[140005] in North American Network Operators' Group
Re: riverbed steelhead
daemon@ATHENA.MIT.EDU (Elijah Savage)
Wed Apr 27 14:22:50 2011
Date: Wed, 27 Apr 2011 14:21:14 -0400 (EDT)
From: Elijah Savage <esavage@digitalrage.org>
To: harbor235 <harbor235@gmail.com>
In-Reply-To: <2617293.29601303928361201.JavaMail.root@ubuntu.digitalrage.org>
Cc: NANOG list <nanog@nanog.org>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
>Anyone out there have experience with Riverbed Steelhead products?
>Do they improve TCP performance over WAN links? is it worth the price?
>mike
-----------------------
James and Eric have done outstanding contributing here. I just wanted to ad=
d a tad bit of information leaving out the name brands. Our environment has=
a large number of these deployed and we have seen business and end user ac=
ceptance. From a business perspective, it could be a cost deferral tool wit=
hin your toolset. Meaning if you are in an environment, which is sensitive =
to MRC (monthly recurring cost), implementing this technology can help you =
defer this cost. For example, I had a remote site that was at 95% of a full=
ds3 in capacity 6 hours of an 8-hour business day, after implementing this=
technology I was amazed at what we found. This sites utilization was reduc=
ed to 40% capacity max, not to mention the reduction we observed at the dat=
acenter headend. We were preparing to increase capacity at this site, that =
was going to require swapping the WAN device as well as an increase in the =
MRC for this site and ongoing operational expense. Introduction of WAN opti=
mization was significantly lower in these categories versus the upgrades.
On the end user side due to the physical location/latency of this site a fe=
w applications was prone to suffer, the developers agreed to fix this would=
possibly require rewriting of the apps. Nothing more than the introduction=
of this technology was done and the support calls into the support center =
went from 100's per month to zero for a few of these applications.
One thing that was important to me WAS MAPI encryption and the fact it was =
reversed engineered. Meaning if Microsoft at some point changed the way MAP=
I worked in a patch or upgrade you would end up with broken MAPI support an=
d possibly a long runway to fixing it. The other company from my understand=
ing (do not quote me on this YET) is collaborating and possibly working wit=
h Microsoft to license such support.
I don't think you could go wrong with either brand/company but understandin=
g your environments traffic patterns and its application usage would go a l=
ong way in guiding you towards the product for your environment.
Good luck
