Source: Cisco's HFR Tips the Scales

COMMENTS

Newest First | Oldest First | Threaded View

<< < Page 4 / 13 > >>

edgecore 12/4/2012 | 11:17:33 PM

re: Source: Cisco's HFR Tips the Scales What I see is this: IBM is doing quite well and Microsoft is creating inferior products. So what is the lession learned?

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

IBM Market Cap 150B

MSFT Market Cap 300B

EC

Reply | Post Message | MESSAGES LIST | START A BOARD

mboeing 12/4/2012 | 11:17:33 PM

re: Source: Cisco's HFR Tips the Scales "And the dual OS approach will never be seen as a plus, as no matter how much code is shared, even a minor
difference will require qualification prior to deployment in any tier 1 network."

If you have a closer look a SP networks based on Cisco routers then you will find different code version on different classes of routers. For example edge boxes will run other code then core boxes. Cisco 7500 routers will run other code then c7600, ESR 10k or GSR 12k boxes.

Service providers typically have very rigid process of lab testing and limited deployment prior to delpoying any new code/box/feature.

My conclusion is that having another box with another OS does not really matter. Service providers have the process to deal with that. Other factors, such as capex/opex/features/stability/performance, are much more important.

Reply | Post Message | MESSAGES LIST | START A BOARD

signmeup 12/4/2012 | 11:17:32 PM

re: Source: Cisco's HFR Tips the Scales Based on the amount of csco FUD included in the article, I thought I would shed some light on the subject.

"As mentioned, Cisco has developed an entirely new operating system for the HFR."

Truth: This is cisco's vaunted ENA code that has been in development for years now - it started on the 7500 platform! I know first hand that they have gone through at least 4 complete purges of the development team, from the coders all the way to the managers. It has been close to scrapped at least 2 times for not being able to meet expectations. It runs approximately 30% SLOWER than IOS does. There is no WAY there can be any consistancy with this code as every 12 months or so everyone on the project either quits or is canned. Without consistancy you get bugs, regardless of whether it is a modular design or not. So you make the call: inexperienced sw coders + brand new hardware + inconsistent development = ??????

"For scaleability, an additional distributed route processor can be installed in any line-card slot. This processor is similar to Cisco's Distributed Cisco Express Forwarding (dCEF), deployed on its lower-end routing platforms"

Truth: So let's see, what version of the PFC is cisco at now? PFC3? So basically what you are saying is that anytime you need a new feature, you'll have to upgrade to the next version of "distributed route processor"? The truth is that cisco lags far behind in development of cutting-edge silicon, and the result is that the customer ends up paying. Sounds like good investment protection to me - for cisco that is...

" One chassis is 1,350 lbs. -- about as much as a very large moose -- and gobbles 12 kilowatts of power."

Truth: How do you even look at this in a positive fashion? Especially given the fact that the competition can provide the same density with 1/2 the space and 1/4 of power requirements!! This was for ONE CHASSIS!

Why do I care about the truth of project "Q"? Because it is laughable how cisco positions the HFR as a next-generation router. There is nothing next-generation about the HFR - Procket and Juniper already have products far superior to it, and are shipping them today. So by the time the HFR actually does hit the street, it won't be next-generation, it will be the same old story: a day (or year) too late, and way too many $$$$.

Reply | Post Message | MESSAGES LIST | START A BOARD

jstuart_99 12/4/2012 | 11:17:32 PM

re: Source: Cisco's HFR Tips the Scales mboeing wrote:
"What I see is this: IBM is doing quite well and Microsoft is creating inferior products. So what is the lession learned?"

Well, in addition to Microsoft's market cap, they forced IBM to completely re-invent itself as a services organization first and a sw/hw company second. Kudos to IBM for being able to pull it off, but the point was that every generation brings out new competition in the market. Cisco has already demonstrated a lack of ability to produce quality products for the service provider market. The market dictates that someone fill that need, and you are starting to see that happen.

