x
Optical/IP

Cisco Stumps for CRS-1

In this election season, Cisco Systems Inc. (Nasdaq: CSCO) appears to have hit the campaign trail, possibly trying to deflect claims that the CRS-1 core router is off to a slow start. Cisco is pressing its argument that the CRS-1 is shipping, has garnered revenue, and is "fully deployed" in carrier networks, as one analyst puts it.

Cisco is saying it has shipped two CRS-1 boxes for revenue, according to reports issued yesterday by analysts Tal Liani of Merrill Lynch & Co. Inc. (the "fully deployed" analyst) and Mark Sue of RBC Capital Markets. Cisco also told the analysts that seven customers have completed CRS-1 testing and seven more have tests underway.

Sprint Corp. (NYSE: FON) and Deutsche Telekom AG (NYSE: DT) have publicly said they're testing the CRS-1. Sprint even announced it was using the box, although some analysts believe it's not carrying any customer traffic (see Sprint Throttles Up Cisco's CRS-1).

But that's different from the question of who the two paying customers are. Sue pegs Sprint and America Online as the buyers. Another source requesting anonymity spins a different tale, saying Comcast Corp. (Nasdaq: CMCSA, CMCSK) is definitely one of the two buyers, but is not putting the box into a live network yet.

The source has also heard AT&T Corp. (NYSE: T) is a possibility, but adds that that seems unlikely due to AT&T's recent contract award to Avici Systems Inc. (Nasdaq: AVCI; Frankfurt: BVC7) for core routing (see Avici Expands AT&T's IP Backbone).

Cisco, by the way, declines comment on all of this; Officials wouldn't even confirm the details published by the analysts, nor would they comment on the status of the Sprint installation.

The CRS-1, formerly known as the HFR, is Cisco's multichassis core router, theoretically capable of linking 18 chassis into one logical router with 46 Tbit/s of capacity. The product was launched to great fanfare in May, with a $450,000 price tag (for one bare-bones chassis) and promises of shipments by July (see Cisco Unveils the HFR).

The product gave Cisco a new core offering, supplanting the ageing Gigabit Switch Router 12000 series. It also gave Cisco an answer to high-end competitors such as Avici's Terabit Switch Router (TSR). Perhaps most importantly, the CRS-1 debuted Cisco's new modular operating system, called IOS XR.

Analysts and competitors have since sniped at the CRS-1, saying the product wasn't ready to ship by deadline (Cisco insists it was, and that the first shipment came in August). The talk is likely to continue -- sources still say Cisco's pending acquisition of Procket Networks Inc. is aimed at gathering the talent to patch up flaws in the CRS-1 technology (see Valley Wonk: The Procket Puzzle and Procket Reaches 'End of Life').

If the CRS-1 is truly making its way into live networks, it could put some pressure on Juniper Networks Inc. (Nasdaq: JNPR). Juniper's got a multichassis product in the wings, the TX Matrix intended to combine the company's T640 core routers in a multichassis configuration. The TX wasn't expected to be released until this year, and there's still no sign of it -- possibly because demand is so low for such large routers (see Juniper Goes Terabit With the T640 and Juniper's TX Waits Its Turn).

So do the CRS-1 shipments give Cisco a killer edge against Juniper? Hardly. Liani notes that a technological lead by either company "will close over time," creating a neck-and-neck race for quite a while. "In the core routing [market], we believe that the comparison between Cisco and Juniper is almost irrelevant. We believe both players now have good offerings and we think that the carriers will continue to dual-source these vendors, in line with their traditional behavior," Liani writes.

It's also still believed that no customer is linking the CRS-1 in a multichassis configuration. In fact, much of the buzz around the system is around a half-sized version that would appeal to a larger percentage of network owners. RBC analyst Sue has taken to calling this box the "CRS-2," and his report yesterday noted that the product probably won't be announced until the end of 2004 or early 2005 (see Cisco Sprints Ahead With HFR).

— Craig Matsumoto, Senior Editor, Light Reading




For more on this topic, check out:


For further education, visit the archives of related Light Reading Webinars:

Page 1 / 2   >   >>
elijed23 1/27/2014 | 9:50:26 PM
Sharks in the Water Though it does seem to carry more bandwith, the concerns are not uncalled for. I do believe the bigger companies will try to take advantage of this change however, and will swarm the waters until every drop is taken. 
elijed23 1/27/2014 | 9:50:26 PM
Sharks in the Water Though it does seem to carry more bandwith, the concerns are not uncalled for. I do believe the bigger companies will try to take advantage of this change however, and will swarm the waters until every drop is taken. 
optical_man 12/5/2012 | 1:10:49 AM
re: Cisco Stumps for CRS-1 Sorry Cisco. I have always respected you and your mission. (technology aside...that's a different story).

