Young, CNN, New York. Thanks.
erbiumfiber
50%
50%
erbiumfiber,
User Rank: Light Beer
12/4/2012 | 10:02:09 PM
re: Did WorldCom Puff Up the Internet Too?
Roy posted an e-mail address way back in post 90...
Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:02:35 PM
re: Did WorldCom Puff Up the Internet Too?
beowulf888: "Access traffic vs. core traffic will have an 800-to-1 ratio only if you consider the core network as being a single pipe. In reality, LOB is correct when he states that, since the core is a complex mesh, the traffic will drop off proportional to the logarithm of network size (as measured in hops). Unicast traffic has a path between the source and destination that traverses only a subset of the total possible network hops in the core."
________________________________________________

I believe that we continue to be talking "apples and oranges". I will discuss a network architecture model to demonstrate what I am talking about. I also want to show how easy it is to G«£manipulateG«• the perception of what G«£growthG«• is when we have been talking about it relative to the Internet. For those that donG«÷t want to wade through the network architecture model, skip down to point (B). (I also recognized as I was doing through this that I was calculating the physical plant bandwidth growth rather than the ratio of the growth but still using the growth ratio labels.)

Perhaps, you may also be talking about the actual effectively delivered data packets that will traverse the multiple architectures of the Internet, while I am talking about the physical plant that has to be in place and provisioned in the different levels of architectures in order to provide for the effective delivery of those data packets. The traffic growth might be at a one for one ratio. The physical plant would not.

What I have not really talked about and may not be understood is that the efficiency of the G«£facilitiesG«• and packet transmission bandwidth G«£usageG«• improves greatly as the data traffic moves toward the core of the network. It is this improved efficiency that is reflected by the G«£gainG«• factors at the different points in the network. I have tried to use a simplified model to express that and the architectural provisioning models that are used to size the provisioned facilities that are used to carry Internet data transmission services.

For those that have worked on very large campus LANs, this will be familiar territory. (I use very large campus LAN because it is a closed system and can be used to demonstrate model without a lot of esoteric G«£traffic modelsG«• that inexperienced people tend to use.) The campus is made up of multiple multi-story buildings. The physical links between the desktop systems today is often by Category 5 twisted pair cable. This cable will support Ethernet at full duplex 100Mb speeds. The applications that are used on the desktop systems seldom saturate that link for greater than a few ms at a time, which is seen as a percentage of utilization per second. The peek of that utilization is often reported at less than 50%, while the majority of the time, the reported utilization is close to 1%.

The links from the desktop systems are aggregated on a floor aggregation switch, which may support more than one floor of a building. I am going to refer to these switches as the first the level aggregation in this model. The links from the desktops to the first level switches, I will refer to as the first level links. Newer facilities will have non-blocking GbE switches with optical GbE between the first level aggregation switches and a building aggregation switch, or router. I will refer to the building aggregation switch/router as the second level of aggregation in this model. I will refer to the links between the first level aggregation switches and the second level aggregation switch/routers as the second level links. The different building aggregation switch/routers are linked by optical GbE in a semi-mesh architecture to provide redundancy between the buildings and any server farms that may be in the different buildings. I will refer to the links between the buildings as third level links.

LetG«÷s get back to architecture of the physical facilities. For simplicity, we will say that there are 6 buildings with 12 floors each with 100 desktop systems on each floor. To simplify the discussion, any server farms in any of the buildings will be terminal network G«£spurG«• off of the second level building aggregation facilities of each building. Each first level aggregation switch supports 2 floors. Each first level switch only has a single GbE link into the second level aggregation. The second level aggregation switch/routers have GbE links to three other second level aggregation switch/routers in different buildings.

The relationships of bandwidth at the different levels of links are defined by the granularity characteristics of the physical link technology at each level. The implementations at the second and third level links are based on not only on the granularity of the links, but the G«£bandwidth gainG«• that is provided at each aggregation level, and the architecture of the link deployments.

The total number of desktop links in each building is 1200 at an aggregate bandwidth of 120,000Mbps per building. The total amount of aggregate first level link bandwidth for the campus is 720,000Mbps. The total number of links between the first level aggregation switches in each building to the second level aggregation switch/routers is 6 at an aggregate bandwidth of 6,000 Mbps per building. The total aggregate bandwidth second level links for the campus is 36,000 Mbps. The total number of third level links between the buildings is 9, for an aggregate third level link bandwidth of 9,000Mps.

To relate to my Internet model, the first level links are equivalent to G«£AG«•, the second level links are equivalent to G«£BG«•, and the third level links are equivalent to G«£CG«•. In the very large campus LAN model, the ratio of A to B links is 20:1. The ratio between B and C links is 4:1. The in this model, total gain of the entire network is only 80:1. This is a very high quality network.

The actual data packet traffic that goes between the buildings, when measured at the different levels of the model, may remain at close to a one for one ratio because it is limited by the aggregate of the third level links in the model. The aggregate bandwidth of physical implementations at each level of the model is, however, very different. The third level traffic model does not include any traffic in the model that may never leave the second level of aggregation. This the fallacy of looking at the traffic models based on the traffic at the third level in the model and not recognizing the limitations created by the characteristics of the physical plant deployment and provisioning at the other levels of the model.

(B)
If I wanted to G«£growG«• the network, the lowest granularity that the first level links can support is 100Mbps. Adding additional desktops to each floor may not, in reality, effect the inter-building traffic as seen at the third level links. The ratio of the different levels might, however, remain the same. However, only when the growth at one level of the network is compared to itself does the ratio of growth of the physical remain the same for each of the levels, ie. the third level can be at a growth ratio of 2:1, the growth of the second level can also be at 2:1, and the growth of the first level can also be at 2:1. More often they are not.

Given the above model, if I were to increase the physical plant bandwidth for the third level links of the network, I would have to add 40Mbps of aggregate bandwidth at the first level of links for each single Mbps of bandwidth that I want to see in the third level of links. The growth of the first level, when compared to the third level would still be a 40:1 ratio. For each GbE link that I added to the third level links, I would have be seeing a growth of 40 Gbps, or 400 additional desktops per single building in the campus.

If I wanted to G«£manipulateG«• the perception of the growth of the network, numbers it would be easy for me use to the desktop connectivity count growth instead of the bandwidth growth of the network. I could also compound the growth at the different levels by saying that each of the three had a growth of 2X, making it a total of 6X. This would be G«£fictionalG«• of course, but if you are a senior manager or stock analyst and donG«÷t know how networks are actually built, you would not realize the fallacy.

If the architecture of the third level links were different, as a star into a single campus aggregation switch/router, there would only be single third level link out of each building. This would make the ratio between the aggregation of the first level links and the third level links to 720:1. This architecture is not only more expensive to implement, it provides a lower quality of service between the buildings in this model. Politicians tend to like this architecture because it G«£centralizesG«• the data network through the single campus aggregation switch/router. You might be surprised how many city wide metro enterprise networks are implemented this way. There are even some Internet facilities that are implemented this way. You can also see how easy it was for some people to manipulate the growth numbers for the Internet.

In the real world, for each Mbps of aggregate traffic growth that is seen on the backbone of the Internet, with a 400:1 ratio, the access would have to see an aggregate growth of 400Mbps. If I were to grow the Internet by 2Mbps, then I would have to see 800Mbps at the access. (This is where I was making a mistake in my earlier posts. Sorry about that everyone.)

Roy Bynum
beowulf888
50%
50%
beowulf888,
User Rank: Light Beer
12/4/2012 | 10:02:46 PM
re: Did WorldCom Puff Up the Internet Too?
Well I couldn't resist responding to flanker's troll...

flanker wrote:
>This is why toll quality VOiP over IP networks
>is an oxymoron: you cant manage the packets over
>the backbone.

You can run VoIP on IP networks with toll quality voice (well, MOS of 3.7 and greater) quite well, thank you very much. But you either have to make sure (A) that you give your voice traffic higher precedence bits than data, or (B) you build a parallel VoIP network. In both cases you have to keep bandwidth issues in mind when you design your VoIP network.

At Cisco, we had 30,000 employees world-wide connected by VoIP -- on the same backbone that our data was traveling over. Of course, the initial deployment was not glitch free, but, once the bugs were smoothed out, Cisco had a carrier-class voice that cost substantially less than their older circuit-switched infrastructure.

Also, China Unicom's pre-paid calling service network is all VoIP. It services over 300 cities in China, and we had better MOS scores than the China Telecom circuit-switched network (of course, that's not saying much in China ;-). In this case, the VoIP traffic was segregated on it's own IP network (so the only data traffic was from snmp, ntp, and routing updates). Guess what? We not only got better MOS scores than the incumbent carrier's circuit-switched infrastructure, but it cost 70 percent less to implement than the old-style big-iron circuit-switched network.

cheers,
--Wulf

beowulf888
50%
50%
beowulf888,
User Rank: Light Beer
12/4/2012 | 10:02:46 PM
re: Did WorldCom Puff Up the Internet Too?
Roy, although I like your math, I think you may be missing an important issue that LOB raised (or perhaps you're just simplifying your arguments for the non-math masses).

Access traffic vs. core traffic will have an 800-to-1 ratio only if you consider the core network as being a single pipe. In reality, LOB is correct when he states that, since the core is a complex mesh, the traffic will drop off proportional to the logarithm of network size (as measured in hops). Unicast traffic has a path between the source and destination that traverses only a subset of the total possible network hops in the core.

So in reality the 800-to-1 ratio for dial-up to core traffic is probably an order of mangnitude too small if you were to use dial-up traffic as indicator of how to scale your core's bandwidth.

best regards,
--Beo

lob wrote:
>Of course, the larger the network is, the
>more circuits a packet has to travel from
>source to destination. This is number of
>hops, or "diameter of the network". It is
>roughly proportional to logarithm of network
>size, and so introduces rather insignificant >correction to end-to-end delivery cost per
>bit.
PhotonGolf
50%
50%
PhotonGolf,
User Rank: Light Beer
12/4/2012 | 10:02:53 PM
re: Did WorldCom Puff Up the Internet Too?

Well, a very interesting thread. Took me a long time to digest!

Since most of us are not in the business of selling fiber (I hope!), I suspect most of us will still see business recovering as the core traffic continues to increase ... routers, transport, electronics, optics, software, services, etc...

So .. given all of this data, when will the long haul start to recover? (relates to the current utilization of existing systems)

Forget 40 .. care to guess how many 10G linecards (1 per lamda) will be consumed in 03, 04, and 05?

What will be the price points? ($20k today, falling).

At what point will it be cheaper for carriers to buy a new transport system (like Ceyba) instead of more linecards for Nortel? (Capex, Opex)

I suspect this will be a better forecast than you could ever get from RHK!

- P
Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:02:55 PM
re: Did WorldCom Puff Up the Internet Too?
lob: "We have network A, with backbone, access and everything - all properly scoped and loaded to capacity. Now, some day we got exactly 2 times more identical customers. We build _identical_ network B - backbone, access and everything. It is also loaded to capacity (because it is identical to network A, and customers are the same). Plus, we need to interconnect those networks."
_________________________________________________

lob,

I think I begin to see where there is a misunderstanding. We are describing two totally different network architectures.

The G«£AG«• in my examples are the access links only, not a network in and of themselves. The dial up access links are independent facilities until the output of the aggregation router at the ISP. Until then, they can be compared to individual Ethernet 100BaseT links between any two ports on any two different systems. It is not like a cable system or a 10Base5 LAN where the first level of aggregation, along with the first level of collisions/blockages, occurs within the G«£etherG«• of the coax facility.

The systems on the different links can not communicate with each other except after the aggregation and routing functions that are in place after the traffic is aggregated over a common G«£planeG«• in the ISP aggregation router. It is at that point that the traffic that originates on the links, G«£AG«•, becomes part of the 100:1 traffic that is pushed up to the NSP peering point to become part of the 4:1 traffic in the backbone G«£CG«•.

In my previous examples, the links G«£AG«• do not become part of a network until after they are part of G«£BG«•, not until the buffered output to the metro link to the Internet backbone. At that point, with the peering facilities between the ISP and the NSP, each access link is part of the Internet, not a separate G«£networkG«•.

For the purposes of this discussion it does no good to evaluate each network as a separate entity because it is the backbone bandwidth of the Internet that we are concerned with, not the additional aggregate bandwidth within each autonomous system, some of which may never get to the backbone. If there is additional traffic with a metro ISP that never gets to the Internet backbone, then the overall aggregate traffic for the metro networks is even higher than the G«£BG«• that I was using, which in turn increases the G«£AG«• as well.

lob
50%
50%
lob,
User Rank: Light Beer
12/4/2012 | 10:02:56 PM
re: Did WorldCom Puff Up the Internet Too?
Before I give up on math education, I'll make one more try.

We have network A, with backbone, access and everything - all properly scoped and loaded to capacity. Now, some day we got exactly 2 times more identical customers. We build _identical_ network B - backbone, access and everything. It is also loaded to capacity (because it is identical to network A, and customers are the same). Plus, we need to interconnect those networks.

Now we have 2x customers, and 2x backbone traffic. Cannot put more customers on A's backbone - it is already loaded. Gluing A and B together doesn't reduce traffic - the traffic patterns in both networks are the same.

Grows 2x in customers (with each customer offering the same load as older customers) yields
2x growth in backbone traffic, and that's it.

I can't explain it any simpler, sorry.

Your mistake is that you calculate absolute growth (which is in our example 400bps for 1bps in backbone) and then for some reason confuse it with relative growth (which is 2x in both backbone and access - because 800/400 = 2 - definitely _not_ 800x!).
Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:02:57 PM
re: Did WorldCom Puff Up the Internet Too?
lob: "The significance of that is that pure self-similar traffic has zero multiplexing gain, i.e. stochastical multiplexing does not reduce burstiness of traffic. Aggregation of several Poisson streams yields smoother traffic flow, thus
reducing the need for extra capacity needed to accomodate traffic bursts (this is known as multiplexing gain)."
________________________________________________

This is nice.

It is too bad that it does not work at "busy hour" within any one demographic facility environment. What you really wind up with, is that regardless of the "gain" you have, is an effective aggregate bandwidth that can never exceed the lowest common facility bandwidth.

Empirically, from watching a whole lot of data analyzers over many years, with a high number of statistical simultaneous attempts to us the same bandwidth within a specific "window" there are large numbers of "collisions" or queue access G«£blockagesG«•, which in turn causes losses of original data packets, which in turn causes the data packets to get retransmitted after a period of G«£random delayG«•, which causes more "collisions" and "blockages" that tend to G«£spikeG«• and G«£quietG«• until all the aggregate of data packets are all correctly transported, or the applications time out. The access links tend to reflect that G«£spikeG«• and G«£quietG«• effect. This is not due to just the delay between the users hitting the G«£sendG«• function in an application.

It is the effect of G«£blockagesG«• with the randomness of retransmission that creates the Poisson streams that you are referring to. The effect of smoother traffic flow is a result of a reacting to failures to transport real time combined with extensive buffering prior to a transmit queue at the egress of an aggregation router. This works very well for low performance applications and users that have a low expectation of delay performance, like dialup users on the Internet. It is the reason that there is a committed information rate that is directly reflected in backbone bandwidth for services that might have slightly higher expectations. It is the reason that there are exclusive use of the bandwidth facility for real time applications and performance expectations.