Reply | Post Message | MESSAGES LIST | START A BOARD

AAL6 12/4/2012 | 11:17:32 PM

re: Source: Cisco's HFR Tips the Scales Only parts of HFR have been done in Ottawa - too few people to do enough for such a big project (BTW, did you see Pentagon Wars movie? - time and money spent there for BFV approaches this one ;). Rumors are that they fired hardware guys couple of weeks ago which could indicate that they are not needed any more?
IOS-NG was based on QNX Neutrino kernel - and IOS ported on top of it.
The HFR development sucked a lot of money and effort - it was supposed to be released in 2001/2002 when the demand for god boxes was looking good (from 1999 perspective).

Reply | Post Message | MESSAGES LIST | START A BOARD

RouterOttawa 12/4/2012 | 11:17:31 PM

re: Source: Cisco's HFR Tips the Scales Only parts of HFR have been done in Ottawa - too few people to do enough for such a big project Rumors are that they fired hardware guys couple of weeks agowhich could indicate that they are not needed any more?

Regrettably, the rumours are fact. They let 20-30 people go.

Reply | Post Message | MESSAGES LIST | START A BOARD

glad2Bgone 12/4/2012 | 11:17:31 PM

re: Source: Cisco's HFR Tips the Scales BobbyMax

All your base are belong to us

Reply | Post Message | MESSAGES LIST | START A BOARD

glad2Bgone 12/4/2012 | 11:17:31 PM

re: Source: Cisco's HFR Tips the Scales BobbyMax

Reply | Post Message | MESSAGES LIST | START A BOARD

Sisyphus 12/4/2012 | 11:17:30 PM

re: Source: Cisco's HFR Tips the Scales
Interesting discussion in general.

No one can blame Cisco for not trying hard to turn routers into the next generation class 5 switch equivalent, though, which does represent a phenomenal challenge and takes major guts. The Avici and Juniper products don't take the vision of the router as a core infrastructure product quite as far as the HFR does.

Nor can Cisco be blamed for going back to the drawing board a few times given the ambitious nature of the project. It is now certainly a more universally usable product than it was initially geared up to be, which is good.

Finally, it is a great thing that the crucial importance of the SW package was recognized, and it's here where the HFR might have the biggest impact.

One thing that's striking in the discussion is how the criticism stems from both accusing the HFR of both being too disruptive and not disruptive enough, which is a sure indication it is as of now a pretty misunderstood product.

Reply | Post Message | MESSAGES LIST | START A BOARD

desikar 12/4/2012 | 11:17:30 PM

re: Source: Cisco's HFR Tips the Scales Tony,

I have lots of respect for your technical views. However, have to disagree with your note on router capacity... Wouldn't the growing traffic need also be served by having more of the proven 1TB routers (instead of new 2 TB routers) and/or better management of traffic and possibly also by lower customer expectations on QoS (in return for lower cost per bit)?
At some point, of course it could be more cost-effective (taking into account technical and operational risks, etc.) to switch to something better - say, when it is just better to have 8x the capacity for example. But till then, falling prices, cost of training and debugging operations, etc. could postpone the drought for new technology, even in greenfield deployments.

IMO, what Cisco would be doing with the HFR if
it really does exist, is to assure customers that they can engage Cisco at any time in the future to tap into advanced technology routers - thus reducing the risk of being stranded if they go
with Cisco. So the HFR needs to keep up with the
latest technologies available to the competition, but does not necessarily need to be deployed to be of benefit to both Cisco and the service provider community, and to be an irritant to competitors bringing out intermediate technology steps.

Looking forward to your thoughts on the above..

-desikar

---------Tony Li wrote---------------------
According to the best available numbers (http://www.dtc.umn.edu/~odlyzk... the Internet is continuing to grow exponentially at 100%/year. If the market needs a terabit machine today, then it will need 2 terabits next year, four the year after that, then 8, then 16, 32, 64, etc.

Tony

Reply | Post Message | MESSAGES LIST | START A BOARD

<< < Page 4 / 13 > >>