Micro MSPPs Are Big

The demand for Ethernet services and the buildout of 3G networks have made a star of micro multiple service provisioning platforms, or micro MSPPs.

A micro MSPP is a small, low-priced, low-capacity switch that aggregates network traffic (in various protocols) onto Sonet/SDH pipes at the edge of the network so that the core network can be simplified and operated more efficiently.

The micro MSPPs stand between 1 and 2 rack units (RU) in height (they can be modular) and usually cost between $3,000 and $8,000. For the small size and small price tag, concessions must be made -- the units' architecture is usually not as fully redundant as larger Sonet platforms.

The micro MSPP concept grew from two realities in the service provider business. First, service providers need to leverage their deep investment in Sonet. Second, they need to satisfy the growing demand for Ethernet-based services. The micro MSPP allows service providers to extend their Sonet/SDH networks to “customer located” sites, making provisioning a smarter, smoother, and less costly affair.

“It is really all about pushing the smarts out to edge of the network where the customers are,” says William Quigley of Clearstone Venture Partners. Quigley says older-generation MSPPs were built to sit in the service provider’s central office (CO), aggregating and switching the traffic for a large area. The old MSPPs therefore needed much more switching capacity (usually 160 Gbit/s to 300 Gbit/s).

But moving all that traffic from the network's edge back to the CO is expensive. Nowadays, much larger traffic loads are being moved in that direction. So carriers need a way to do that switching at the edge so as to eliminate all the backhauling to the CO. The way to do it is to distribute that switching power more finely to more locations at the edge of the network. Since such a scenario creates more switching locations, these new MSPPs can get away with having far less switching capacity. Still, they need to be far less expensive than their $40,000 predecessors to justify the scheme, Quigley says.

The micro MSPPs look good on paper; but what applications are propelling their popularity? As discussed in a Light Reading Webinar held last week, the delivery of Ethernet services will be a major carrier capital expenditure (capex) priority this year. The micro MSPP will likely emerge as a crucial piece of hardware for carriers racing to delivering those services efficiently, according to Andrew Knott of White Rock Networks Inc., which has been selling a micro MSPP since August.

“The reality is the carrier will offer Ethernet to new customers it is trying to take away from somebody else providing basic DSL service,” Knott says, “Some of the smart ones will offer Ethernet to their existing customers so that nobody can steal them, but usually [carriers offering Ethernet] are trying to take subscribers away from competing carriers.”

Besides Ethernet service delivery, 3G network requirements may also be driving a big demand for micro MSPPs. Specifically, micro MSPPs are being sought out for the ability to streamline the movement of wireless traffic among access points.

Clearstone's Quigley says it is no coincidence that places where 3G technology is hot -- like China, Korea, and Japan -- are also the places where demand for micro MSPPs will most likely turn into real sales. China Telecommunications Corp. (NYSE: CHA), Korean incumbent KT Corp., and India’s Bharat Sanchar Nigam Ltd. (BSNL) all have RFPs or RFQs out for micro MSPPs, according to Paul Nahi, CEO of Crimson Microsystems Inc., which supplies components to the makers of micro MSPPs.

The new technologies are nice, but the central appeal of the micro MSPP, from the service provider's point of view, is simply economic: It leverages Sonet, the world's dominant telecom networking technology.

“There has been a substantial investment globally in SDH and Sonet technology in core networks -- I think its something upwards of $300 billion of installed base,” says Michael Crossey of Transmode Systems AB. “And with having those micro MSPPs with those interfaces, it really gives those networks a new lease on life, and allows operators to deploy new data services, high-capacity services, which traditionally weren’t supported on legacy SDH/Sonet.”

Micro MSPPs also support storage networking protocols such as Escon, Fibre Channel, and Ficon, making them a good fit for data centers and medium to large enterprise customers. In fact, many service providers first began buying micro MSPPs to serve their enterprise customers.

“In a full-blown MSPP it is really expected to have a fully redundant system where there is no single point of failure. In the case of the micro MSPP very often there is a tradeoff made, because one its big attributes is cost,” says Transmode’s Crossey.