Voice steams are real time and can not be statistically buffered and delayed relative to the load a voice network. This is the reason that voice services G«£blockG«• calls during busy hour over loads.

This a distinction between voice networks that are sized directly for G«£busy hourG«• loads with exclusive use of the IMT facilities and data networks that are more often sized relative to the port density of the vendors routers, or the cost for transmission services that a budget will allow.
Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:02:57 PM
re: Did WorldCom Puff Up the Internet Too?
lob: "Ok. Before: A0 = 100*B0, B0 = 4*C0,
After: A1 = 100*B1, B1 = 4*C1, C1 = 2*C0

C0 = A0 / 100 / 4
A1=100*4*(2*C0) = 100*4*2*(A0/100/4) = 2*A0.

I.e. growth 2x in backbone matches growth 2x in
access. This is not even a university math, this
is middle school.
_________________________________________________

lob,

I begin to realize why I could not understand the accounting that was being presented by the WorldCom accountants at the last all hands meeting. It should not take a complex formula to figure out what a given growth in the backbone would require for growth in the access. I will do it G«£long handG«• with G«£long proofG«• as we learned in 8th grade algebra.

To find out what a 2X growth in the backbone would require for growth in the access:
C represents the given growth of the aggregate bandwidth in the backbone
B represents the aggregate bandwidth in the metro relative to the aggregate bandwidth in the backbone
A represents the aggregate bandwidth in the access relative to the aggregate bandwidth in the metro
Given: C = 2
B = 4*C
A = 100*B
Find A
Replace G«£BG«• with G«£4*CG«•: A = 100*(4*C)
Replace G«£CG«• with G«£2G«•: A = 100*(4*2)
First calculation: A = 100*(8)
Final calculation A = 800

Results: For growth 2X in the backbone, the access will have 800X growth.
lob
50%
50%
lob,
User Rank: Light Beer
12/4/2012 | 10:02:58 PM
re: Did WorldCom Puff Up the Internet Too?
Aw, forgot to put the disclaimer that I did not
made a correction for multiplexing gain dependence
on scale (relatively recent results show that
heavy-tailed distribution (aka self-similar, or fractal processes)) of packet rates observed
in the LANs and smallish networks gets a Poisson
component which grows with the scale of the network.

The significance of that is that pure self-similar
traffic has zero multiplexing gain, i.e. stochastical multiplexing does not reduce burstiness of traffic. Aggregation of several Poisson streams yields smoother traffic flow, thus
reducing the need for extra capacity needed to accomodate traffic bursts (this is known as multiplexing gain).

However, the multplexing gain factor is not that
large: 2x tops. Growth of the network above some level does not increase that factor much.

But, then, this is second-order effects which do not significantly alter fundamentals.
lob
50%
50%
lob,
User Rank: Light Beer
12/4/2012 | 10:02:58 PM
re: Did WorldCom Puff Up the Internet Too?
> Perhaps you can do the math for us. I want to
> determine how much growth it would take at two
> different points in the network to represent a
> specific growth at a third point in the
> network. Consider this a set of compounded
> linear sets:

> Point A is a ratio of 100:1 to point B,
> Point B is a ratio of 4:1 to point C,
> Point C is 2 (2X growth or 100% growth
> depending on how you represent it.)

> What is the factor and percentage at Points A
> and B?

Ok. Before: A0 = 100*B0, B0 = 4*C0,
After: A1 = 100*B1, B1 = 4*C1, C1 = 2*C0

C0 = A0 / 100 / 4
A1=100*4*(2*C0) = 100*4*2*(A0/100/4) = 2*A0.

I.e. growth 2x in backbone matches growth 2x in
access. This is not even a university math, this
is middle school.

Now, given that level of familiarity with basics
of algebra shown by _engineers_, why do we act surprised when investors couldn't put 2 and 2 together?

Anyone wants to challenge my result? :)
Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:03:01 PM
re: Did WorldCom Puff Up the Internet Too?
fiber_r_us: "The approach was to take a statistical sample of customers' leased lines at various locations from test ports on crossconnects. I believe they did the sampling during the workweek over 24hrs on some number of circuits. I also remember that they were looking at the busy-hour average (that is the average of the busiest hour from each of the monitored circuits). The vast majority of the circuits were running data (ie. PPP, POS, ATM, FR). For the data circuits, utilization is pretty easy to measure."
_________________________________________________

Thank you for this information.

For the ATM and FR data services, the circuits were probably fractional CIR because of the way that the services are sold. I have been told that the AT&T sales people were told to not sell CBR or full rate for those two services. An amount of stat gain is needed to break even on the cost of providing the services. Was this accounted for?

What was the sampling rate granularity? Was it reported as %use per second for each second in the test period? Was the utilization averaged over a 24 hour period, over an hour period? How was it reported? Did it report peak at busy hour? Were there multiple busy hours? (Most of the enterprise networks that I have worked with had between two and four separate busy hour periods each day.) Was the non-busy hour utilization reported?

Given my experience with the difference between what the transmission people reported for utilization on my circuits versus what my systems reported the utilization to be, please forgive me if I am suspicious of the numbers that were reported by the AT&T transmission people. I have found that the way that transmission people view a facility and monitoring that facility and the way that data people view the same thing are very different.
Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:03:04 PM
re: Did WorldCom Puff Up the Internet Too?
lob: "I'm not sure what school you've attended, but it surely wasn't a good place to study math. If x/y=n (x is "connecivity", or customer capacity, and y is "bandwidth", or backbone capacity, and n is the scaling factor, for example 8), then 10x*n = 10y. I.e. backbone capacity is linearly proportional to end-user capacity."
_______________________________________________

Perhaps you can do the math for us. I want to determine how much growth it would take at two different points in the network to represent a specific growth at a third point in the network. Consider this a set of compounded linear sets:
Point A is a ratio of 100:1 to point B,
Point B is a ratio of 4:1 to point C,
Point C is 2 (2X growth or 100% growth depending on how you represent it.)
What is the factor and percentage at Points A and B?
fiber_r_us
50%
50%
fiber_r_us,
User Rank: Light Beer
12/4/2012 | 10:03:07 PM
re: Did WorldCom Puff Up the Internet Too?
Roy said:
When you talk about the utilization of private line circuits in the AT&T network, what was the test period, what was the granularity of the test results, and how could the contents of the leased circuit private line circuits be determined without invading the customersG«÷ leased circuit to monitor the contents?

____________________________________________

The approach was to take a statistical sample of customers' leased lines at various locations from test ports on crossconnects. I believe they did the sampling during the workweek over 24hrs on some number of circuits. I also remember that they were looking at the busy-hour average (that is the average of the busiest hour from each of the monitored circuits). The vast majority of the circuits were running data (ie. PPP, POS, ATM, FR). For the data circuits, utilization is pretty easy to measure. I believe they ignored the few voice circuits that they found, but, since they were far in the minority for customer circuits, it doesn't really matter.
LightSeeking
50%
50%
LightSeeking,
User Rank: Light Beer
12/4/2012 | 10:03:07 PM
re: Did WorldCom Puff Up the Internet Too?
To add to my last post, there does not have to be a clear relationship between the growth in connectivity itself and the growth in overall bandwidth: I understand connectivity to mean distribution links and points that depends on the network topology and ongoing provisioning of new clients.

Perhaps, LOB is commenting on the bandwidth in all the connections??

LS
LightSeeking
50%
50%
LightSeeking,
User Rank: Light Beer
12/4/2012 | 10:03:09 PM
re: Did WorldCom Puff Up the Internet Too?
Roy, lob:

While I'm not an expert in network architecture, it seems that the growth in connectivity could be bigger than the growth in overall bandwidth, if you want to build redundancy in the network.

I once heard from a network guru from a giant networking company that there was a lot of redundancy build into AT&T's phone network.

Any comments?

LS
-----------------------------------------
> That means that it takes a 800% growth in
> connectivity to show a 100% growth in overall
> bandwidth.

I'm not sure what school you've attended, but it surely wasn't a good place to study math. If x/y=n (x is "connecivity", or customer capacity,
and y is "bandwidth", or backbone capacity, and n is the scaling factor, for example 8), then
10x*n = 10y. I.e. backbone capacity is linearly proportional to end-user capacity.

lob
50%
50%
lob,
User Rank: Light Beer
12/4/2012 | 10:03:11 PM
re: Did WorldCom Puff Up the Internet Too?
Roy Bynum wrote:

> In reality, the G«£growth of the Internet was
> limited more to connectivity than bandwidth.
> This can be seen when you recognize that
> most dial up ISPs have a statistical gain
> of 100:1 between their customers and the
> backbone, and the NSPs like Internet MCI,
> UUNet and others have a 4:1 gain between
> the ISP and backbone.

The figure 100:1 is max. customer capacity vs
backbone link-up capacity. This incorporates
low duty cycles common for dial-up customers.
The high-duty cycle customers moved on to DSL
and cable, so that figure is hardly representative.

4:1 seems to be valid. Oversubscription (this is the technical term :) 3:1 to 5:1 is common.

> That means that it takes a 800% growth in
> connectivity to show a 100% growth in overall
> bandwidth.

I'm not sure what school you've attended, but it surely wasn't a good place to study math. If x/y=n (x is "connecivity", or customer capacity,
and y is "bandwidth", or backbone capacity, and n is the scaling factor, for example 8), then
10x*n = 10y. I.e. backbone capacity is linearly proportional to end-user capacity.

Of course, the larger the network is, the more circuits a packet has to travel from source to destination. This is number of hops, or "diameter of the network". It is roughly proportional to logarithm of network size, and
so introduces rather insignificant correction to end-to-end delivery cost per bit.

> This is easy to see in retrospect, but hard to
> see unless you are one of those insiders that
> are developing and implementing the
> connectivity/backbone technologies.

In restrospect, your statements still make no sense, sorry.
flanker
50%
50%
flanker,
User Rank: Light Beer
12/4/2012 | 10:03:11 PM
re: Did WorldCom Puff Up the Internet Too?
"AT&T did a study a while back that showed that thier private lines had a 4% utlization. Since private lines can't be oversubscribed through the backbones, the backbones are significantly overbuilt from a capacity standpoint. Of course, the customer pays for the bandwidth whether they use it or not, so why should the carrier care!"

On the contrary, the only way to make money is to allocate 10x the access capacity to 1x backbone capacity. This is a standard metric for second tier data carriers and ISPs.

This means I run 200 to 240 Internet subscribers per T1 line, not 20 to 24. Dedicated access is another story, but few consumers pay for dedicated access. Even when a corporate user buys dedicated IP access, they invariably buy less than they need because dedicated access is expensive.

Speaking of voice/data split, you can't run toll quality voice and data over the same logical channel, so the comparison is useless. While backbone networks carry far more data than voice traffic, from a traffic engineering perspective, you still need to buffer voice over the backbone and allocate more bandwidth than would be necessary under ideal conditions. This is why toll quality VOiP over IP networks is an oxymoron: you cant manage the packets over the backbone.









Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:03:15 PM
re: Did WorldCom Puff Up the Internet Too?
fiber_r_us: In fact, this is exactly what happens in the private-line space from a capacity standpoint. AT&T did a study a while back that showed that thier private lines had a 4% utlization.
_________________________________________________

In 1995 we deployed a transmission surveillance network. We were forced to use the Frame Relay network for political reasons. When we indicated that a fiber cut would cause a massive spike in bandwidth we were scoffed at. When a fiber cut did occur and the bandwidth utilization figures came back, were astounded to find out that we only had a 3% utilization for almost a Megabyte of data over 512kb links. We got to asking questions and found out that the utilization measurements were taken over a 24 hour period. The lack of granularity totally hid any bursts of traffic or utilization that might have occurred.

When we went to leased circuit T1s, the data services organization could not track the utilization so we relied on the routers to tell us. During quite periods, the normal background traffic was about 3% utilization. At the time of a fiber cut our utilization went to 60% for a very short period. When we showed these numbers to management, they went back to the Frame Relay people to improve the performance reporting. We found that their systems could not easily report a better than 24 hour granularity because of the massive processing requirements needed. We settled for being able to do it on individual customersG«÷ services on a one by one basis. It took an external network analyzer to do that.

When you talk about the utilization of private line circuits in the AT&T network, what was the test period, what was the granularity of the test results, and how could the contents of the leased circuit private line circuits be determined without invading the customersG«÷ leased circuit to monitor the contents?
Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:03:16 PM
re: Did WorldCom Puff Up the Internet Too?
In early 2001 Fred Briggs held an G«£all hands meetingG«•. At that meeting, we were told that the revenue from data services had already exceeded the revenue from voice services. He talked about how the margins from data services such as the Internet were not anywhere near what they were for voice or for other data services. I asked him later, where things like leased facilities, wholesale bandwidth, and private line fell within the service model that he was talking about. He said that these services were considered to be part of G«£dataG«•.

Since the revenue per transmitted bit is much greater for voice than it is for something like private line, it should be safe to assume that the bandwidth for data services had already exceeded the bandwidth for voice some time ago, at least for MCI/WorldCom.

I recently did a presentation about the ratio of packet data bandwidth in the backbone to leased circuit bandwidth. A former AT&T employee that was at that presentation, indicated that AT&T had similar service bandwidth ratios.

While these ratios may not be the same for the local service providers, I believe that they would be close for all of the legacy long haul service providers. The legacy free providers that only offer packet type services will, of course, have different numbers. They may also be siphoning off some of the packet services from the legacy providers which would change those service numbers.

The bottom line is, even if the revenue from voice and data services are close, the bandwidth from data services is greatly exceeding the bandwidth from voice services.
fiber_r_us
50%
50%
fiber_r_us,
User Rank: Light Beer
12/4/2012 | 10:03:16 PM
re: Did WorldCom Puff Up the Internet Too?
Stat-muxing on dial-up modem banks is not a *bad* thing and has almost nothing to do with QOS. Consider that most modern modem systems have 500 to 2000 modems in a chassis that is connected to an Ethernet switch with 100Mbit Ethernet. Even in the 2000 modem case, that is 2000x56k =~ 100Mbit if all modems were maxed-out. The 100 to 1 oversubscrption that Roy is talking about is *observed* oversubscription. That is, the 100Mbit Ethernet rarely peaks over 1Mbit utilization because there is no *offerred load* greater than that! QOS would suffer if the 100Mbit Ethernet port on the modem bank became overloaded.

At the ISP/NSP level, the same is true. If the NSP has an edge router with ten 1-Gbit links to customers, and an OC-48 uplink to core backbone routers, yet the OC-48 never experiences congestion, is this 4-to-1 oversubscription *bad*?

Oversubscription is only a problem if congestion develops due to the oversubscription. Any reasonable ISP/NSP operator closely monitors individual link utilizations and packet queues/drops on those links in order to ensure that congestion is not occuring.

To suggest that one should not oversubscribe is a sure way to bankruptcy for any provider.

Roy's points about calculating backbone bandwidth simply by summing-up access-port capacities as being off by a factor of ~800 is an accurate portrayal of real networks. Without oversubscription, backbones would be built out hundreds of times too large!

