Chiaro Debuts a Big, Bad Router

COMMENTS

Newest First | Oldest First | Threaded View

<< < Page 4 / 4

signmeup 12/5/2012 | 12:01:14 AM

re: Chiaro Debuts a Big, Bad Router You are correct, but that does not change the fact that Juniper is most vulnerable to a startup that differentiates itself on superior technology. Everyone knows that cisco is behind the curve when it comes to next-generation router technology - people don't buy cisco for that reason. They buy cisco because of its size and its support (or perhaps to have an end-to-end cisco network). Cisco will be around for awhile, given its billions of dollars in cash.

Juniper has always touted its technical superiority over cisco. Again, my question is "what happens when a couple of startups do one better?"

As far as 3rd party testing is concerned, I would love to see someone test the Procket 8812 with 48 OC-192's to see what it can do. Anyone from Procket care to comment on this?

signmeup

Reply | Post Message | MESSAGES LIST | START A BOARD

signmeup 12/5/2012 | 12:01:14 AM

re: Chiaro Debuts a Big, Bad Router >>http://www.aloha.net/~mahalo/c...

Now there's some technology I would LOVE to get my hands on !!

Sorry, couldn't resist...

signmeup

Reply | Post Message | MESSAGES LIST | START A BOARD

skeptic 12/5/2012 | 12:01:13 AM

re: Chiaro Debuts a Big, Bad Router Several vendors have tried this clustering stuff in vain. Avici talked about this but have they deployed any routers in such a configuration?

Has Juniper?
--------------
There is a rumor that they did a lab demo, but
I've never heard anyone whose seen one in the
field.

Reply | Post Message | MESSAGES LIST | START A BOARD

joeshmo 12/5/2012 | 12:01:13 AM

re: Chiaro Debuts a Big, Bad Router I would argue that Juniper is less vlunerable than you think. The technology works. Not in just in a lab under the best possible conditions but in the live networks you cannot beat them. The edge gear is phenominal and continually improving. They are out there winning most of the "new" business at the core and especially the edge even with the small amount of spending the market has seen. The large carriers seem to be aligning themselves with Juniper in a way that will put them in a great spot when (not if) the spending resumes. Sure Cisco is still selling to their customers but what new projects have they won lately. The most recent example is this Verizon deal where everyone thought it would be Cisco or a startup. We're talking hundreds of millions in the next two years and that's just one customer.

Sadly, by the time startups like Procket prove they have a better technology, Juniper will have a foothold and their next gen equipment will be as good or better than the startup.

Reply | Post Message | MESSAGES LIST | START A BOARD

skeptic 12/5/2012 | 12:01:12 AM

re: Chiaro Debuts a Big, Bad Router I don't see any more information regarding clustering on the web site; perhaps still in development?
----------------

Even in the more detailed presentations, they
don't talk about it much. I don't remember
seeing it even roadmapped.

They had bigger plans for scalability at one
time, but what I heard was that they abandoned
most of them.

I think its also going to be harder because
their switch is supposedly shared memory
based.

The problem they have now is that they have a
line of three different routers with a confusing
upgrade path between them.

Procket has so many features that need to be done
(with less staff now), that I can't see them
spending their resources on clusters.

The big task for procket is to figure out what
they want to build. They have to pick a direction
(and a set of features) and move toward finishing
off "a" product. They can either go toward
the edge or the core, but I dont think they
can afford to do both anymore.

Even after the layoffs, their burn rate is
way too high. And the overall product is
still months from really being done in all
features based on

Reply | Post Message | MESSAGES LIST | START A BOARD

rjmcmahon 12/5/2012 | 12:01:11 AM

re: Chiaro Debuts a Big, Bad Router The big task for procket is to figure out what they want to build.

Maybe knowing the target will help our tech industry figure out the tools and weapons needed? I'll suggest an initial target (tech's version of Iraq) as Louisiana -- where Bell South needs to be replaced by a regime that embraces structural separation.