Micro MSPPs on the market now include ADVA Optical Networking's (Frankfurt: ADV) FSP 1500; Verilink Corp.'s (Nasdaq: VRLK) Orion 5000; Tejas Networks India Ltd.’s TJ100; Transmode’s TSE; and White Rock’s VLX2006. Cisco Systems Inc. (Nasdaq: CSCO), ECI Telecom Ltd. (Nasdaq/NM: ECIL), and Luminous Networks Inc. have also announced gear in this space (see Luminous Thinks Small, ECI Intros Small MSPP and MSPPizza to Go at Cisco).

— Mark Sullivan, Reporter, Light Reading

douggreen 12/5/2012 | 3:27:59 AM
re: Micro MSPPs Are Big Viewpoint,

Agreed on China, India, and Europe. It is already happening.
douggreen 12/5/2012 | 3:28:03 AM
re: Micro MSPPs Are Big Viewpoint,

Some of the startups have been shipping over a year, and some of them have experience in the US carrier market. A few of them do already offer both ends of the system. I would give the edge in the experience factor to the US and Japanese (i.e. Fujitsu) companies as they are closer to the customer. Believe it or not, the RBOCs and IXCs expect you to give them what they want without telling you what it is.
Viewpoint 12/5/2012 | 3:28:03 AM
re: Micro MSPPs Are Big Doug,

Those US and Japanese vendors will get the big USA deals. The Chinese will get the deals in places like India and China. You also see the Chinese making headway in Europe. Between the big USA/Japanese vendors who have existing relationships with the carriers and Chinese vendors with their low cost equipment, the startups are in tough position. I don't see RBOCs and IXCs buying anything from smaller companies. The partner companies of Turin (UT and Motorola) and Luminous (SA and Ciena) don't know how to sell to the RBOCs and IXCs. Tellabs may be only slightly better. I don't know how much it has helped White Rocks. The other startups like Overtures have even tougher time.
Viewpoint 12/5/2012 | 3:28:12 AM
re: Micro MSPPs Are Big >As I posted before, the CPE box is only half the
>equation. Unlike the larger MSPPs, who can carry
>data in one STS and voice in another, micro-mspps
>have to mux voice and data into a single STS or one
>or more VTs for efficiency.

This is the reason the Chinese have an advantage over startups. Demultiplexing the voice and data upstream is the key. I very much doubt if one micro-MSPP vendor's multiplexing of voice and data in GFP, Cendor'oncatenated SONET payloads or RPR on its OC3/OC12 trunk can be demultiplex by another vendor's MSPP.

It is possible to have good OAM features while keeping the cost under $5,000. It is very hard to make startup understand carrier requirements. In my experience it takes a startup at least a year after FCS to have a shippable product that carrier can buy.
Viewpoint 12/5/2012 | 3:28:12 AM
re: Micro MSPPs Are Big Exalted moved its operations to India (for cost reasons?). They are far from having any shipping MSPP product. India is not a country where you can find ready made talent for creating products like MSPP. As other poster pointed out Huawei, UT and ZTE are ahead of the curve. They all have good traction in Indian market. Their main advantage is willing to sell at less than 10% margins.

douggreen 12/5/2012 | 3:28:12 AM
re: Micro MSPPs Are Big At first glance, it would appear that lowest cost is all that matters. However, these boxes have to be cheap operationally as well. They have to be easy to configure (preferably self-configuring). Since they sit on the customer demark, they also have to have good OAM features (test, loopback, etc).

This means that there is room to compete partially on features rather than 100% price, but not a LOT of room. Enough, I think, to allow some US companies who understand carrier networks to compete, provided they meet the cost clip-level.