In fact, this is exactly what happens in the private-line space from a capacity standpoint. AT&T did a study a while back that showed that thier private lines had a 4% utlization. Since private lines can't be oversubscribed through the backbones, the backbones are significantly overbuilt from a capacity standpoint. Of course, the customer pays for the bandwidth whether they use it or not, so why should the carrier care!
Light-bulb
50%
50%
Light-bulb,
User Rank: Light Beer
12/4/2012 | 10:03:18 PM
re: Did WorldCom Puff Up the Internet Too?
Your more than Welcome Lob on the correction. We all do it from time to time...
Very well written and I do see some of your points. However understand that I was talking theoretical bandwidth. In your post you mention multiple connections etc. While you are right, there are numbers that you didn't quite take into account. Namely TDM traffic.
From the Post I take that you are anticipating all traffic to hit multiple Hubs. That doesn't have to be the case especially in TDM. Transient traffic is everywhere however in a properly engineered network transient traffic should stay as low as possible to avoid repeated signal and avoid re-transmission. In your post I take that you believe from A-D it must traverse B and C. This doesn't have to be the case. I'm not only referring to Internet Backbone, though If I was only referring to IP traffic your post is Dead on. I'm talking about usable bandwidth between end-points of the fiber. I still believe that number to be ~5000Pbps and perhaps at much as double based on current technology. As far as distance and capacity, understand that I did not take into account the dispersion sensitive(today) 40Gbps (OC-768) 10Gbps can easily be pushed down the fiber especially Level-3s relatively young strands. To get to around 5000Pbps we would need basically (Based on Today's Technology) 5000Pbps/1.76Tbps=~2840 pairs of Fiber. Thats not much. If we just evaluated two end-points... Lets say Denver --> Salt Lake how many strands do you think exist between the two points? I believe that Just Level-3 has ~430 Pairs in those conduits. How about Qwest? MCI? ATT? Global Crossing? I think its safe to say that just between those two end-points at least 1000 pairs of fiber has been laid, already 1/3 of the way to 5000Pbps. Think if we added the fibers laid from DC to NY. In all reality if we use new technology to drive even more bits from our fiber We will have enough Fiber for many many moons is it a 1000 years I don't know... Your calculations failed to take that into effect. Yes while traffic increases your calculations were based on 10g for 35 years. We may see OC-3072 by then. :) For as surely as the traffic increases so does the available speeds for the glass.

I do disagree with you on your final point though Lob. PCs today (General PCs) can not fill 100Mbps. They may have a capability to burst to 100 but to sustain? No way. For even if everyone had a 100Mbps pipe to their home they wouldn't know how to fill it. Also remember that Mpeg-4 allows nearly DVD quality streams @ 1.5Mbps. Lets not even Discuss additional Codecs such as VP5. What I'm saying is the typical 100Mbps connection would be able to sustain 64 DVD quality Concurrent video feeds! I believe Video is the only way we will be able to even come close to saturating the bandwidth. And if we choose to build efficient transport systems to our end-users only the channel they watch will be casted out to them. (Multicast Group Join)
Honeslty you don't think the average PC can fill 100Mbps with Internet surfing? With Grid computing your getting closer but good grief data-blocks don't need that much bandwidth. Of course that all depends on what we are processing. [email protected] comes to mind. Arguably the most succesful Distributed system. People with dial-up have no problems.
Very good discussion though its nice having a decent conversations with an educated person. You should re-run your numbers taking into account Multiple Point A-B-C-D links, and also taking into account link speed migration... OC-192 -->OC-768 -->OC-3072 etc. Don't forget in that calculation the channel increase in the DWDM system. 168 @10G Channels=1.68Tbps -->80x40G 3.2Tbps etc...

Thanks for the post,

Cheerio,
flanker
50%
50%
flanker,
User Rank: Light Beer
12/4/2012 | 10:03:18 PM
re: Did WorldCom Puff Up the Internet Too?
It amazes me how some of the posters on this thread allow their hatred for venture capitalists/entrepreneurs/lawyers/investment bankers/politicians to accept hearsay as fact and just throw objective evidence out the window.

No wonder the market bubble happened in the first place.

cripes.
erbiumfiber
50%
50%
erbiumfiber,
User Rank: Light Beer
12/4/2012 | 10:03:19 PM
re: Did WorldCom Puff Up the Internet Too?
THANKS! Anyone who can say with a straight face that, as Vice-Chairman of the Board of Directors, he was not "around" during the accounting frauds lacks ALL credibility. And since he is very proud of all his work going around to conferences and analyst meetings (apparently, according to him, his "only job" when he was Vice-Chairman), he should be only to happy to support and "prove" that his numbers were "correct" (maybe merely "agressive accounting" for Internet bandwidth? LOL). But this topic is a little too esoteric for congressional investigations, unfortunately. Anyway, he has more to be concerned about from hostile laid-off WCOM employees and burnt shareholders than he does from Congress...will read your upcoming report with great interest...

Regards,
erbiumfiber
Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:03:20 PM
re: Did WorldCom Puff Up the Internet Too?
Beowulf888:"However, you don't seem to take into account the large number of web-surfers sitting at their desks in corporate environments (who have T-1 or greater connections to the ISPs). I would think that 100:1 ratio wouldn't necessarily hold up for corporate customers."
_________________________________________________

Actually, depending on the company business model, there could be a much greater than a 100:1 ratio of user bandwidth to Internet usage bandwidth. Except for those sites that are directly Internet based service related, such as a "support" center, or a content provider, such as a "porn" site, the vast majority of the user traffic does not go out to the Internet.

In your case, you sited a situation where the peek of Internet traffic occurred during lunch. This indicates that your company did not have any directly Internet related services or requirements. The vast majority of the Internet "users" were people that were staying at their desks during lunch. In summer time, more people went out to lunch than stated in. (You may have been in a Northern, or G«£rainyG«• part of the country.) This is what is referred to as a G«£busy hourG«• demographic pattern. During the rest of the day, the data communications activities, for the most part, were not Internet related. The growth that you sited, about 5% per year, was more likely related to growth in number of people working at your company than anything else, because the nature of the applications used over the Internet has not changed very much for the last few years.

The way that people use the Internet, or any other functionally data system, is referred to as G«£operant behaviorG«•. It is this operant behavior that is the primary determinant of Internet usage and bandwidth. This behavior creates different busy hour demographic patterns for different business and residential service environments. The different demographic environments produce different busy hour patterns, many with different G«£hoursG«• for the busy hour peaks. These different busy hour peaks are another reason that there is so much statistical gain in Internet access. While a person is at work, they will not be using their access from home, and while they are at home, they will not be using it from work. Even though there may be multiple points of connectivity for any one individual, they can not use all of them at the same time. This is why that the aggregation of business Internet access and residential access needs to be discounted as well.
skeptic
50%
50%
skeptic,
User Rank: Light Beer
12/4/2012 | 10:03:20 PM
re: Did WorldCom Puff Up the Internet Too?
In talking with ISP engineers, I have found that that level of statistical gain is very common. Some dial up services push that gain up to provide a lower cost service while remaining profitable. Other services charge more to the retail customer for bandwidth and can provide a service that has less gain in the access.
-------------------

What you say is correct. There
services who play with oversubscription based
on statistical gain.
And further, they will almost never talk about
that number in public or disclose what it is
to a customer. They don't want to compete on
it, even though in many cases its a very good
predictor of the quality of their service.

And its important (in my opinion) to note
that making assumptions about the statistical
gain in network design doesn't always work
out. If you guess wrong, you can end up paying
a huge price for it (losing customers and
buying new equipment).

fknight
50%
50%
fknight,
User Rank: Light Beer
12/4/2012 | 10:03:21 PM
re: Did WorldCom Puff Up the Internet Too?
Rational analysis of Internet growth was available, even at the height of the mania, if you knew where to look and who to listen to.

In Jan 99, BCR published an article by Peter Sevcik, entitled "The Myth of Internet Growth" which challenged Sidgmore's assertions about growth. Peter's analysis found that demand was doubling every 11 months and that capacity was doubling every 10 months.

As to the implications of his findings Peter wrote: "Let's assume you're putting together a business plan for an Internet-based venture, and you rely on Sidgmore's predictions that the Internet will double in size every 3.5 months. If he's wrong, even by a little about -- let's say doubling takes 10 months instead...-- 95 percent of your market won't be there when you expected. That makes for a nasty surprise."
(http://www.bcr.com/bcrmag/1999...

Peter followed in his column for our Jan 2001 issue, which took another look at the mismatch between bandwidth supply and demand, and again he took a shot at Sidgmore's claims. He wrote: "The Intenet has not been doubling every 3.5 months, despite what you might hear.....But the myth continues because it's useful -- it helps get money out of investors and keep stock prices high."
(http://www.bcr.com/bcrmag/2001...

Peter's coming at this topic again in our Sep issue, so stay tuned.

Fred Knight
Editor/Publisher
BCR
beowulf888
50%
50%
beowulf888,
User Rank: Light Beer
12/4/2012 | 10:03:22 PM
re: Did WorldCom Puff Up the Internet Too?
Roy:
I see. So Even though 100 customers would be dialed in at 56Kb/s (for an aggregate bandwidth of 5.6Mb/s), statistically as an aggregate they'd only be generating an average of 56Kb/s of traffic on the dial-up ISP. Yes, that makes perfect sense. And the 4:1 ratio for ISP-to-NSP traffic would follow.

However, you don't seem to take into account the large number of web-surfers sitting at their desks in corporate environments (who have T-1 or greater connections to the ISPs). I would think that 100:1 ratio wouldn't necessarily hold up for corporate customers. Depending on the ISP's percentage of corporate bandwidth vs home bandwidth, I would expect that this ratio would be less than a 100:1 (probably closer to 50:1 for corporate connections?). Still it's a big drop as you move inward towards the core.

Back in 96, when I was a Network Engineer at a medium-sized company, I watched the peak traffic to and from our ISP increase by 5 percent a month (less in the summer months, more during the rest of the year). Natually, this traffic was also bursty (diminishing to almost nothing in the wee hours of the AM, and peaking during lunch hours). I could believe the estimates that the Internet usage was doubling every year. But of course, what one sees at the edge isn't necessisarily reflected in the core.

Thanks, Roy!
--Beo

Roy Bynum wrote:
>beowoulf888: "Could you flesh out your 'statistical gain' argument a little more?"
_______________________________________________

>In 1995 I was part of group of peole working on the Multiple Integrated Access Processor, or MIAP. The MIAP was actually a bank of modems/routers that were directly linked to a service termination point of a Class 5 switch. The MIAP included access authenication and other secuity, monitoring, and metering functions. It was designed to support the dialup access for several different retail ISPs. It also provided for intermachine truncks between the Internet side interfaces of the service access routers to what ever NSP service the ISPs. One of the interesting aspects of the MIAP was that the customer facing side bandwidth was 100 times the Internet service facing bandwidth. This creates a statistical gain between the customer and the service IMT. This is based on the premise that a single customer's utilization is spike bursty which can be aggregated with other customers' bandwidth to produce a more even level of bandwidth utilization. For access Internet traffic this works very well. The aggregate retail customer access bandwidth of 5.6Mb would translate to only 56Kb from the ISP to the core NSP.

>In talking with ISP engineers, I have found that that level of statistical gain is very common. Some dial up services push that gain up to provide a lower cost service while remaining profitable. Other services charge more to the retail customer for bandwidth and can provide a service that has less gain in the access.

>Between the ISP and the Internet core there is the NSP peering router. Even with the smoothing effects of the statistical gain at the ISP, IMT utilization is still somewhat bursty. The aggregate port bandwidth on the ISP access side of the NSP access routers is often 4 times that of the port bandwidth on the core transmission side. This provides an additional 4:1 statistical bandwidth gain between the retail customer and the core of the Internet. The 5.6Mb aggregate bandwidth from the Internet retail customers now becomes 14Kb in the core.
Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:03:23 PM
re: Did WorldCom Puff Up the Internet Too?
erbiumfiber - Please contact me directly at [email protected]
erbiumfiber
50%
50%
erbiumfiber,
User Rank: Light Beer
12/4/2012 | 10:03:23 PM
re: Did WorldCom Puff Up the Internet Too?
I was laid off in April of 2002 as my company quietly died- "consulting" now and may go back to school this fall- see my second post to you about the "God box" thing- I think you and I were on opposite sides of the customer/vendor transaction at about the same time...(except I am a patent attorney and function quietly behind the scenes...)
Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:03:24 PM
re: Did WorldCom Puff Up the Internet Too?
erbiumfiber: "So are you still a WCOM employee? Or have you left for greener pastures?"
______________________________________________

With the completion of the development of Ethernet service deployment technology project, I was downsized by WorldCom in November of 2001. I have yet to find steady work. I hope others in the industry are having a better time of it than I am.
Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:03:24 PM
re: Did WorldCom Puff Up the Internet Too?
beowoulf888: "Could you flesh out your 'statistical gain' argument a little more?"
_______________________________________________

In 1995 I was part of group of peole working on the Multiple Integrated Access Processor, or MIAP. The MIAP was actually a bank of modems/routers that were directly linked to a service termination point of a Class 5 switch. The MIAP included access authenication and other secuity, monitoring, and metering functions. It was designed to support the dialup access for several different retail ISPs. It also provided for intermachine truncks between the Internet side interfaces of the service access routers to what ever NSP service the ISPs. One of the interesting aspects of the MIAP was that the customer facing side bandwidth was 100 times the Internet service facing bandwidth. This creates a statistical gain between the customer and the service IMT. This is based on the premise that a single customer's utilization is spike bursty which can be aggregated with other customers' bandwidth to produce a more even level of bandwidth utilization. For access Internet traffic this works very well. The aggregate retail customer access bandwidth of 5.6Mb would translate to only 56Kb from the ISP to the core NSP.

In talking with ISP engineers, I have found that that level of statistical gain is very common. Some dial up services push that gain up to provide a lower cost service while remaining profitable. Other services charge more to the retail customer for bandwidth and can provide a service that has less gain in the access.

Between the ISP and the Internet core there is the NSP peering router. Even with the smoothing effects of the statistical gain at the ISP, IMT utilization is still somewhat bursty. The aggregate port bandwidth on the ISP access side of the NSP access routers is often 4 times that of the port bandwidth on the core transmission side. This provides an additional 4:1 statistical bandwidth gain between the retail customer and the core of the Internet. The 5.6Mb aggregate bandwidth from the Internet retail customers now becomes 14Kb in the core.
erbiumfiber
50%
50%
erbiumfiber,
User Rank: Light Beer
12/4/2012 | 10:03:24 PM
re: Did WorldCom Puff Up the Internet Too?
Another fabulous post! You wrote:

>>The G«£Internet cadreG«• had gained a lot of political power because it had been predicted that they would become the dominant service, so a lot of resources were being fed into developing what was called G«£multi-serviceG«• access technology. Before a lot of real cost/revenue modeling had time to be done, the G«£industryG«• considered a cost effective way to provide voice, private line, and packet based Internet services over the same common channel facility. It was only after there had been time to get into the reality of the economic modeling, that it was found out that it would not work. (G«£That is a whole G«ˇnother story.G«•) By then, much of the damage had been done to create the G«£bubbleG«•.<<

Sounds like the "God box" solution- I was associated with a (now-defunct) start-up that "bought" that vision and WCOM was a major target customer. As you point out, the whole "multi-service" access "industry" fell out of favor- we switched (oops! no pun intended) to developing another part of our system that we had been focusing on in the hopes of catching that market, which also soon evaporated. "Research" firms, such as RHK, also fueled this whole industry vision. While supposedly "independent" it was always amazing how the results ended up being exactly what the recipient wanted to hear in terms of projected direction of the industry. As you say, plenty of blame to go around.
erbiumfiber
50%
50%
erbiumfiber,
User Rank: Light Beer
12/4/2012 | 10:03:24 PM
re: Did WorldCom Puff Up the Internet Too?
An amazing post! Unfortunately, that could be the chronicle of some of my experiences at AT&T (prior to its Lucentification) and probably mirrors those of other employees in large organizations where politics trumps reality. For me, it was WDM (it had not yet become DWDM yet). I watched as the more politically powerful group (AWGs) trumped in-fiber Bragg gratings for stupid political reasons; I left to join a start-up and never looked back...

