FUD the WAAS way

Well, I'm back in San Francisco after a successful Interop show last week in Las Vegas. I say "successful" even though Cisco WAAS was named "Best of Interop" instead of us.

But, it's hard to know if they really were Best of Interop or not - they didn't have the new WAAS product in their booth (as required by the contest rules), only a video about it and six PowerPoint slides.

Maybe they should have won a different award:
"Best PowerPoint presentation about a product that might be demonstrated someday at Interop".

Anyway, I digress. What I really wanted to write about was those same folks at Cisco continuing to insist that "Riverbed will break VOIP" or "Riverbed will screw up your security" if you are so foolish as to deploy it. Please...

It happened again during my WAN optimization panel on Wednesday. There's kind of an existence proof that what they claim can't be true: our 4,000 customers. Furthermore, our overall customer satisfaction is 8.75 out of 10 (and 87% of the roughly 1,000 respondents gave us an 8, 9 or 10 as an overall score).

If we were really breaking our customers' voice or security implementations, I don't think we'd have 4,000 customers, nor would they be giving us such high ratings.

The fact is that, just like our slogan, Cisco customers love Riverbed. Actually, Riverbed is a Cisco customer too! We love Cisco... for routing and switching.

The other angle we've been hearing from these guys lately is that the products (all the competitors' products) are about the same now technically, that they've caught up (and therefore you should just go with the vendor you know and love... Cisco).

The example given in the same panel was that the CIFS optimizations in our respective products are essentially the same now, so customers shouldn't differentiate on that dimension any longer (and by extension, that the other technical features are about the same now too).

Actually, that's not true either in the specific instance of CIFS, or more generally. Cisco still has the file cache left over in their WAAS product from their acquisition of Actona back in 2004 and their optimization of Windows file sharing is still largely based on caching copies of files locally, combined with byte-level deduplication.

Caching is just completely different from the way Riverbed optimizes bandwidth and optimizes for high latency. The reality is that the products in the WDS market have many differences, some subtle and some not so subtle, and you really have to test them to figure out which product performs the best, and which ones are easiest to deploy and manage at scale. Glossing over important differences will only lead to disappointment. 

Comments

[Mark]

I had to add passthru rules to my Steelheads as they were interfering with the call control on our Cisco CallManager implementation...