As I posted before, the CPE box is only half the equation. Unlike the larger MSPPs, who can carry data in one STS and voice in another, micro-mspps have to mux voice and data into a single STS or one or more VTs for efficiency. This can be done by carrying everthing over packets using circuit emulation for voice, or by concatenating VTs or STSs for data. There has to be a corresponding box at the other end within the carrier network to demux the voice and packets. This article only covers half the story.
quikshe12 12/5/2012 | 3:28:14 AM
re: Micro MSPPs Are Big
Technically, yes, it is a US company. It is incorporated in Delaware but all operations are in Bangalore. Just a sales office in US I think.

sigint 12/5/2012 | 3:28:14 AM
re: Micro MSPPs Are Big quikshe12:
Xalted Networks is another company (indian based) that is coming out with a new MSPP. Looks impressive so far and being built at a low cost...

Xalted does have an office in Bangalore - but aren't they a US start-up?
quikshe12 12/5/2012 | 3:28:14 AM
re: Micro MSPPs Are Big
It makes sense that the market is getting crowded, it is one of the few comm equipment categories that has a strong economic case and many willing service provider customers. I agree with the earlier comment, however, that this will be a market won by the likes of ZTE, UTStartcom, Hauwei, and other offshore vendors. Xalted Networks is another company (indian based) that is coming out with a new MSPP. Looks impressive so far and being built at a low cost...
quikshe12 12/5/2012 | 3:28:14 AM
re: Micro MSPPs Are Big
You raise a good point regarding limited gross margin opportunities for products selling under $10K. One correction, however, the financial impact would be recorded on the income statement, not the balance sheet. A balance sheet records a company's assets and liablities. An income statements records a company's operating results.
douggreen 12/5/2012 | 3:28:15 AM
re: Micro MSPPs Are Big Oldpots,

I see a combination of opportunistic one-at-a-time deployment for the masses with targeted fiber builds to strategic locations. Right now, the cash outlays for an "if you build it, they will come" network seem a bit unpalateable for a CEO trying to please Wall Street. This is especially true as attention focuses on growth via consolidation. At the same time, they have to keep up a good fiber show to apear "visionary."

I expect the larger buildout to be based on the financial success of the initial deployments.
OldPOTS 12/5/2012 | 3:28:24 AM
re: Micro MSPPs Are Big Speaking of the App.

Do you think they will continue be do one at a time apps/impementation using Micro MSPPs, at low volume $$$ costs?

Or, DSL with pairs or fiber access links to the CO, like FTTU, for more penetration and with lower volume costs?

(Including customer device, transport and CO aggregator costs)

I see a migration to a valuable enterprise market(1/2) with higher ARPU. Less truck rolls to instal with higher ARPU. Less OPEX risk. I have seen practicle proposals that include FTTU/B to leverage both home, apartments, enterprises and small business coverage. But with full buildout to enterprises and shopping centers first.
These also included the oppertunity to propose reduced prices Micro MSPPs as an alternative migration.

douggreen 12/5/2012 | 3:28:27 AM
re: Micro MSPPs Are Big This article, while informative, tries to define the micro MSPP market by the technology inside the box rather than the application. I often find it more helpful to look at the products in terms of the application.

The issue that SPs face is that they want to deliver both voice and data, price it lower than competition, and still make money (duh). This is easy to do if you build a business case based on a massive network buildout and assumed take-rates (like DSL). This is NOT what the service providers want to do, however. They want to be opportunisic in responding to individual customer requests, and still make money. They can do this by leveraging the existing infrastructure.

Aside from ongoing operational costsw, there are three aspects to their cost: the cost of the transport link, the cost of the CPE, and one other piece of equiment that I'll talk about later. Larger MSPPs allow you to lower line cost by using a single link, but the business case never closes until you have several customers on the same box. The service providers HATE having to speculate on getting more customers. They want to do a per-application business case. The goal of a micro-MSPP is to get the product cost low enough to justify a per-customer business case. (Of course, they would still like to make even more money with multiple customers). With ATM based micro-MSPPs back in the 1990s (which tried to solve the exact same problem), the clip level was somewhere between $2000 and $5000.