You ARE SO RIGHT about people believing their own spin and hype. You didn't mention Sidgmore but from all we have seen from him these past few months, he is VERY IMPRESSED with himself. I can only imagine how much fun it would have been to deal with him as a UUNET person (and, of course, as he himself asserted, he wasn't "there" because he was always traveling to conferences and analyst meetings- the better to hype the Internet...).

So are you still a WCOM employee? Or have you left for greener pastures? I'll go back and look at your earlier posts.
beowulf888
50%
50%
beowulf888,
User Rank: Light Beer
12/4/2012 | 10:03:25 PM
re: Did WorldCom Puff Up the Internet Too?
Skeptic:
Very cogent point...
"- I think that there is far less 'router' capacity
than 'transmission' capacity. The fiber (and
the DWDM multiplier effect) have made more fiber
capacity available than be used with the existing
number of core routers. (i.e. growth for routers,
no growth for fiber/DWDM)."

...Supported by my router contacts at the 800lb Gorilla, who say that their core routers are still selling (but not at the pace they were during the boom). While my optical contacts in the industry are seeing a much worse sales environment. Of course, that 800lb Gorilla gets a much smaller percentage of its revenue from its optical business. So despite the over-hyped growth of the Internet, routers seem to be better business to be in than optical switches. I predict we'll see the 800lb Gorilla bulk up some more after this recession.

--Beo
beowulf888
50%
50%
beowulf888,
User Rank: Light Beer
12/4/2012 | 10:03:25 PM
re: Did WorldCom Puff Up the Internet Too?
Hi, Roy:
Could you flesh out your 'statistical gain' argument a little more? I'm not really sure what you mean when you say "most dial up ISPs have a statistical gain of 100:1 between their customers and the backbone." And likewise I'm not sure what you're trying to say when you state: "NSPs like Internet MCI, UUNet and others have a 4:1 gain between the ISP and backbone." What do you mean by 'gain'? 100:1 vs. 4:1 gain in traffic?

I agree with your analysis of the 'Internet Cadre' and how they affected (manipulated?) the market outlook. However, I'd also like to point out that these sort of technology/marketing fads were not restricted to the Internet. I remember the ATM-is-the-future hype of the early 90s, and I got sucked up into it myself. I remember going to ATM conferences where TCP/IP, and every other non-ATM protocol, were called "legacy protocols" (as if they had already died). And the Internet was "a toy". Well, ATM turned out to be a very useful technology, but it didn't take over the networking universe. Likewise, TCP/IP and the Internet have changed the face of computing and the way we do business -- but it's clear that many of the predictions that the IP-everywhere gurus will never come to pass.

--Beo

Roy_Bynum wrote:
"In reality, the growth of the Internet was limited more to connectivity than bandwidth. This can be seen when you recognize that most dial up ISPs have a statistical gain of 100:1 between their customers and the backbone, and the NSPs like Internet MCI, UUNet and others have a 4:1 gain between the ISP and backbone. That means that it takes a 800% growth in connectivity to show a 100% growth in overall bandwidth. This is easy to see in retrospect, but hard to see unless you are one of those insiders that are developing and implementing the connectivity/backbone technologies.
Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:03:25 PM
re: Did WorldCom Puff Up the Internet Too?
In a previous post I mentioned that I was a MCI employee in Richardson, Texas, when WorldCom took over. I was working on IP over DWDM and Ethernet optical networking and the architecture of optical switched networks. This type of project is considered to be a three to five year project before any new services would be deployed and start being sold to customers.

At first, a lot of this work was based on knowing what our customers were asking for and were willing to pay for and how services and applications evolve over a long period of time. I had spent a few years working with business markets and had access to the market research that MCI was doing and was paying others to do. One of the first things that I and several other G«£MCI old timersG«• noticed was that WorldCom shut down a lot of the business markets and market research efforts. Also, third party market research was no longer available. WorldCom was investing in stock market analysis and meetings with stack market analysts. (You can guess where their priorities were.)

In order to find out what was going on, I had to do my own internal research. I went to the provisioning engineers and the internal service deployment people to find out what was really going on. From this I could determine what customers were actually buying, at what bandwidths, and in what quantities. This worked for everything, voice, private line, Frame Relay, ATM, managed services, etc, except the Internet services.

For internet services, I had to deal with the people in Virginia. The people at VBNS, what remained of the old Internet MCI, were very cooperative. We even had one of their engineers to come to Richardson to do original research on the inherent characteristics of optical Ethernet as an IP service transmission technology. This group did not last very long under WorldCom.

Every time that I tried to get some real world information from UUNet, I got G«£the run aroundG«•. This was particularly true of anything that was Ethernet based. I had contacts in Canada that were telling me that UUNet was using GbE for access and peering connections. I had long time friends that were working for vendors that gave me copies of documentation from UUNet that showed the use of Ethernet for access interconnects and intraconnects. For a long time, UUNet denied the use of Ethernet as a service technology. It made me suspicious of everything that came out of their engineering group, including their support for MPLS.

The political power of UUNet within WorldCom was immense. All the transmission network architecture and planning people got out of UUNet were bandwidth requirement predictions that were in the 8X to 12X range. Most of the more realistic transmission architecture people scaled that back to 4X. Even at that, it was easy to see that the Internet service alone would totally dominate the transmission facilities by 2005. The problem was that the revenues that UUNet was reporting (we find out now that those could have been inflated as well) were only growing at 2X, so it would have been about 2004 that the current SONET/SDH technology was too expensive to support the continuation of the service growth.

Since Internet services are over a common channel facility, the work that I was doing with optical service protocols was very important. A lot of people thought that peer level optically (OO, not OEO) switched 10GbE would be the solution to that problem. With the growth of the Internet taking over the existing transmission facilities, 40Gb would be needed for the continued growth of the existing services. This is where the market for terabit routers originated.

You can start to see the domino effect that the political power that such inflated predictions can have on, not just Internet technology providers, but the whole industry. As time allowed the reality to of the market to set in, and more realistic economic modeling was done, UUNet engineering lost a lot of their power to be an independent group.

When Fred Brigs (an old MCI executive, and one of the few executives that I trust) took over UUNet engineering, most of the old UUNet people left and we got a more accurate picture of what UUNet was doing. By this time, I was deep into the architecture and deployment of Ethernet services within WorldCom, so I only got a limited picture of what was going on. I do understand that it was found that UUNet engineering was measured on how many ports and how much bandwidth they could deploy, regardless of the customersG«÷ usage. When I visited the cache server co-location facilities, they were mostly empty.

UUNet evidently bought into their own hype. In buying into their own hype, the political power that they had caused a lot of other organizations and companies to answer to that hype. This was not a good time to be politically incorrect. If the WorldCom management had been more in contact with the reality of the customers and the business markets, and wanting to I do not believe that UUNet would have gotten to be the political power that they became. If the WorldCom management had not seen the stock market value of UUNet growth, I do not believe that UUNet would have become the market influence that they became. (Remember, WorldCom management evidenced more concern for the value of their stock than the needs of their customers.) Maybe this can be a lesson to the management of other companies and to the industry as a whole.
skeptic
50%
50%
skeptic,
User Rank: Light Beer
12/4/2012 | 10:03:26 PM
re: Did WorldCom Puff Up the Internet Too?
which do you think was more correct? Odlyzko's 1x (plus or minus X%), Robert's 3x calculation, or UUnet's 10x per year?
---------------------
Again, it depends on where in the network you
are measuring and what sort of equipment you
want to measure the growth in. I think there
was tremendous growth in the 1998-99 timeframe,
but its difficult to seperate out growth that
was due to new customers (connectivity) vs.
bandwidth growth.

It also depends on where you make the measurement.
There are cities/locations which grow far faster
than other locations. Despite what some people
think, there just isn't an easy number for
"internet growth".

In terms of bandwidth growth, anything over 100%
(assuming that it did happen) was growth from a
small base and anyone with any sense would have
seen that it wasn't sustainable at those
extraordinary rates.

The other thing is that despite what lots of
people think, in many cases there isn't a linear
relationship between "bandwidth" and purchasing.

- Nobody (sane) runs their links at full capacity
- Connectivity requirements can grow networks in
unexpected ways.


-----------------------
What do you think the growth rate for Internet traffic is today? If it is even 2-3x, why aren't the ISPs buying more C&J routers? You would think they would be running out of capacity by now.
-----------------------

Believe it or not, C&J are still selling routers.
And if you went around to some of their customers
you would see:

- Multiple core routers meshed together to get
extra ports toward the edge.
- GSRs and M160's that are running out of slots in
high-growth locations.
- Routers that were purchased long ago that are
still adding cards.

And you have to understand that the economy and
debt situation at the carriers is so terrible that
even if certain investments make sense from a
network point of view, they will not be made if
there is a way to work around the problem with
less cost.

To summerize:

- I believe that long-haul is overbuilt and
unless someone can show HUGE cost savings with
new equipment, there is going to be very little
new stuff deployed. Nobody in rural idaho is
demanding "wavelength services".

- The only way to really understand whats going
on in service provider networks is through
direct contact with them. I've never seen any
public information on traffic, growth or even
network design that was all that trustworthy.


- I believe that internet traffic in the core
of the network is still growing at a healthy
rate.

- I think that there is far less "router" capacity
than "transmission" capacity. The fiber (and
the DWDM multiplier effect) have made more fiber
capacity available than be used with the existing
number of core routers. (i.e. growth for routers,
no growth for fiber/DWDM).















In my opinion, transmission (fiber) was overbuilt
because the implications of DWDM didn't sink in
to some people's heads fast enough.
Roy_Bynum
50%
50%
Roy_Bynum,
User Rank: Light Beer
12/4/2012 | 10:03:27 PM
re: Did WorldCom Puff Up the Internet Too?
I was a MCI employee when the MCI executives were looking for someone to buy the company and make them rich. When WorldCom took over, I was working on IP over DWDM and Ethernet optical networking and the architecture of optical switched networks. At first, a lot of this work was based on knowing what our customers were asking for and were willing to pay for and how services and applications evolve over a long period of time. This type of project is considered a three to five year project before any new services would deployed and start being sold to customers. As time progressed, more and more resources were siphoned off to respond to G«£internal and industry political pressuresG«•.

(I am going to use a term: G«£Internet cadreG«• to refer to a whole group of people that took on an almost religious dedication to the supremacy of the Internet and Internet Protocol. While I am the original host master for MCI, I am somewhat pragmatic about where, when, and how to use protocols. You will see in this post that I am somewhat disdainful of those that show an emotional and political attachment to the G«£InternetG«• and IP.)

The G«£Internet cadreG«• had gained a lot of political power because it had been predicted that they would become the dominant service, so a lot of resources were being fed into developing what was called G«£multi-serviceG«• access technology. Before a lot of real cost/revenue modeling had time to be done, the G«£industryG«• considered a cost effective way to provide voice, private line, and packet based Internet services over the same common channel facility. It was only after there had been time to get into the reality of the economic modeling, that it was found out that it would not work. (G«£That is a whole G«ˇnother story.G«•) By then, much of the damage had been done to create the G«£bubbleG«•.

The political power of the G«£Internet cadreG«• within the industry was based on the G«£growthG«• of bandwidth and connectivity that was supposed to be occurring within the Internet. If the growth had really been there, there would have been major markets for all sorts of hardware and software products. It would not have mattered how many vendors with similar technology there were, all of them would have a market and customers clamoring for their products.

In reality, the G«£growth of the Internet was limited more to connectivity than bandwidth. This can be seen when you recognize that most dial up ISPs have a statistical gain of 100:1 between their customers and the backbone, and the NSPs like Internet MCI, UUNet and others have a 4:1 gain between the ISP and backbone. That means that it takes a 800% growth in connectivity to show a 100% growth in overall bandwidth. This is easy to see in retrospect, but hard to see unless you are one of those insiders that are developing and implementing the connectivity/backbone technologies.

The problem with getting this type of information out to the public is the politics of the G«£Internet cadreG«• and their marketing needed to keep it sequestered. This would be particularly true of companies whose revenue projections were based on the growth of Internet service bandwidth. While it could not be called a G«£conspiracyG«•, none of those companies, vendors or service providers, wanted anyone to G«£rain on their paradeG«•. It became a G«£feeding frenzyG«• of bandwidth and service predictions and the requirements for routers, switches, fiber, access, etc, to support them.

There is enough G«£blameG«• to go around to almost everyone in the industry between the years of 1995 and 2001.
mma
50%
50%
mma,
User Rank: Light Beer
12/4/2012 | 10:03:29 PM
re: Did WorldCom Puff Up the Internet Too?
DCME is Digital Circuit Multiplication Equipment. It is typically used on undersea cable and satellite digital carriers to obtain from 100 to 240 trunks per 1.544 Mb/s DS1. This is done by using a combination of fast switching, where the long haul channel is held only during speech bursts, and speech encoding at about 32 Kb/s or 16 Kb/s. DCME cause delay, clipping, and low bit rate encoding impairments of the speech signal, so that using multiple DCME in a switched connection would be poor practice. This is particularly true if wireless links are used at the ends, since they also add low bit rate encoding impairments.

Modem and fax signals diminish the circuit multiplication factor, since they do not have silent intervals that can be detected and removed. Higher rate modem signals are not transmissible through the lower bit rate speech codecs. Either 40 or 64 Kb/s channels must be allocated, or the modem signals must be demodulated at one end and remodulated at the other. Training and fax handshake signals must also be cared for.

See ITU Recommendations G.763 and G.767.
broadbandboy
50%
50%
broadbandboy,
User Rank: Light Beer
12/4/2012 | 10:03:32 PM
re: Did WorldCom Puff Up the Internet Too?
"Odlyzko also talks about ATT dismissing his claims that the internet was growing slower than worldcom said."
---------------------------------

Andrew worked for AT&T (Labs) at that time, and had access to their internal growth statistics. That knowledge must have been incorporated into his estimates.

Still, some people within the company disagreed with his thesis that growth would average only about 100% per year over the long run. AT&T was officially claiming >200% annual growth at the time, and their real growth rate may have actually been higher.

Skeptic, since you seem to know something about routers and the Internet, which do you think was more correct? Odlyzko's 1x (plus or minus X%), Robert's 3x calculation, or UUnet's 10x per year?

To be fair, we have to talk about the 1998-99 time frame.

What do you think the growth rate for Internet traffic is today? If it is even 2-3x, why aren't the ISPs buying more C&J routers? You would think they would be running out of capacity by now.


BBboy


broadbandboy
50%
50%
broadbandboy,
User Rank: Light Beer
12/4/2012 | 10:03:33 PM
re: Did WorldCom Puff Up the Internet Too?
"Our story was published on July 18 which would lead us to believe it came before the Economist. But I don't know when that issue of the Economist hit the newstands, nor have I seen that issue of the Economist."
-----------------------------------------------

Who cares who was first?

The BroadbandWorld article cited earlier was published in June 2001.