Is it me, or is this the beginning of Cisco's slide into becoming Alcatel, Nortel, Lucent, Ericsson?

They are now having to defend the same stories that the Old Guard has to every other month.

"Older, Mature, Company with a good product years ago, cannot seem to innovate"

Which company does that describe? One or all of them.
Welcome to the club Cisco. It's not a bad club, it's a profitable one, but the WOW is now gone.
It's time for all the fun stuff, like layoff's, politics, lack of direction, and retiree pension reductions, or at least the Stock Option version of same.

Sign of the times, methinks.
jepovic 12/5/2012 | 1:10:48 AM
re: Cisco Stumps for CRS-1 The CRS-1 is a helluvalot more innovative than the T640. Its software architecture is exactly where routers need to go, when the operators migrate all their traffic to IP/MPLS. Juniper has nice stuff, but I doubt that monolithic software will do in the long term.

Right now, though, the GSR carries enough traffic, so it's tough for carriers to build a business case for a brand new core. Also, fewer and fewer carriers enjoy being the vendors' guinea pigs, and prefer to wait until the software stabilizes.
dpb 12/5/2012 | 1:10:45 AM
re: Cisco Stumps for CRS-1 A little quick searching will give you the answer to the deployment of CRS-1 and if it is passing customer traffic.

Go here:
http://www.prnewswire.com/cgi-...

Then here:
http://proj.sunet.se/LSR3/

-David Bannister
myoptic 12/5/2012 | 1:10:44 AM
re: Cisco Stumps for CRS-1 Cisco is nearing the "cut bait" decision point for the CRS-1. After 5 years and over $500 million in development costs and extensive arm twisting of its customer base, Cisco has yet to register a single legitimate customer win with the CRS-1.

Fact is that Cisco got blinded by ego and built a box which is out of touch with carrier requirements. The CRS-1 is waay too big, doesn't have the interfaces to replace the GSR, and has fundamental ASIC performance and sw stability issues.

Cisco is already hedging their bets with the 12816, and is using Procket resources to develop the CRS-2. The rumblings are that the CRS-2 will be built on a new hardware architecture and the CRS-1 may go down in history as a very expensive science project.
AAL5 12/5/2012 | 1:10:44 AM
re: Cisco Stumps for CRS-1 myoptic said
"Fact is that Cisco got blinded by ego and built a box which is out of touch with carrier requirements."

- Many aspcts of the CRS-1 were built as a result of direct customer feedback, this is the type of box they want in their networks.

"The CRS-1 is waay too big"

- '640Kb is enough for anyone'?

"doesn't have the interfaces to replace the GSR"

- What core router can you name had POS OC-3 to OC-768, GigE, ATM, channelized interfaces all supported in its initial release?

"has fundamental ASIC performance and sw stability issues."

- What Asic performance and s/w stability issues exactly?


"Cisco is already hedging their bets with the 12816, and is using Procket resources to develop the CRS-2."

- LOL, you are really grasping at straws. If you had any clue you would know that Procket resources are *not* being used for this purpose. I don't know why people like you think they can type this B.S. as there are many other people reading this that know you are just making this up.


"The rumblings are that the CRS-2 will be built on a new hardware architecture and the CRS-1 may go down in history as a very expensive science project."

- Ehh, no, wrong again.


AAL5
chipsales 12/5/2012 | 1:10:42 AM
re: Cisco Stumps for CRS-1 Good lord-it sounds like the republican versus the democrats when reading the posts of you router wanks.

Give it a rest. Both companies (Juniper/Cisco)are solid. Both have good products.

Do you know most people read your "slanted" posts and simply chalk it up to internal marketing from one or the other.

-chipsales.

tsat 12/5/2012 | 1:10:42 AM
re: Cisco Stumps for CRS-1 "Juniper has nice stuff, but I doubt that monolithic software will do in the long term."

This is very funny coming from a Cisco cheerleader...

-tsat
opticalwatcher 12/5/2012 | 1:10:42 AM
re: Cisco Stumps for CRS-1 "Cisco is already hedging their bets with the 12816, and is using Procket resources to develop the CRS-2."
"LOL, you are really grasping at straws. If you had any clue you would know that Procket resources are *not* being used for this purpose. "

Too bad. Despite all of the talk of Procket's massive overspending, they accomplished a lot more (in terms of customer traction) than Cisco for a LOT less money. Cisco would be nuts not take advantage of what Procket accomplished. But you know how it can be at big companies--it wasn't invented here so politically it is probably looked at like dirt. Companies often care more about their egos than what is best for the company.

And no, I never worked at Procket.
Page 1 / 2   >   >>
HOME
Sign In
SEARCH
CLOSE
MORE
CLOSE