The other issue that this article doesn't address is this: the CPE is only one end of the problem. The voice and data are multiplexed at the customer end using a variety of mechanisms. You can turn everything to packets and use circuit emulation to carry the voice. You can carry voice native and use virtual concatenation to carry data. Both have their advantages, depending on how the rest of your network is built. Regardless, you have to have some sort of device at the other end to de-multiples the voice and data to feed them into their respective backbones. That device is also a part of the cost of the service as well, and it has a whole different set of requirements.

Photon_Got_Mad 12/5/2012 | 3:28:27 AM
re: Micro MSPPs Are Big Anyone has any idea what is the cost on the Opti6100?
particle_man 12/5/2012 | 3:28:29 AM
re: Micro MSPPs Are Big "Don't look for a Micro MSPP to pull any company out of troubled waters. At best you'd sell these products through a VAR at volume rates. "

There is another angle here which is the "CLE" discussed in some length on this site already. If Verizon were to decide to dump their lame Canoga-Perkins solution and offer real e-line and e-lan services they would need an access box like the one being discussed here. At that point the game changes due to the incredible volumes that are possible from an incumbant.

If a company like White Rock lands a large RBOC CLE deal, liquidity issues will be substantially mitigated. The other side of this sword is that the price point will be draconian and as Ringed? rightly points out margins will be under pressure. I would expect(temporary?) margin salvation may come in one of the emerging VLSI solutions - Arrive, Parama, PMC - that will crank the cost down a bit.
papychon 12/5/2012 | 3:28:29 AM
re: Micro MSPPs Are Big There are several more MicroMSPPs on the market that were not mentionned in this article: UTStarcom's NetRing600, TelcoSystems OTM1000, Lucent's (DMXplore, DMXpress, Metropolis-AM), T-PACK (Denmark), Anda Networks EtherReach 3000, RBN's GigaEdge, AXXESSIT (Norway), etc, the list goes on...

In my opinion, this market is already getting pretty crowded...

The barrier to entry to this market has become very low, anybody and his monkey can now stick a couple PMC chips together...
fgoldstein 12/5/2012 | 3:28:29 AM
re: Micro MSPPs Are Big >I'm not sure what chip you are talking about here. I would think that TranSwitch Ethermap-3 would be ideal in this sort of application. This is a $125 chip. Any thoughts here?

HR put together a report which named a whole heap of chips. They cover a wide price range; I was picking $300 as sort of a midrange number. The costlier chips tend to support higher speeds, and may have more features.

And yes the Adtran Opti6100 does look good -- it was in fact my recommendation for a space-constrained application. It tops off at OC-12, and assumes that you have another vendor's SONET backbone -- Adtran is focused on access, so this MSPP doesn't get very far on its own. I seem to recall a missing feature or two, but it does more than most of the (cheaper) micros.
OSXman 12/5/2012 | 3:28:33 AM
re: Micro MSPPs Are Big ">Think about it. How high can the margin get on products like this? Anyone know?"

I don't know, but I'm pretty sure the margins will be lower. On the other hand, there are a lot of customer premises and the unit volume can be a lot higher. Sounds like a classic tradeoff to me.

"The gist is that the bulk of the work of a micro MSPP is done on one chip, which costs about $300."

I'm not sure what chip you are talking about here. I would think that TranSwitch Ethermap-3 would be ideal in this sort of application. This is a $125 chip. Any thoughts here?
spsol 12/5/2012 | 3:28:34 AM
re: Micro MSPPs Are Big It is unforgetant they skipped the OPti6100. It is very full featured and most important has superior warranty. The weakness is strickly SONET, no GBE plans yet. The market at Entrerprise level seems to be going from ATM/Sonet to GBE.
jmunn 12/5/2012 | 3:28:35 AM
re: Micro MSPPs Are Big They missed ADTRAN's OPTI-6100 MSPP. It has much more flexibility than some of these pizza boxes.
douggreen 12/5/2012 | 3:28:35 AM
re: Micro MSPPs Are Big Tellabs originally OEMed White Rock to front end their optical cross connect (for lower speed interfaces), not to sell as a stand alone product. THis may have changed, but if it did not, there is no pull-though for the White Rocks products. Perhaps someone familiar with the current relationship can comment.