Reply | Post Message | MESSAGES LIST | START A BOARD

Belzebutt 12/5/2012 | 12:01:10 AM

re: Chiaro Debuts a Big, Bad Router Procket has so many features that need to be done
(with less staff now), that I can't see them
spending their resources on clusters.

That's the hard part for those startups. They're a man-century of work behind Juniper, let alone Cisco when it comes to features, and they have to catch up using their hundred or so engineers whose ranks keep shrinking.

Reply | Post Message | MESSAGES LIST | START A BOARD

nbwaite 12/5/2012 | 12:01:08 AM

re: Chiaro Debuts a Big, Bad Router So, your article has:

_____"capital spending slump"
_____"It seems a bit crazy"
_____"an unfortunate time"
_____"significant economic hurdles to cross"

for an IP routing company.

Okay, adjectives aside aside, suppose we look at
some audited reasonably objective numerical data
about external reality?

One 'barometer' of "capital spending" on IP routing
is the revenue of Cisco. So, suppose we take a look
at that.

In the Cisco 10-K report of 9/18/2002 and from the
SEC Edgar Web site, we can see

_____________As_of_Date_______Revenue_(Millions)
__________-------------_______------------------
__________July_27,_2002__________________$18,915
__________July_28,_2001__________________$22,293
__________July_29,_2000__________________$18,928
__________July_31,_1999__________________$12,173
__________July_25,_1998___________________$8,489

Here Cisco's revenue looks nearly as high as it has
ever been and with excellent growth over the years
shown (yes, there were some acquisitions).

Of course, this is data from 2002, and maybe Cisco
has gone out of business since then? Well, more
recently at Edgar we can see for Cisco:

_____NET SALES FOR THE FIRST SIX MONTHS of fiscal
_____2003 were $9.6 billion, compared with $9.3
_____billion for the first six months of fiscal
_____2002, an increase of 3.2%.

Look, guys, this audited numerical evidence is that
"capital spending" on IP is about as high as it has
ever been and not in a "slump".

Yes, it's easier to have just two 'stories': (1) up,
up, up, forever up to infinity and beyond or (2)
down, down, down, melt-down.

Your story here is a case of (2) with "slump that
shows few signs of roaring back" for nostalgia for
(1).

Reply | Post Message | MESSAGES LIST | START A BOARD

hayward 12/5/2012 | 12:00:42 AM

re: Chiaro Debuts a Big, Bad Router Actually they did not win the deal. Wavesmith is in the process of completing the contract. We found several problems with the ERX. The system still melts chips on the T3 cards when the fan tray is pulled along with an inherant problem with the ICC. If over run with traffic the ICC causes the system to crash. Sorry for the bad news especially if you work for Juniper. Better luck next time.

Reply | Post Message | MESSAGES LIST | START A BOARD

andropat 12/5/2012 | 12:00:42 AM

re: Chiaro Debuts a Big, Bad Router signmeup,

It's funny to hear comments about performance and stability, etc., from people like yourself who obviously have never worked on these routers or tested them. I am a customer who has tested the T-series and I can tell you that the router does 40-byte packets WITH services at line-rate over Ge/10Ge interfaces given L2 encap overhead. In any case, it is leaps and bounds better than the same tests run against GSR. After multiple linecard swapouts through all the various engine cards that my managers foolishly bought from the now rich cisco team I still didn't get anywhere near the performance of the T-series.

From a s/w perspective its ridiculously tilted as well. Try flooding these routers with lots of routes, do convergence tests, etc., the difference is very very apparent. and at the end of the day simply look at the s/w engineering team....i've met them all on various trips to our account. Credibility is huge!

Lastly, I wish I could share with you the downtime results by vendor in my network. Again, ridiculous when you compare cisco to juniper. Juniper didn't win market share for no reason. What is exciting is that startups have a chance to differentiate given the rise in popularity of needs like IPVPNs, CoS, Security,etc., which weren't huge deals when guys like juniper and cisco were developing their current products. As a core router, I would say forget about competing. As a MSE or core product with serious NEEDED differentiation there is a chance for an upcoming startup.

My .02 cents.

Reply | Post Message | MESSAGES LIST | START A BOARD

<< < Page 4 / 4