It was a response to an earlier McKinsey/JP Morgan research report alleging Internet growth had slowed. The McKinsey report was heavily influenced by Odlyzko's earlier research.

BroadbandWorld did not allege Sidgemore et. al. were lying, rather, that too many people assumed the claims of network capacity (router ports) doubling every 3-4 months equated to traffic growth, or even overall Internet growth, of 1000% per year.

The UUnet/Worldcom growth rate was frequently cited in presentations at conferences like NGN and elsewhere. Eventually the "myth" of Internet traffic growth became an urban legend.

The rest, as they say, is history.

Reference:

www.broadbandpub.com/broadband...


LightSeeking
50%
50%
LightSeeking,
User Rank: Light Beer
12/4/2012 | 10:03:33 PM
re: Did WorldCom Puff Up the Internet Too?
Skeptic,

Your knowledge and assessment about this industry are impressive and painstakingly correct. I am wondering what you do for living!

LightSeeking
Phonon-Ex
50%
50%
Phonon-Ex,
User Rank: Light Beer
12/4/2012 | 10:03:36 PM
re: Did WorldCom Puff Up the Internet Too?
Flanker

Please - what is a DCME? A"digital compresser????"

PEX
skeptic
50%
50%
skeptic,
User Rank: Light Beer
12/4/2012 | 10:03:36 PM
re: Did WorldCom Puff Up the Internet Too?
Our story was published on July 18 which would lead us to believe it came before the Economist. But I don't know when that issue of the Economist hit the newstands, nor have I seen that issue of the Economist.
================

Odlyzko is the main source. But the quotes from
him are far more agressive in the economist
than in your story. Their story was also much
more specific in nailing down what claims worldcom
made about traffic and when.

Odlyzko also talks about ATT dismissing his
claims that the internet was growing slower
than worldcom said.

He also claims (or the article claims for him)
that companies were doing "hollow swaps" to
boost sales and traffic figures. Though, they
make the claim quoting him in a sparse way.

As far as when they did the story, its hard to
tell. The dates are so close (20th vs. 18th),
that its difficult to tell anything from that.
And I dont know what the lead-times are on
a weekly print magazine. The magazine might
have been available mid/late-last-week.



cfaller
50%
50%
cfaller,
User Rank: Light Beer
12/4/2012 | 10:03:36 PM
re: Did WorldCom Puff Up the Internet Too?
with the stylings of willywilson:

skeptic wrote:

1. "If the numbers were silly and inflated, why
has it taken several years for anyone to
figure that out? Wouldn't, for example, worldcom's financials for 1998 have shown in
1999 that their actual build-out didn't reflect
1000% growth?"
2. "That would seem to indicate that the analysts who used the numbers without validating
them are guilty of fraud and negligance on
a rather extraordinary scale."
------------------------------
1. Not if WorldCom took operating expenses and assigned it as capex. That would boost the capex number. Oh, wait, they DID do that!
2. Yup.
skeptic
50%
50%
skeptic,
User Rank: Light Beer
12/4/2012 | 10:03:37 PM
re: Did WorldCom Puff Up the Internet Too?
Sure they were after M&A. But any knucklehead
VC would have asked the question: how long
is this going to continue?
======================

But thats the whole thing about a bubble.
VC's don't ask that question. Don't you
remember the "new economy"? Or those long
pieces about how "bricks and mortar" companies
were being made obsolete by dot-coms.

And if you did bring up the "hard questions",
you were going to be labeled "out of touch" or
"out of date". Nobody was listening.



skeptic
50%
50%
skeptic,
User Rank: Light Beer
12/4/2012 | 10:03:37 PM
re: Did WorldCom Puff Up the Internet Too?
So if John was the busy boy planting a bunch of silly, inflated Internet numbers at analyst conferences, I am pleased to know that he was "there" for at least ONE WorldCom fraud...
---------------

If the numbers were silly and inflated, why
has it taken several years for anyone to
figure that out? Wouldn't, for example,
worldcom's financials for 1998 have shown in
1999 that their actual build-out didn't reflect
1000% growth?

That would seem to indicate that the analysts
who used the numbers without validating
them are guilty of fraud and negligance on
a rather extraordinary scale.

But all "good" people know that its worldcom's
fault now. How likely was it, after all,
that the financial community would ever be held
responsible for its actions.

skeptic
50%
50%
skeptic,
User Rank: Light Beer
12/4/2012 | 10:03:37 PM
re: Did WorldCom Puff Up the Internet Too?

Scott,

What I'm telling you is that a bunch of ignorant
fools can convince themselves of anything. (See
the dot-com's)

Most of the people who were making "money"
decisions were doing so based on relative
valuations and stock prices. Most VC's cared
far more about Juniper's market cap than traffic
growth in the internet.

As far as those "research reports",
they were often put together by people with
a vested interest in hyping the market (brokerage
firms) or were being paid by startups to do
the research.

I suppose its easy for someone like you to excuse
all of those people for their ignorance, but
letting them off the hook and blaming worldcom
is absurd.

If you want to know what hyped the market, start
with:

- Vendor-based financing
When the large companies started inflating
their sales by essentially selling to
themselves, the economics of the business
started to go crazy.

- IPO's and M&A
Believe me when I say that most VC's were
more motivated by the market cap of Juniper
and the fact that billions was being paid
out for startups without finished or
even working products
.....than by any "projections" of internet
growth.

- Kickbacks between startups and vendors
When startups started giving pre-IPO stock
out to their customers (and management at
their customers), a feedback loop was created
that led to utterly absurd situations.


Look, there were VCs pumping money into startups
and companies buying startups....whose business
plans and products were so non-sensical or non
functional that even if the internet had been
growing at 1000%, they still would have failed
for lots more basic reasons.

There was a time in 1999 where anyone could get
funding. All they had to do was put the word
"internet" on it and the money would pour in.

Scott Raynovich
50%
50%
Scott Raynovich,
User Rank: Light Sabre
12/4/2012 | 10:03:37 PM
re: Did WorldCom Puff Up the Internet Too?
Our story was published on July 18 which would lead us to believe it came before the Economist. But I don't know when that issue of the Economist hit the newstands, nor have I seen that issue of the Economist.

I am not familar with the Economist story so I can't comment on how they are related, if at all. Possibly both reporters used the same source? Was a source named in the Economist?

skeptic
50%
50%
skeptic,
User Rank: Light Beer
12/4/2012 | 10:03:38 PM
re: Did WorldCom Puff Up the Internet Too?

As you are being given attribution in a variety
of places for this story, can you shed any
light on your story's relationship to a
similar story that appears on p. 60 of the
July 20th Economist? (re: the power of worldcom's
puff: Exaggerated figures for internet traffic
inflated the telcoms bubble).

(i.e. which came first and were the authors
& editors aware of the other story before the
one on Light reading appeared).
erbiumfiber
50%
50%
erbiumfiber,
User Rank: Light Beer
12/4/2012 | 10:03:38 PM
re: Did WorldCom Puff Up the Internet Too?
In his on-line interview at the Washington Post, Sidgmore explained that as Vice-Chairman of the Board it was bascially his job to just go around and speak/show slides at tech conferences (see, THAT'S why he never had time to look at all those financial reports). He said before Congress, in the interview, and in today's e-mail to employees that he "wasn't there" for the accounting fiasco (although I feel if you are a BOARD MEMBER that you ARE "there.")

So if John was the busy boy planting a bunch of silly, inflated Internet numbers at analyst conferences, I am pleased to know that he was "there" for at least ONE WorldCom fraud...
jamesbond
50%
50%
jamesbond,
User Rank: Light Beer
12/4/2012 | 10:03:38 PM
re: Did WorldCom Puff Up the Internet Too?
skeptic wrote:

The "crash" of the stock market wasn't caused
by "traffic predictions". If anything, it was
caused by several companies making false
statements about earnings

The VC's didn't care about traffic predictions.
It just never came up. What they cared about
were stock valuations and M&A valuations for
companies in the space. The IPOs in 1999-2000
and the M&A's meant a whole lot more than
internet traffic growth ever did.

------------------

Sure they were after M&A. But any knucklehead
VC would have asked the question: how long
is this going to continue? Answer would have
been as long as the Internet traffic is
exploding. And hence the WCOM reports definitely
contributed in some way to the crash.

Call it sour grapes but I am just glad that the Stanford's old boy network isn't making money
anymore.
skeptic
50%
50%
skeptic,
User Rank: Light Beer
12/4/2012 | 10:03:39 PM
re: Did WorldCom Puff Up the Internet Too?
You can adjust my number around but the bottom line is if the revenue do not grow as faster than traffic, revenue per bit went down and you get chapter 11..
--------------

The bottom line you should consider is that
much of worldcom's debt was related to M&A,
not necessarly capex. This was a company
(particularly after the MCI acquisition) that
was built on leverage and debt.

Even if revenue per bit had not changed, it
would not have materially altered what happened
at worldcom.

Scott Raynovich
50%
50%
Scott Raynovich,
User Rank: Light Sabre
12/4/2012 | 10:03:39 PM
re: Did WorldCom Puff Up the Internet Too?
Wow, Skeptic, I've never been more skeptical of your knowledge than now.

You're basically saying that Internet traffic figures were never used to puff up these companies. Hmm, you mean all those analysts and industry experts showing power-point slides of massive industry growth at investment and technology conferences weren't hyping the market? You mean that all of the massive Internet growth figures alluded to in Wall Street research reports through the years had no contribution to the puffing of the bubble? You mean that none of those glowing press reports, based on misleading information about Internet traffic, didn't contribute to people getting interested in the market? You mean that all of those times that equipment vendors cited the pace of Internet expansion and equipment demand on investor conference calls, circa 1999, had nothing to do with stock prices going up? And are you telling me that VCs had absolutely nothing to do with this -- they stood objectively aside and said, "there is no massive Internet growth fueling a bubble environment, we are value investors and believe companies should be valued at 10X projected revenues"?

I think that is the most mixed-up post-bubble analysis I have ever heard.

fgoldstein
50%
50%
fgoldstein,
User Rank: Light Sabre
12/4/2012 | 10:03:39 PM
re: Did WorldCom Puff Up the Internet Too?
AOL gets its modems from several sources. They used to own ANS, which was at the time their largest provider, but sold ANS to Worldcom. Plus there are some other UUNET numbers. Then there are the Genuity and Sprint numbers, and possibly others.

Worldcom is probably still their biggest supplier. But it's a going business, so even if WCOM were liquidated, somebody would probably take over that operation. I'm sure Howard Jonas would offer a few cents on the dollar for it. ;-)

mma> I used CO in the sense of switching office or wire center. For NJ, whenever a town has a UUNET dial access number, there is almost always an AOL dial access number. Usually the AOL dial access number is either one greater or less than the UUNET dial access number. This may indicate that UUNET and AOL numbers are in a hunt group that directs traffic to the same access equipment.

There are, however 151 AOL dial access numbers and only 60 UUNET access numbers in the state.
lowbandwit
50%
50%
lowbandwit,
User Rank: Light Beer
12/4/2012 | 10:03:40 PM
re: Did WorldCom Puff Up the Internet Too?
First, my 1000 year remark was a joke- don't you people have a sense of humor? :-) I don't know about many of the estimates in this thread but, VERY conservatively, we should be looking at something less then 100 years before we need more capacity then we have buried right now.

beowulf:

I know about incremental cost to add additional fiber. I don't have a problem with that at all. What I do have a problem with is the number of players who went for the brass ring by trying to outdo the others. L3's business model was to design their network so that they could expand and upgrade it quickly and efficiently- smoking the incumbents. The problem there is that, published or not, there were other carriers doing the exact same thing. It's like everybody forgot that the competition was more then just AT&T. They were all so busy building their networks to cream traditional networks that they ran their debt into the sky. (It would be ironic if the only one left is AT&T, don't you think?)

You also left out the fact that many of the carriers buried empty conduit in addition to dark fiber. That was to allow cheap and easy pulls of 'next gen' fiber when such a thing came along. Again, they would just be paying for the fiber and labor to pull it while others were trying to re-trench.

Last thing: on the computing side... I have a AT&T digital cable box (new, with optical digital audio and S-video). If that thing is performing a gigaflop, I'm a monkey's uncle. I'd be very surprised if it was even in Pentium class performance-wise. I'm basing this on menu refresh rate and channel switch times on the digital portion.

Apple's G4 has been capable of gigaflop for a couple of years.
boozoo
50%
50%
boozoo,
User Rank: Light Beer
12/4/2012 | 10:03:40 PM
re: Did WorldCom Puff Up the Internet Too?
"revenue per bit went down and you get chapter 11.."

No true, because you are missing an important variable: "cost per bit to support that traffic"
If the "cost per bit" is also negative, than you can arguably sustain negative revenue per bit and be profitable.


Boozoo
sgan201
50%
50%
sgan201,
User Rank: Light Beer
12/4/2012 | 10:03:41 PM
re: Did WorldCom Puff Up the Internet Too?
Hi All,
You are asking the wrong questions.
And, you are looking at wrong thing..
Let assume the Internet traffic at the beginning of the year as T.
The revenue at the begining of the year as R..
If we say the Internet is growing at 10 times, then, the traffic at the end of the year is 10X.
Since we know the revenue earn by transporting bits did not grow at 10 times, let assume it grows at 2 times = 2T..
So what we get is,
at the begining of year,
Revenue per bit = R/T
At the end of the year,
Revenue per bit = 2R/10T = R/5T..
To keep up with the growth, CAPEX sky rocket..
But revenue per bit is dropping 5 times per year..
What you get is a recipe for chapter 11..
If the internet growth is actually 10 times, most ISP will be bankrupted a long time ago as opposed to now..
You can adjust my number around but the bottom line is if the revenue do not grow as faster than traffic, revenue per bit went down and you get chapter 11..

skeptic
50%
50%
skeptic,
User Rank: Light Beer
12/4/2012 | 10:03:42 PM
re: Did WorldCom Puff Up the Internet Too?
Most of the companies that were in a position to benefit from the increased internet traffic did lie. Start-ups also lied to VCs as they wanted to get funding. All these traffic predictions lead to excess inventories and brought a false sense of prosperity that eventually led to the crash of the stock market.
-------------------

I wish people would take a step back.

- The "crash" of the stock market wasn't caused
by "traffic predictions". If anything, it was
caused by several companies making false
statements about earnings.


- The VC's didn't care about traffic predictions.
It just never came up. What they cared about
were stock valuations and M&A valuations for
companies in the space. The IPOs in 1999-2000
and the M&A's meant a whole lot more than
internet traffic growth ever did.




victorblake
50%
50%
victorblake,
User Rank: Lightning
12/4/2012 | 10:03:44 PM
re: Did WorldCom Puff Up the Internet Too?
As a point of additional information. Subscriber lines are usually pulled out of a CO or LSO and backhauled to a POP where the dial equipment (Ascend/Lucent, Cisco, 3Com/USR, etc.), switches, and routing reside. Lines are typically backhauled on PRI's. So the CO or LSO location are often independent of the dial-ISP POP location. However, there are sometime in the same building where the scale is either very large or very small.
billyjoebob
50%
50%
billyjoebob,
User Rank: Light Beer
12/4/2012 | 10:03:45 PM
re: Did WorldCom Puff Up the Internet Too?
My point (which I never made) was that as an entertainment media it will have to compete with other entertainments for the residential entertainment budget. Cable TV, Satelite TV, DVDs/VHS rentals and Video Games all have a share of that market. For the Internet to compete it must maintain a similar pricing structure to the incumbents. Hence the prices that broadband access can charge are bound by what cable and satelite can charge. I don't know if there can be enough revenue from residential pricing to pay down the debt loads of the QWests, UUNets and PSInets to make the last three years of hyper investment profitable.