I have always liked the product. The company is very focused on doing a few things well. If it had been around in the good old CLEC days (MFS and TCG), it would have made a real killing.

By the way, since I now live in the RTP area, I have to point out a local company that has a pretty nice micro-MSPP offering: Overture Networks. It has SONET or DS-3 uplinks with Ethernet and T1 downlinks.

(P.S. I have no consulting relationship with either White Rock or Overture).

Ringed? 12/5/2012 | 3:28:36 AM
re: Micro MSPPs Are Big Turin announced a Micro MSPP and had one at their booth at Supercom. Don't know if it worked but the lights were on. :)


White Rock is an odd company. The sell their VLX product line from a price point as though it were a Micro MSPP. Why?

I think Viewpoint and MrDid have brought up key issues. White Rock will need to IPO to make it. No large company will pay for them when they can wait until July and buy it at liquidation. Therefore going back to the VC's is going to be difficult.

Don't look for a Micro MSPP to pull any company out of troubled waters. At best you'd sell these products through a VAR at volume rates. But at these price points you would likely need at minumum 40% margins to sell through a VAR. If a VAR would take 20% then there is only 20 left. Is a 20% margin enough for an equipment vendor to positively impact the balance sheet? Are components that cheap now?

Think about it. How high can the margin get on products like this? Anyone know?

fgoldstein 12/5/2012 | 3:28:36 AM
re: Micro MSPPs Are Big >Think about it. How high can the margin get on products like this? Anyone know?

Hard to tell what the margin is, especially considering development costs. But I did attent LR's SONET seminar in November, and got HR's report on SONET chips. The gist is that the bulk of the work of a micro MSPP is done on one chip, which costs about $300. So you add some glue logic, SFPs, packaging, power, and the basic box still probably costs less than a grand to produce. So if one's to believe Scott Raynovich, there will be a raft of these things selling in the $1-5k range pretty soon.

I'm actually disappointed by what's in the market now. I'm running into applications for a small but not micro MSPP -- say, 3RU with a few plug-in cards, to squeeze into a telco collocation rack, for instance. Many of the small products are too bounded. White Rock, for instance, has a great deal if you need that particular combination of ports, but if you need a bunch of DS3s, or a bunch of DS1s, or a particular combination of Ethernet and DS-whatever, then you could be out of luck. Especially at OC-12 or up. Even Cisco's 13-series products seem bounded. Huawei of all companies seems to be the one closest to the target. I'm open to suggestions.
Viewpoint 12/5/2012 | 3:28:37 AM
re: Micro MSPPs Are Big The American MSPP vendors don't have cost structure that BSNL will demand. BSNL is not going to buy in quantities that will make American MSPPs profitable. More likely beneficiaries of BSNL build out would be Indian vendors like Tejas, low cost Chinese vendors and the big telecom vendors like Siemens, Cisco, Alcatel.
Viewpoint 12/5/2012 | 3:28:37 AM
re: Micro MSPPs Are Big
Calling modular design as building block strategy is nice marketing spin. These companies have been too long in a depressed market. It is time to put them to sleep. How many micro-MSPPs and MSPPs are ILECs going to buy from little boys? Only choice is that either they get acquired for paltry sum of money invested in them or gracefully shut down. It has been perfect storm for them.
optical 12/5/2012 | 3:28:37 AM
re: Micro MSPPs Are Big With White Rock's building block strategy, they can just build a new "building block" to keep up with the technology race. the key is, do they have enough runway to pull this off. Their concept is really smart if they have a revenue stream that supports this model. My guess is due to the depressed market, the answer is not positive, but you have to give these guys credit for surviving, which is a feather in their mangement's cap. Lonnie Martin and company have done a good job keeping this going.
Liberty Valence 12/5/2012 | 3:28:40 AM
re: Micro MSPPs Are Big Do you mean that the Indian PTT, which needs to deploys massive E1 connections for public switches is really interested in those boxes ? hum...
bb55 12/5/2012 | 3:28:41 AM
re: Micro MSPPs Are Big Although those guys get some significant funding few years back, they are burning through those fast. I don't believe that their sales in last 2 years support operational costs with any significancy.
The aggrement with Tellabs could be just a cooperation aggrement (ie. Movaz with Lucent and Motorola). Tellabs simply uses WR to get into customers' accounts and win more DSC sales. If the box was so wonderful, WR would be picked up by now.
Remember that the box is based on 4-5 years old technology. All the new uMSPP contenders will have much more advanced, flexible and economical boxes.
Viewpoint 12/5/2012 | 3:28:43 AM
re: Micro MSPPs Are Big Turin doesn't have anything resembling Micro-MSPPs. If one is in works we won't know.