My apologies for not making myself clear (if indeed I have done that yet.)
BobbyMax
50%
50%
BobbyMax,
User Rank: Light Beer
12/4/2012 | 10:03:47 PM
re: Did WorldCom Puff Up the Internet Too?
Most of the companies that were in a position to benefit from the increased internet traffic did lie. Start-ups also lied to VCs as they wanted to get funding. All these traffic predictions lead to excess inventories and brought a false sense of prosperity that eventually led to the crash of the stock market.

What happened to Onternet, the same thing is happening to Storage where over 80 companies have sprung up in the US. Most of them are exactly identical.
papabear
50%
50%
papabear,
User Rank: Light Beer
12/4/2012 | 10:03:47 PM
re: Did WorldCom Puff Up the Internet Too?
Same report on CNN Money

http://money.cnn.com/2002/07/2...

papabear
50%
50%
papabear,
User Rank: Light Beer
12/4/2012 | 10:03:47 PM
re: Did WorldCom Puff Up the Internet Too?
Breaking news on www.foxnews.com "Worldcom Inc. says it will file for bankruptcy"

billy_fold
50%
50%
billy_fold,
User Rank: Light Beer
12/4/2012 | 10:03:48 PM
re: Did WorldCom Puff Up the Internet Too?
UUNet (including Sige) always lied about Internet growth. They always thought that they had to lie to get Worldcom to give them more bandwidth. I don't think that they ever stopped to consider that the more bandwidth you throw at TCP/IP, the more bandwidth it will use.


billy
who is john galt?
50%
50%
who is john galt?,
User Rank: Light Beer
12/4/2012 | 10:03:48 PM
re: Did WorldCom Puff Up the Internet Too?
Billy,

Perhaps UUnet was using Ms. Clio to predict bandwidth growth.
flanker
50%
50%
flanker,
User Rank: Light Beer
12/4/2012 | 10:03:48 PM
re: Did WorldCom Puff Up the Internet Too?
The bottom line is that if the 100% growth rate for the Internet is true, it may translate into 100% capacity demands for things like routers. However, since the Internet is a small slice of the telecom transmission market, it doesn't translate into a 100% growth rate for fiber, fiber terminals, DWDM wavelengths, SONET equipment, and other types of transmission equipment.

I think if you are an ILEC, RBOC (or to a lesser degree) a CLEC, voice is still a big part of the picture. But you have a lot of data carriers that are just moving around layer 2 traffic (Global Crossing, Level3, Williams, Broadwing) or IP traffic over some layer 1 or 2 protocol/physical medium.

The network segmentation between profitable RBOCs and money-losing data carriers is confusing the issue about traffic demand. Not to mention the fact that even if an RBOC hands off LD traffic to a third party, the third party may compress it over the backbone.

Jamesbond asks about private peering and this gets back to my point that some backbone traffic (or a lot, depends on who you talk to) doesn't cross the major NAPs like MAE East.

You can set up private peering where two carriers like LVLT and Williams swap data traffic and it may never cross a public NAP (I could be wrong, but I get the impression private peering is more common that people admit).

These agreements are to some degree confidential and proprietary, so I can see why it isn't talked about much in the press.




mma
50%
50%
mma,
User Rank: Light Beer
12/4/2012 | 10:03:49 PM
re: Did WorldCom Puff Up the Internet Too?
--------------
"You are mad! Data carrier traffic is not 68% voice! Maybe for Verizon, but... You can compress LD voice using a DCME and the bandwidth usage drops off a cliff."

--------------

Quite possibly so. But I was estimating the demand for inter-office transmission, metro and long haul. This would include the demands due to growth in voice, leased line, data networks like frame and ATM, and Internet.

Voice, while a slow grower, is a massive business, and it is almost always carried interoffice as 64 Kb/s PCM signals. DCME, PCME, voice over frame relay, VoIP, etc have not been applied in a significant way to the domestic US network. Compression is used in the wireless link.

The percentage leased line seemed low to me, but point-to-point special services were a small fraction of the total specials, and many of them were voice services that may have migrated to the switched network due to innovations in call processing and especially cheaper billing rates.
Speech compression may also be having some impact, and there has also been a tendency to rip out lightly used point-to-point data circuits and roll the traffic onto the corporate intranet to take advantage of the statistical multiplexing gain and the cheaper rates for widerband circuits.

As for the Internet, the routemaps that Boardwatch showed for the major backbone providers always struck me as having mostly pretty thin routes. Even at the 100% growth rate, the Internet was small compared to the rest of the telecom network by the end of '99, and it would not equal it until about 2003.

The bottom line is that if the 100% growth rate for the Internet is true, it may translate into 100% capacity demands for things like routers. However, since the Internet is a small slice of the telecom transmission market, it doesn't translate into a 100% growth rate for fiber, fiber terminals, DWDM wavelengths, SONET equipment, and other types of transmission equipment.
mma
50%
50%
mma,
User Rank: Light Beer
12/4/2012 | 10:03:49 PM
re: Did WorldCom Puff Up the Internet Too?
---------------------------------

"so I'm paying about 30c per (potential) Gig :)

with IP transit at, say, $100/meg these days you come to a similar number for global delivery of IP packets.

The issue, as I see it, is more about having a sufficiently fat pipe to the home to stream movies in real time."

---------------------------

You are correct. I did intend $1 per GigaByte. That was the most that we could envision flowing to the carrier for sending a pay per view movie to the home using MPEG 1 coding, which would take about 1 GB and could be done over the ADSL contemplated at the time.

Today, however, things have changed. Hard disk drives are about $1/GB, DVDs can be produced for about $2 per 4.6GB disk (10,000s), and the processors in DVD drives, digital satellite and cable modem boxes easily beat the $100 per 1 GOp/s. In this competitive landscape there is little chance for entertainment video on the telecom network, and cable, satellite and video stores will continue to be the primary distribution.

That said, the Internet does provide a very nice "magazine with audio" service.
jamesbond
50%
50%
jamesbond,
User Rank: Light Beer
12/4/2012 | 10:03:50 PM
re: Did WorldCom Puff Up the Internet Too?
I think the numbers for IP growth are somewhere between 100% per year and 100% per quarter. As I said, I think Andrew O's numbers are VERY conservative. I think private line and privately peered IP traffic has to be factored in.

------------------------------

flanker,

can you give an example of a private peering
point? are these non-service provider companies?

thanks
flanker
50%
50%
flanker,
User Rank: Light Beer
12/4/2012 | 10:03:51 PM
re: Did WorldCom Puff Up the Internet Too?
"In early 2000 I did some estimates of transmission demand growth rates. At the time (end of '99 data) it appeared that the traffic load was 68% voice"

You are mad! Data carrier traffic is not 68% voice! Maybe for Verizon, but... You can compress LD voice using a DCME and the bandwidth usage drops off a cliff.

If your line traffic is 68% voice, your economics are all screwed up and your network voice costs are way to high.

flanker
50%
50%
flanker,
User Rank: Light Beer
12/4/2012 | 10:03:51 PM
re: Did WorldCom Puff Up the Internet Too?
I am not trying to criticize but if I take
your argument then we should start counting
LAN traffic as "Internet" traffic.


I see your point. But there is a lot of data traffic out there on backbone networks, IP or not, that does not transit a public peering point.

Of course I would not include LAN traffic because it never transits a backbone, but why exclude private line traffic and traffic engineered to avoid congested peering points as not being part of "IP traffic" growth, broadly speaking?

I think the numbers for IP growth are somewhere between 100% per year and 100% per quarter. As I said, I think Andrew O's numbers are VERY conservative. I think private line and privately peered IP traffic has to be factored in.

That said, I think many backbone networks are upgradeable to 10x or 100x there current capacities using wavelength multiplexing. It's not very sexy to talk about installing wavelength cards to upgrade capacity, as opposed to wasting $200mln on virgin fiber and ADM nodes.


jamesbond
50%
50%
jamesbond,
User Rank: Light Beer
12/4/2012 | 10:03:52 PM
re: Did WorldCom Puff Up the Internet Too?
There is a lot of data traffic out there that never transits an IP peering point. There is private peering, and private line IP traffic, not to mention data traffic not transported over IP.

-----------------------------

I am not trying to criticize but if I take
your argument then we should start counting
LAN traffic as "Internet" traffic.

Data traffic not transported over IP cannot
be "Internet" related. yes?




mma
50%
50%
mma,
User Rank: Light Beer
12/4/2012 | 10:03:52 PM
re: Did WorldCom Puff Up the Internet Too?
"I also don't get the reference to COs. The term CO is usually used to refer to a central office of an ILEC (such as an RBOC, etc.)."

I used CO in the sense of switching office or wire center. For NJ, whenever a town has a UUNET dial access number, there is almost always an AOL dial access number. Usually the AOL dial access number is either one greater or less than the UUNET dial access number. This may indicate that UUNET and AOL numbers are in a hunt group that directs traffic to the same access equipment.

There are, however 151 AOL dial access numbers and only 60 UUNET access numbers in the state.
flanker
50%
50%
flanker,
User Rank: Light Beer
12/4/2012 | 10:03:52 PM
re: Did WorldCom Puff Up the Internet Too?
The issue of Internet growth was argued on this webwsite last year. It isnt old news because there is no consensus as to how to measure it.

Andrew O. uses some fairly conservative metrics, and only considers traffic that transits a peering point between tier one Internet backbone.

There is a lot of data traffic out there that never transits an IP peering point. There is private peering, and private line IP traffic, not to mention data traffic not transported over IP.

If the connections are revenue generating, you can't fault Worldcom for saying port growth is 100% per three months.

Get real. It's not a conspiracy, and WCOM is a "real" company with real traffic, real hardware and real revenues.





PresterJohn
50%
50%
PresterJohn,
User Rank: Light Beer
12/4/2012 | 10:03:53 PM
re: Did WorldCom Puff Up the Internet Too?
-------
-------
By the end of the MPEG 1 work ('91), we knew that entertainment required 1 Giga Op/s processors for <$100, 1 GB disks for <$100, and 1 GB to the home for <$1. Alas, the telecom industry didn't make the last number, while the semiconductor and storage industries have made theirs. So CD-RW, DVD and TIVO rule, and video over the telecom network is going nowhere (unless you consider satellite and cable to be telecom networks).
------
------

------
are you sure about that? (the telecom industry failing to make <$1 per GB)
------

I'm sure he meant 1 GB/s, since a 1 GB/month data rate doesn't deliver all that many movies.

Not sure how he's letting the semiconductor industry off the hook by declaring >1 giga-op processors for < $100 though. A 2 Ghz Pentium does not run in the giga op range.
victorblake
50%
50%
victorblake,
User Rank: Lightning
12/4/2012 | 10:03:53 PM
re: Did WorldCom Puff Up the Internet Too?
AOL, like many other companies, does buy services from various Worldcom companies dating as far back as Wiltel, and including UUnet, MFS, and MCI Metro services. These include local SONET transport, Internet access, VPN, and of course dial-up services. Microsoft and AOL in fact both purchase dial-up services from UUnet.

The reference to UUnet and AOL POPs doesn't make sense. AOL's POPs in its WAN may in some cases be in the same buildings (Carrier hotels) as UUnet facilities, but they do not share space. I also don't get the reference to COs. The term CO is usually used to refer to a central office of an ILEC (such as an RBOC, etc.). Since neither AOL nor UUnet operate such business, they don't use those terms. I suppose some folks (but even then not all) within MCI and MFS companies (part of Wcom) would use those terms for some facilities, but again -- not applicable here.

UUnet, along with Sprint, Genuity, and others (now likely Level 3 and Qwest as well) provide outsourced dial-up and other access (xDSL) access and aggregation.

Would UUnet shutting down its network affect AOL. Yes. But that wouldn't be the real news. For AOL, UUnet is one of many service providers. For most UUnet customers, they are the only form of internet connectivity they have. The real news would be the disappearance off of the Internet of all of those Fortune 1000 companies and numerous small ISPs, .coms, etc.

In my well informed and not so humble opinion, the growth numbers have been way overinflated for years. But the real reason is explained by a kind of double counting of numbers. One simple example is in an outsourcing deal where AOL buys dial capacity from UUnet, they count a number, then AOL counts the same number. In fact both can count the number twice (ingress and eggress from their respective edges). Sidgmore's UUnet and MFS groups were and are the shining star of Worldcom. I hope those groups do survive beceause they have built something useful and well.

Victor Parente-Blake
Titanic Optics
50%
50%
Titanic Optics,
User Rank: Light Beer
12/4/2012 | 10:03:53 PM
re: Did WorldCom Puff Up the Internet Too?
This story was talked about in Barron's Technology Week column. Congrats to LR--nice mention as the authority on all things Internet backbone.

If anybody is a subscriber to Barrons, you can go to www.barrons.com and post the text; here's the last paragraph:

"WorldCom's phantom toll booths, in turn, helped fuel the myth that Internet traffic was doubling every 3.5 months, or 1000% a year. In fact, say researchers cited by Light Reading, Internet traffic appears to be doubling every year--a nice growth rate, but less than advertised. Sidgmore didn't return calls for comment from Light Reading (www.lightreading.com) or from Barrons."
mma
50%
50%
mma,
User Rank: Light Beer
12/4/2012 | 10:03:54 PM
re: Did WorldCom Puff Up the Internet Too?
"The entertainment industry (radio, TV, movies, porn, etc.) probably generates more 100 gigabucks per year. "

-------------------------------

I think you're off by a factor of at least 5 for paid media (movie box office, video rental, in-flight, pay per view, premium channels). $100G is order of magnitude right for media advertizing revenue from all sources (TV, radio, newspapers, magazines, direct mail, etc).
mma
50%
50%
mma,
User Rank: Light Beer
12/4/2012 | 10:03:54 PM
re: Did WorldCom Puff Up the Internet Too?
Once upon a time (Sept 8, '97) AOL and Worldcom did business together.

--------------------

http://news.com.com/2100-1001-...

<quote> WorldCom (WCOM) said today it would buy CompuServe (CSRV) for $1.2 billion in stock from H&R Block (HRB) and then shed CompuServe's interactive services division to America Online (AOL).
As part of the complex three-company deal, WorldCom also will pay AOL $175 million and receive AOL's ANS Commmunications division. Under a five-year contract, WorldCom will become AOL's largest network service </quote>

--------------------------

Lots of AOL POPs still have numbers similar to UUNET POPs and are in the same COs.

Does anyone know how dependent AOL is on UUNET's network?
giles0
50%
50%
giles0,
User Rank: Light Beer
12/4/2012 | 10:03:54 PM
re: Did WorldCom Puff Up the Internet Too?
-------
By the end of the MPEG 1 work ('91), we knew that entertainment required 1 Giga Op/s processors for <$100, 1 GB disks for <$100, and 1 GB to the home for <$1. Alas, the telecom industry didn't make the last number, while the semiconductor and storage industries have made theirs. So CD-RW, DVD and TIVO rule, and video over the telecom network is going nowhere (unless you consider satellite and cable to be telecom networks).
------

are you sure about that? (the telecom industry failing to make <$1 per GB)

My 512kbit/s ASDL service can deliver:

(512 * 1000 * 60 * 60 * 24 * 30) /
(8 * (2 ^ 30))
= >150 GB of data per month.

It costs me about $45.

so I'm paying about 30c per (potential) Gig :)

with IP transit at, say, $100/meg these days you come to a similar number for global delivery of IP packets.

The issue, as I see it, is more about having a sufficiently fat pipe to the home to stream movies in real time.

Having said that my ADSL modem is rated at 8Mb/s, so I suspect the technology has already solved this problem also (for those of us close enough to the local exchange).

Giles
mma
50%
50%
mma,
User Rank: Light Beer
12/4/2012 | 10:03:55 PM
re: Did WorldCom Puff Up the Internet Too?
Odlyzko in Table 1.1 of http://www.research.att.com/~a... provides the relative traffic loads on voice versus Internet networks. I was surprised at the low percentage of traffic for private line, since the conventional wisdom has been that the faster growing data traffic segment will overwhelm voice any day now for the last three decades. However, the notions of "data", "digital" and "special service circuits" have been somewhat fuzzy, and I decided that Odlyzko's percentages might be right. Although "special service circuits" versus "interoffice trunks" achieved parity during the early '70s, most specials were things like foreign exchange lines, PBX tie trunks, and other short-haul voice circuits. The category of "conditioned data lines" was much smaller. Although I don't have first hand knowledge of the current statistics, it is very plausible that voice specials went INA and data specials went leased lines, with the longer haul part of the latter reflected appropriately by Odlyzko's data.

As a sanity check, I looked at the unfortunately incomplete FCC statistics for '98 and '99 and for those years, the combination of Bell South, US West, Ameritech, Pacific Bell, and SBC, and General Telephone (NYNEX and Bell Atlantic missing) reported a 35% increase in T1 equivalent transmission facilities. I thought that this was in reasonable agreement with an anticipated 30% increase in offered load for the next year givent that much of the Internet access would ride on the ILECs access facilities.

As for the holding time on Internet access, this doesn't necessarily translate into metro/metro core/long haul transmission capacity growth. Indeed, it shows the inefficiency of local access where a 64 kb/s transmission path is held continusously, while a much lower actual information rate has been transmitted beyond the first router. Indeed, I've been holding up a 64 Kb/s local access path while typing the meager number of bytes in this reply which will eventually be sent over the internet (may the gods of dial service adminstration forgive me).
brown
50%
50%
brown,
User Rank: Light Beer
12/4/2012 | 10:03:55 PM
re: Did WorldCom Puff Up the Internet Too?
broadbandboy wrote....

More likely, planners will try to project demographic growth trends out 10-15 years, and start planning the transportation system that will be needed to accommodate the expected future population.

-------------------------------------------

In this day and age, in a high tech industry like telecom, does it really make sense to install fiber for capacity that may materialize 15 years from now?

After all, who knows what kind of technology we will have 15 years from now? These fibers may become obsolete. Someone may come up with a fiber material that is infinitely better than what we have. Or maybe we'll have all have switched to satellite communications because the bandwidth and reliability will have vastly improved.

Who knows? Maybe we'll be using quantum entanglement to communicate instaneously around the world.

I'm not saying that any of these things are going to be a reality. I know most of them are a long way off. But they will happen someday, how can you be so sure it won't happen in 15 years?

Look at the story with computers. A processor that seems like the state of the art now will be agonizingly slow in 2 years, and in 5 years will be so far behind it can't even run the latest operation system. Why assume telecom won't evolve the same way?

mma
50%
50%
mma,
User Rank: Light Beer
12/4/2012 | 10:03:55 PM
re: Did WorldCom Puff Up the Internet Too?
By the end of the MPEG 1 work ('91), we knew that entertainment required 1 Giga Op/s processors for <$100, 1 GB disks for <$100, and 1 GB to the home for <$1. Alas, the telecom industry didn't make the last number, while the semiconductor and storage industries have made theirs. So CD-RW, DVD and TIVO rule, and video over the telecom network is going nowhere (unless you consider satellite and cable to be telecom networks).

OTOH, video is pretty exclusively an entertainment medium. There is no experimental evidence that I've seen that shows that moving pictures are more effective for collaborative work, training and education, or any other business purpose. There are reports that even motor skills are taught just as well using stills, audio and text.
beowulf888
50%
50%
beowulf888,
User Rank: Light Beer
12/4/2012 | 10:03:56 PM
re: Did WorldCom Puff Up the Internet Too?
mma:
How did you derive your figures? I don't doubt you saw what you saw, but I've also heard organizations like Telegeography state that, for the long haul carriers at least, data bandwidth exceeded voice around '98/'99 (sorry I don't have a link available at the moment).

The fact that the RBOCs were whining (in the '98 to '00 time frame) that dial-up Internet usage tying up the ports on their switches doesn't seem to jibe with your figures. Would you consider dial-up Internet to be "voice" because it was using a voice circuit?

--Beo

mma wrote:
"In early 2000 I did some estimates of transmission demand growth rates. At the time (end of '99 data) it appeared that the traffic load was 68% voice, 19% Internet, 3% other data networks, and 10% leased line. The growth rates were estimated to be 10%, 100%, 100%, and 30% respectively. If you project the loads based on the relative size of each traffic load times its relative growth rate, the total traffic load would increase 31% during 2000, 41% during 2001, 53% during 2002, and 66% during 2003."
beowulf888
50%
50%
beowulf888,
User Rank: Light Beer
12/4/2012 | 10:03:56 PM
re: Did WorldCom Puff Up the Internet Too?
rjmcmahon wrote (with tongue in cheek?):
>To the residential market the internet is >entertainment - nothing more.

It sounds like you're knocking entertainment as a source of revenue, but maybe I'm reading more into your koan than I should. The entertainment industry (radio, TV, movies, porn, etc.) probably generates more 100 gigabucks per year. If the carriers were smart, they'd subsidize high-speed access to the home. Then entertainment vendors could pay carriers to transport their wares. Of course, most of the media companies are scared of the Internet at the moment...

But I think home use of the Internet is more than just entertainment. Even in this recession, revenue from on-line transactions is still growing at an immense rate.

Still an optimist...
--Beo
beowulf888
50%
50%
beowulf888,
User Rank: Light Beer
12/4/2012 | 10:03:56 PM
re: Did WorldCom Puff Up the Internet Too?
Lowbandwit and Light-bulb:
FYI, it makes economic sense to lay as much fiber as you can in one fell swoop. I forget how many of hundreds of strands are in a PVC sheath (and I'm sure vendors sell a multiplicity of sheaths with different fiber counts). But once you dig the ditch, it's most cost-effective to lay as much cable as possible -- so you never have to go back and lay any more. Even if you leave most of strands dark, laying the cable is a one-time cost.

So the vast oversupply of dark fiber isn't really an economic albatross for the long haul carriers. Rather it was the vast investments in the equipment to light the fibers and move the data that were the albatrosses to the carriers.

cheers,
--Beo

Lowbandwit wrote:
>Now THAT's what I'm talking about! L3 put so
>much fiber in the ground it's mind boggleing.
>But, at the same time so did several other
>players. I don't care WHO you are, that takes
>balls- and ignorant investors.

Light-bulb wrote:
>We have so much fiber in the ground its s
>imply unbelievable! If you put routes together
>from multiple carriers to identify totals
>like you mentioned 100Tbps... using my math
>above the capacity we have in the ground is >probably more around ~5000Pbps Yes that
>Petabits! Of course why use DWDM when you have
>so much excess fiber you can by routes for
>pennies on the dollar vs. cost of installation!
cyber_techy
50%
50%
cyber_techy,
User Rank: Light Beer
12/4/2012 | 10:03:57 PM
re: Did WorldCom Puff Up the Internet Too?
Right on the money, mama
lob
50%
50%
lob,
User Rank: Light Beer
12/4/2012 | 10:03:57 PM
re: Did WorldCom Puff Up the Internet Too?
LightBulb -- thank you for the correction (OC-48 -> OC-192).

WRT actual fiber capacity -- I do not think you can get that fast with the actual fiber and ability to place repeaters; going fast requires shorter spans which may be economical in Japan, but in US with its large distances the costs are going to be excessive. There's also an issue of space and power in COs and repeater locations.

Even assuming that this figure is real, we end up with log2(768e12/10e9) ~ 16 years, or 2.5 years off my initial estimate (13.5 yrs).

As you see, my estimate is pretty much the same; exponents are growing _fast_. In fact, the time estimate for capacity consumption does not depend much on the amount of fiber in the ground, but is very sensitive to the rate of traffic growth. My estimate used 100% growth per year (per Odlyzko); conservative estimates of unknown origin (i.e. 30%) would yield about 40 years to get fiber exhausted. Still not 1000 years, and I hope to live to see the new backbone build-up :)

BTW, the figure 768Tbps applies to the single route - so you should compare it to the todays single route in Internet backbone (i.e. 10Gbps),
because, obviously, today's total capacity is far beyond OC-192. Apples to apples :)

Now, let's estimate the relation of the total carrying capacity to the circuit speed. A typical backbone has something like 70% of traffic going to other backbones, and thus crossing exchange points. Let's assume total of
40 exchange circuits between our provider and other Tier-1 backbones (going to larger number doesn't scale due to routing issues). That gives total carries capacity of a backbone on the real-life traffic pattern at about 60x circuit speed. With 5 Tier-1 backbones, the total carrying capacity is, therefore 300x circuit speed, or 3Tbps today.

With your figure of 768Tbps per circuit we end up with 230 Pbps (not 5000!). Packets generally have to cross more than one circuit, and that's why your result is off.

The upper limit on traffic is, obviously, in the ability of end-nodes to generate traffic. Today's PCs can easily generate 100Mbps; let's assume there is 100mil of them in the US. This gives total peak traffic of 100Tbps. With Moore's law (at 2yr doubling) that will be 1600Pbps in 14 years. Obviously, the duty cycle is rather low (1/1000 or so) and that would delay saturation by 20 years (i.e. 35 years is about right). This result does not take into account the fact that new applications (grid computing, for example) tend to create a lot more traffic than human users.

So... Different methods of estimating traffic vs capacity yield spread of approx 15 to 35 years to get all the fiber utilized. I would not say that the money was all wasted, however the ROI is not attractive. The fiber build-out should've been a federal program, like interstates, not a commercial project - since most economical benefits of having transport infrastructure are not gathered by the network/road owners.

OTOH, one may consider conning gullible investors a form of tax :) In any case, this fiber will be the workhorse of US economy for many years.
mma
50%
50%
mma,
User Rank: Light Beer
12/4/2012 | 10:03:58 PM
re: Did WorldCom Puff Up the Internet Too?
In early 2000 I did some estimates of transmission demand growth rates. At the time (end of '99 data) it appeared that the traffic load was 68% voice, 19% Internet, 3% other data networks, and 10% leased line. The growth rates were estimated to be 10%, 100%, 100%, and 30% respectively. If you project the loads based on the relative size of each traffic load times its relative growth rate, the total traffic load would increase 31% during 2000, 41% during 2001, 53% during 2002, and 66% during 2003.

However, due to the slump in the economy and due to technical advances, I believe that these growth rates are too optimistic in the later years. Less transmission capacity is used when a person interacts directly with a company's web server than when a person talks to a customer representative at a CRM terminal at a company. Companies such as banks are known to be removing CRM seats as their web services grow. Even less transmission capacity is used when a server at one company talks to a server at another company, now becoming more prevalent in B to B interactions.

Also, most likely 10% growth is too high for voice, and in the future voice may shrink due to VoIP and the use of packetized compressed speech coding, which is typical of wireless services.

So long term, the network will probably grow closer to the 30% long term growth rate for data than rising asymptotically to the 100% annual growth rate of the Internet.

At the 30% annual growth rate it takes about 13 years to soak up the capacity multiplication of going from 1 wavelength per fiber to 32 wavelengths per fiber using DWDM -- or about 18 years if you also go from OC48 to OC192 with the same upgrade.
broadbandboy
50%
50%
broadbandboy,
User Rank: Light Beer
12/4/2012 | 10:04:01 PM
re: Did WorldCom Puff Up the Internet Too?
jepovic: "Bah. Here, at one of Europe's larger backbones, we use MRTG graphs like everyone else. When the traffic on a link exceeds some value, say 50%, you upgrade. If you need a trend, look at the one-year graph. Sure it's based on "rearview-mirror analysis", but so is all forecasts, right?"

-------------------------------------------

I agree that is how it is done, and that's the problem. Trying to predict the future based on current trends in a volatile market like the Internet is dangerous. When things were hot, ISPs placed orders for equipment well in advance of demand, becuase of delays in getting the equipment.

The big system vendors, worried about shortages of critical components that would degrade their ability to meet customer demand, sought to increase the supply of key optical components.

Component vendors, seeing a big backlog of orders, added new capacity. Then all of a sudden, something happened. ISPs started delaying purchases of more routers, and vendors up and and down the supply food chain were left holding the bag. The channel was stuffed full, with no place to go.

The rest is history. Rear-view analsys led to massive overinvestment and massive layoffs. I say we need to do a better job of predicting both bandwidth demand and variability of that demand, not to mention doing a better jop of calculating the price elasticity of that demand. I would say that is where we really blew it.

---------------------------------------

"Compare with the highway system: Choose between trying to calculate the number of cars on a certain stretch by 1) Building a model based on number of cars, population growth, gas price, etc, or 2) Measuring the actual amount of cars. Only
difference is that measurements are built-in in routers."

------------------------------------------

You are helping to prove my argument. 1. is much better. No competent urban planner would just wait until the daily number of cars exceeds a certain threshold, as in 2., and then start planning to widen the highway. The result would be either years of congestion (daily rush hour traffic jams) or huge overcapacity and wasted investment.

More likely, planners will try to project demographic growth trends out 10-15 years, and start planning the transportation system that will be needed to accommodate the expected future population.

BBboy
broadbandboy
50%
50%
broadbandboy,
User Rank: Light Beer
12/4/2012 | 10:04:02 PM
re: Did WorldCom Puff Up the Internet Too?
"It's good to hear the LR boards are such a newsbreaking mechanism!"

Scott, I don't know about newsbreaking, but these boards are the best place to discuss the news after it's broken.

We're a lot more fun than NANOG...

BBboy
lowbandwit
50%
50%
lowbandwit,
User Rank: Light Beer
12/4/2012 | 10:04:04 PM
re: Did WorldCom Puff Up the Internet Too?
If you put routes together from multiple carriers to identify totals like you mentioned 100Tbps... using my math above the capacity we have in the ground is probably more around ~5000Pbps Yes that Petabits!

________________________

Light-bulb-
Now THAT's what I'm talking about! L3 put so much fiber in the ground it's mind boggleing. But, at the same time so did several other players. I don't care WHO you are, that takes balls- and ignorant investors. I mean, THAT'S A LOT OF PORN! ;-)
rjmcmahon
50%
50%
rjmcmahon,
User Rank: Light Beer
12/4/2012 | 10:04:05 PM
re: Did WorldCom Puff Up the Internet Too?
To the residential market the internet is entertainment - nothing more.
_____________

The following document surveys internet usage patterns.

http://www.ntia.doc.gov/ntiaho...