>Did you ever think that they might not want to be

That depends on how long VCs are willing to support them. Are they operationally profitable ?

At cost point of $3000 to $5000 you have to sell a hell lot of these to make up your revenues. You are not selling $100,000 OC-192 boxes. It is a different ballgame.

>They have folks signed up for product but are
>unfortunately not in a position to deliver as of
>yet..."hope springs eternal"

May I know why are they not in position to deliver ?
mrdid13 12/5/2012 | 3:28:43 AM
re: Micro MSPPs Are Big Lonnie and his team are indeed in a good position with good products. Paul is doing a bangup job in marketing but I wonder about their sales crew...they've lost some business in key accounts to Turin Networks. Yes....mrdid13 has been here too and I do indeed have personal connections.

They need to pin down more business in the independent telco space as well as possibly the larger ILECs...too bad they lost Consolidated.

Did you ever think that they might not want to be acquired? Phoenix-Lite is another optical startup that is winging it without the thought of getting acquired. Some companies want to get their own bang out of running things for as long as they can. They have folks signed up for product but are unfortunately not in a position to deliver as of yet..."hope springs eternal"
Viewpoint 12/5/2012 | 3:28:44 AM
re: Micro MSPPs Are Big Yes White Rock has a good product lineup. A potential acquirer likes to see a big account and market traction demonstrated by revenues. Not lab trials. Has Tellabs being able to move Whiterock ?

Blame it on market conditions and other factors. Big boys are not spending. Not enough to give profitability to bubble era start ups like White Rocks and others. The technology is few years old. Meanwhile you can buy better chipsets from PMC or AMCC.
optical 12/5/2012 | 3:28:45 AM
re: Micro MSPPs Are Big White Rock has a trememdous product offering and I'm just shocked that no one has acquired them to date, or maybe they have, but the price just wasn't high enough due to depressed market. Maybe this is their year. Great product in my opinion.
Viewpoint 12/5/2012 | 3:28:45 AM
re: Micro MSPPs Are Big
It had a mirco-MSPP two years back. I remember testing it. Had more features than we require but it was one of the first with a shipping micro-MSPP. Very over-engineered box with no support for native SONET/SDH uplinks. It had very good features on the user side interfaces. Better than Atrica which also has no support for SONET/SDH uplinks. Atrica had very good modularity. Closer to prototype version than deployment product. The native SONET/SDH OEMs lacked user interface features that Atrica and Luminous offered.

I am disappointed with this article. I expected it to have more depth on what micro-MSPP really means. micro-MSPP must support following:

OC3 and OC12 uplinks in native SONET. Face it guys most of our equipment in the ground can't be plugged in with Ethernet or RPR interfaces.

User side interfaces should support edge functionality. This means rate limiting, VLAN translation, modularity on interface mix, user level provisioning, some ability to do IP lookup.
Viewpoint 12/5/2012 | 3:28:46 AM
re: Micro MSPPs Are Big Since when did a security company start manufacturing and selling MSSPs ? I didn't know they were so hot :)

You meant Tejas Networks, a Banglore based company and not Tejas securities.
Toad680 12/5/2012 | 3:28:47 AM
re: Micro MSPPs Are Big Anyone heard how they are doing? Last I heard they had a reseller deal with Alcatel for their EoS box, but that it was coming along slowly. Rumor has it initial production shipments have just begun.

Anyone else hearing anything?