We are only at the beginning. It will go way beyond entertainment. We should be careful not to underestimate what ubiquitous participation brings to the table.
switchrus
50%
50%
switchrus,
User Rank: Light Beer
12/4/2012 | 10:04:06 PM
re: Did WorldCom Puff Up the Internet Too?
According to Yahoo, the "Fat Lady" seems to be warming up for the end of the tragic opera called World Com, well maybe not the end of the opera.

http://story.news.yahoo.com/ne...
Light-bulb
50%
50%
Light-bulb,
User Rank: Light Beer
12/4/2012 | 10:04:07 PM
re: Did WorldCom Puff Up the Internet Too?
Lob,

Check out this site. Also a technicality on your post... OC-48=~2.5Gbps OC-192=~10Gbps
http://pr.fujitsu.com/en/news/...
Now take that into account on some of Level 3's ditches. With 4x216strand fibers ducts in the ground... you end up with 1.75Tbps x 108Pairs x 4 = 756Tbps capable per route! Don't forget if you are using Diverse routes two seperate ditches to each major hub destination... Then think PCA.
We have so much fiber in the ground its simply unbelievable!
If you put routes together from multiple carriers to identify totals like you mentioned 100Tbps... using my math above the capacity we have in the ground is probably more around ~5000Pbps Yes that Petabits!
Of course why use DWDM when you have so much excess fiber you can by routes for pennies on the dollar vs. cost of installation!
But we're talking theoretical here :)

Now think about that Capacity! Thats a whole lot of DS3's!!! Let me put it another way... Think every single person in the US, oh around 285Million of us... All with an OC-48. Thats how much theoretical bandwidth exists in our fiber today with Todays technology. Think about 5 years later with the advances in DWDM... Eeek!

Cheerio,
let-there-be-light
50%
50%
let-there-be-light,
User Rank: Light Beer
12/4/2012 | 10:04:07 PM
re: Did WorldCom Puff Up the Internet Too?
Today, many of us dismiss the erstwhile "visionary leaders" of Lucent and Nortel's optical businesses as supreme idiots, however, I don't think any of us believes that they were really so idiotic as to spend hundreds of millions of dollars of R&D money on pipedreams.

I think it is more realistic to believe that they were "gently pressured" into investing huge sums of money into projects such as the all-optical switch that can handle THOUSANDS of wavelengths, by companies such as WorldCom, who told them that they would be needing this tomorrow, or at the very latest, a day after that. In turn, the system vendors put enormous pressure on the component vendors.

Just goes to prove that

IT'S ALL WORLDCOM'S FAULT!!
billyjoebob
50%
50%
billyjoebob,
User Rank: Light Beer
12/4/2012 | 10:04:08 PM
re: Did WorldCom Puff Up the Internet Too?
This is a joke - right?? Rip out all the copper in NA to replace it so (with little price increase it bankrupt every ILEC out there) we can push a high bandwidth application to justify the overbuilding??

But this points out a long standing paradym - the "Killer Application" that will prove once and for all the value of the "Internet".

Streaming videos, video on demand, real time video conferencing have all be touted as the Killer app - trials have been done, and are continuing - but take up rate is slow.

There seems to be strong evidence that to the business community "Internet" = reduced costs over current WAN costs. No bottom line impact (make my real dollar costs go down) not percevied bottom line (over x years with x growth and x connections Y dollars are saved by reduced mangement/downtime/upgrades . . .)and no move to internet connections.

To the residential market the internet is entertainment - nothing more. Sure it's a usefull tool but most use it just to get email, surf for porn and let their kids chat online. The slowing of DSL deployements (I didn't say flattening) seem to indicate that we may be in sight of maximum market penetration.

IF (a big if I grant) that these two assumptions have some value - then realistic growth figures for Internet traffic in NA could be done. So for those who say we can't project forward to what the growth will be - I disagree - I would hazard a guess that it has been done.

lob
50%
50%
lob,
User Rank: Light Beer
12/4/2012 | 10:04:09 PM
re: Did WorldCom Puff Up the Internet Too?
> We've known for a long time that any one of the
> major carriers had enough fiber capacity to
> carry all the Internet traffic, but now... does
> this mean that we're set for the next 1,000
> years or so of growth?

Boy, the math is hard!

Today's Tier-1 Internet backbones are OC-48 (10Gbps), carriers have about 100Tbps worth of fiber capacity (projecting use of fancy DWDM) in the ground. This is growth 10000 times. With 1-year traffic doubling (which still holds accouding to Odlyzko's research) this corresponds to 13.5 years of growth (10000 ~ 2^13.5).

Building infrastructure which is going to be full in less than 15 years is not stupid. Thinking that revenues will grow in line with traffic is.

The $1B questions are - what kind of equipment will be switching all that traffic, and how to haul it across tail circuits.
st0
50%
50%
st0,
User Rank: Light Beer
12/4/2012 | 10:04:09 PM
re: Did WorldCom Puff Up the Internet Too?
"Maybe that's what happened to the dinosaurs?"
------------
or same as the cancer cell growth mechanism...We are going through the chemotherapy period now...

st
jepovic
50%
50%
jepovic,
User Rank: Light Beer
12/4/2012 | 10:04:09 PM
re: Did WorldCom Puff Up the Internet Too?
The ability to predict future traffic growth and bandwidth demand with reasonable accuracy is critical for planning future network builds and anticipating router demand by service providers and vendors. Yet I don't know of any reliable method for doing so. I have talked to traffic engineers at some Tier-1 ISPs, and it sounds like a lot of rear-view mirror analysis projected into the future.
_________________________________________________

Bah. Here, at one of Europe's larger backbones, we use MRTG graphs like everyone else. When the traffic on a link exceeds some value, say 50%, you upgrade. If you need a trend, look at the one-year graph. Sure it's based on "rearview-mirror analysis", but so is all forecasts, right?

If you instead build your network according to theoretical models, like UUnet seems to have done, it can awfully complex and still completely useless. Planning based on actual measurements is really something my mom could do.

Compare with the highway system: Choose between trying to calculate the number of cars on a certain stretch by 1) Building a model based on number of cars, population growth, gas price, etc, or 2) Measuring the actual amount of cars . Only difference is that measurements are built-in in routers.
Scott Raynovich
50%
50%
Scott Raynovich,
User Rank: Light Sabre
12/4/2012 | 10:04:13 PM
re: Did WorldCom Puff Up the Internet Too?
"this story was extensively discussed on these boards at least a year ago"

It's good to hear the LR boards are such a newsbreaking mechanism!
wkdpssh
50%
50%
wkdpssh,
User Rank: Light Beer
12/4/2012 | 10:04:13 PM
re: Did WorldCom Puff Up the Internet Too?
I think the real killer was DWDM and its relatives that allowed bandwidth to be subdivided into smaller and smaller pieces. Even if the internet continued growing at much higher rates the technology would be there to make the bits cheaper and cheaper. Not the fault of Ciena who just did what was right for them to make money, but shows that it's dangerous to bank on the value of a fiber optic pipe when the amount of data that can be transmitted is only limited by the equipment on either end. The pipe itself is virtually unlimited in capacity. I think what happened to telecom and the fiber glut was bound to happen (I've personally been predicting it since 1998). It just happened a bit sooner because the growth was "only" 100% instead of 300% or whatever.
broadbandboy
50%
50%
broadbandboy,
User Rank: Light Beer
12/4/2012 | 10:04:14 PM
re: Did WorldCom Puff Up the Internet Too?
Skeptic, I know for a fact that some VCs were using the UUnet stats to justify funding for certain start ups.

I also know that not everybody swallowed that number. My contacts working on Cisco's GSR talked to their own customers, and said growth was more like doubling every six months. UUnet was considered an anomaly even then.

BBboy
lightdimming
50%
50%
lightdimming,
User Rank: Light Beer
12/4/2012 | 10:04:14 PM
re: Did WorldCom Puff Up the Internet Too?
Pass a law requiring all the phone lines must be upgraded to handle 10Mbps or faster without raising much the price. Box-Bluster starts delivering/streaming movie on internet at a fractional cost. You will see internet traffic grows much more than Worldcom expected.
skeptic
50%
50%
skeptic,
User Rank: Light Beer
12/4/2012 | 10:04:14 PM
re: Did WorldCom Puff Up the Internet Too?
But it is amazing to me that supposedley intelligent people would take that one data point and use it to base vc funded business plans.
------------------

Many people didn't even have business plans and
whatever they are saying now, the ONLY thing
most of them cared about was M&A valuations
and stock prices in the space.

Nobody cared about how fast worldcom said the
internet was growing. Just like nobody cared
what Enron was doing as long as they said they
made their quarterly numbers.


vapa
50%
50%
vapa,
User Rank: Light Beer
12/4/2012 | 10:04:14 PM
re: Did WorldCom Puff Up the Internet Too?
Maybe that's what happened to the dinosaurs?

According to the movie, Reign of Fire, they were killed by fire breathing dragons. :D
FinBurger
50%
50%
FinBurger,
User Rank: Light Beer
12/4/2012 | 10:04:15 PM
re: Did WorldCom Puff Up the Internet Too?
Au contraire, mon frere...

Removal of bribes (options) == New-found objectivity

Not accounting for concentration in data traffic is about as wrong as you can get.

Anybody who believed those exponential traffic graphs would continue forever was a fool. The real question was, "when will it saturate?". The answer, we now know, was 18:25 EST, March 13, 2000. Too bad.

Thank God it happened when it did. Imagine if the insanity went on for a few more years. We might have reached the Telecom Event Horizon, where it would be unprofitable to do anything but work on the Internet. Food production would have ceased, and mankind would be extinct.

Maybe that's what happened to the dinosaurs?

:-)
broadbandboy
50%
50%
broadbandboy,
User Rank: Light Beer
12/4/2012 | 10:04:15 PM
re: Did WorldCom Puff Up the Internet Too?
This article was posted here a year ago, but is still relevant. It explains the issue about as well as anything I have seen in print.

http://www.broadbandpub.com/br...

BBboy
BobbyMax
50%
50%
BobbyMax,
User Rank: Light Beer
12/4/2012 | 10:04:15 PM
re: Did WorldCom Puff Up the Internet Too?
Many companies in the telewcom sector lie to boost their stock value. They bring the stock up and up --- and finally they are ready to excercise their stock options. Using this practice Cisco was able to split its stock several times a year. In the year 2000 the stocks were increasing without bounds. All on-line stock trade center were publishing false and misleading information almost on every stock.

Maximum number of companies went public during the year 2000. It should have alerted someone, but it never happened.

In effect, many companies became highway robbers with no ethics and morality. They are criminals who are sitting on millions of dollars where as the employees and share holders lost trillions of dollars.

Under no circumstances can Board of Directors be trusted. Almost allthe companies arew packed with the friends and relatives of CEOs.

Our system has built-in-corruption which cannot be erased unless the basic laws of corporate governance is changed. The corruption is so widespread that it is unlikely that any thing would change.
oc-3072
50%
50%
oc-3072,
User Rank: Light Beer
12/4/2012 | 10:04:16 PM
re: Did WorldCom Puff Up the Internet Too?
skeptic is right. if WCOM had to provision a T1 private line, it doesn't matter whether the customer used the whole thing or left it empty... that's a pipe provisioned on the network.
broadbandboy
50%
50%
broadbandboy,
User Rank: Light Beer
12/4/2012 | 10:04:16 PM
re: Did WorldCom Puff Up the Internet Too?
Hello--this story was extensively discussed on these boards at least a year ago, in response to the Larry Roberts studies of Internet traffic growth.

It is clear today that Sidgmore and O'Dell were talking about growth of the network infrastructure, and that was confused with growth of actual traffic.

I have seen ISP internal data on the growth of router ports in certain cities that did indeed show a doubling every 3-4 months during that time period.

But it is amazing to me that supposedley intelligent people would take that one data point and use it to base vc funded business plans.

The big question today is whether the growth rate is 1x, 2x, 3x, etc. An even more important question is, what will the rate of traffic growth be in 6 months, 12 months, 18 months, 24 months, etc.

The ability to predict future traffic growth and bandwidth demand with reasonable accuracy is critical for planning future network builds and anticipating router demand by service providers and vendors. Yet I don't know of any reliable method for doing so. I have talked to traffic engineers at some Tier-1 ISPs, and it sounds like a lot of rear-view mirror analysis projected into the future.

That's why so many companies got caught with their pants down when the bottom fell out.

BBboy

skeptic
50%
50%
skeptic,
User Rank: Light Beer
12/4/2012 | 10:04:18 PM
re: Did WorldCom Puff Up the Internet Too?
Here's how it worked, according to the former Worldcom employee: Worldcom would hook up new customers with connections capable of handling, say, up to 1.5 Mbit/s of data, knowing that for most of the time the lines would only carry a fraction of this amount. Worldcom would then use the 1.5 Mbit/s figures, not the actual traffic figures, when citing Internet traffic growth statistics.
======================

Which is a perfectly valid and meaningful way
of measurment in order to make certain
decisions about the network.

Your article and your thinking is terribly
confused. There are lots of different ways
to measure growth in the network. And often,
different numbers for growth/traffic are
meaningful for different sectors of the
equipment marketplace. The core isn't the
same as the edge isn't the same as access.

Worldcom has to build/operate an entire network
and just because they didn't cite traffic
growth numbers that were appropriate for vendors
of core optical equipment and routers doesn't
make those numbers invalid.








lowbandwit
50%
50%
lowbandwit,
User Rank: Light Beer
12/4/2012 | 10:04:19 PM
re: Did WorldCom Puff Up the Internet Too?
Well, if they did work those numbers... how many billions got wasted by companies building out their networks to meet that demand? We've known for a long time that any one of the major carriers had enough fiber capacity to carry all the Internet traffic, but now... does this mean that we're set for the next 1,000 years or so of growth?
jggveth
50%
50%
jggveth,
User Rank: Light Beer
12/4/2012 | 10:04:19 PM
re: Did WorldCom Puff Up the Internet Too?
"Global Crossing and 360networks will battle for global supremacy, but in a trillion-dollar market, there will be no loser."
- George Gilder, Forbes, 2/19/01

"Back to nature"
- inquisitive, LR, 7/18/02
cessna
50%
50%
cessna,
User Rank: Light Beer
12/4/2012 | 10:04:20 PM
re: Did WorldCom Puff Up the Internet Too?
During WCOM/MCI merger, UUNET, Compuserve, ANS and MCI provided extensive IP traffic data to the the European Commission and DOJ. That data (much to WCOM/MCI's dismay) did show the fantastic traffic growth rates for 1996-1998(Q1) period. SOB @ Harvard may be a good person to interview since he testified on behalf of MCI/WCOM.

Not sure if those records are public but check'em out.
GreatFallsNetworks
50%
50%
GreatFallsNetworks,
User Rank: Light Beer
12/4/2012 | 10:04:20 PM
re: Did WorldCom Puff Up the Internet Too?
Ex worldcom employess == sour grapes
Featured Video
Flash Poll
Upcoming Live Events
September 17-19, 2019, Dallas, Texas
October 1-2, 2019, New Orleans, Louisiana
October 10, 2019, New York, New York
October 22, 2019, Los Angeles, CA
November 5, 2019, London, England
November 7, 2019, London, UK
December 3, 2019, New York, New York
December 3-5, 2019, Vienna, Austria
March 16-18, 2020, Embassy Suites, Denver, Colorado
May 18-20, 2020, Irving Convention Center, Dallas, TX
All Upcoming Live Events