Re: [load balancing] Netscaler

From: Tal Klein <Tal.Klein [izzat] citrix.com>
Date: Mon May 07 2007 - 17:23:28 EDT

Hi Leo,

 

I don't like bringing politics into this forum because it's the one
place I can still just be a geek. I come from a hosting background where
my primary areas of LB expertise revolved around NetScaler, Cisco
(CSS's), and F5, so I should tell you I don't have any first hand
experience with Crescendo, but even if I had first-hand experience with
their devices, I wouldn't make any sort of statements about press
releases or claims they've made. However, you do bring up an interesting
article. I doubt Citrix itself has any sort of comment on the NWW test.
I don't really have any comment on it except to point out that there
really is no standardized testing methodology for ADC/AFE's, which is
both a curse and a blessing. See, there's no absolute set of tests that
will tell you which device is the absolute best device on the market
because ADC's (I don't like the term AFE) do different things for
different environments. It's tough to do these sorts of analyses, and I
didn't envy David Newman (the guy who ran the tests) because his task of
creating a "Clear Choice Test" of these devices was utterly impossible.
Allow me to defer to tidbits from the article:

 

"As is often the case with any new market category, there are
substantial differences in terms of form factor, topological
requirements and supported features.

Because of those differences, and because this is a relatively new
product category and no device aced all tests, we're not scoring
products this time around. As our results clearly show, different Web
front-end vendors have put their development dollars in different
places."

 

"It's important to note that we're not claiming either set of tests
represents "real-world" behavior for all users. There are too many
variables in application load testing for a one-size-fits-all definition
of that term to be meaningful."

 

"No device aced every event; rather, the results suggest different
vendors optimized for different aspects of device performance."

 

"Faced with all these differences, the best advice we can offer in
choosing a Web front-end device for your network is to assess which of
these metrics matters most, and go for the box that delivers the biggest
benefit in that area."

(http://www.networkworld.com/reviews/2006/011606-wfe-intro.html)

 

So we basically ran a bunch of devices through a bunch of tests and the
conclusion was that they all do things differently and their benefits
are entirely dependant upon the particular implementation scenarios.

 

What I recommend to anybody who's in the market for one of these is:
First identify the vendors you're interested in based on references in
your industry. Create your own test bed based on staging environments of
your actual networked applications, and then invite your shortlisted
vendors in for a bakeoff of your own. The only way to make an
intelligent decision on which product is right for you is to test it
under your unique set of circumstances.

 

-Tal

 

---
Tal M. Klein
Technical Marketing & Strategy
Citrix Application Networking Group
Direct: 408-790-8536
Mobile: 415-336-3779
E-mail: tal.klein@citrix.com
 
 
________________________________
From: lb-l-bounces@vegan.net [mailto:lb-l-bounces@vegan.net] On Behalf
Of Leo Chan
Sent: Friday, May 04, 2007 8:03 AM
To: Load Balancing Mailing List
Subject: Re: [load balancing] Netscaler
 
Hi Tal,
Do you have any comment on Crescendo Networks ? 
There is the information found in their website
"Crescendo Networks CN-5080E Satisfies "Need for Speed" in Network
World's Landmark Application Front End (AFE) Test" 
http://www.crescendonetworks.com/engsite/html/newsItem.asp?itemID=1&Pid=
47&Sid=9
Leo
On 4/21/07, Tal Klein <Tal.Klein@citrix.com> wrote:
Hi guys,
What a great thread, I feel kindof lame coming it to it so late. First
of all thanks Chuck for stepping up and also Tony for the NIC vs.
throughput discussion which I was going to get into, but having read 
through it I think everyone covered it rather nicely. The NetScaler
12000 currently has several gigs of throughput (I don't have the spec
sheet in front of me but I think its max throughput is like 6 gig) which
can be achieved through link aggregation and in fact many of our
customers use the box that way.
However, the need for 10G ports has been heard loud and clear. We wanted
to be certain that our 10G solution would be solid, but to answer 
Chuck's original question, I'm pretty sure we'll have a 12000 with 10G
ports available this quarter.
If any of you are coming to Interop in Vegas, stop by the Citrix booth
and we can talk shop. 
Cheers,
-Tal
---
Tal M. Klein
Technical Marketing & Strategy
Citrix Application Networking Group
Direct: 408-790-8536
Mobile: 415-336-3779
E-mail: tal.klein@citrix.com
-----Original Message-----
From: lb-l-bounces@vegan.net [mailto:lb-l-bounces@vegan.net] On Behalf 
Of Chuck Adkins
Sent: Friday, April 20, 2007 8:13 AM
To: Load Balancing Mailing List
Subject: Re: [load balancing] Netscaler
Thanks for the feedback on the 10G ...
Anyone using Netscaler that can speak to stability, support, ease of 
administration, availability of metrics for uptime/capacity planning/etc
Regards,
Chuck Adkins
-----Original Message-----
From: lb-l-bounces@vegan.net [mailto:lb-l-bounces@vegan.net] On Behalf
Of Simon Hamilton-Wilkes
Sent: Wednesday, April 18, 2007 11:40 AM
To: Load Balancing Mailing List
Subject: Re: [load balancing] Netscaler 
Correct, it's their biggest box and has 8 gigabit SFP ports on the
front.  It is a nice box though and has handled almost everything we've
tried (very large scale SSL offload was the exception - not all possible
algorithms are hardware accelerated and our client couldn't change which
ones their app used) I agree a couple of 10 gig ports would be nice -
I'm increasingly having the need for 10 gig on projects, big pipes are 
way more efficient than port aggregation, just wish the switch ports
from it would come on down in price, the Woven switch is promising, and
Force 10 have their 24 port 1U box.  On the economy front right now I'm 
using 8 port 1U switches from HP.
Simon
On 4/18/07, Chuck Adkins <chuck.adkins@theice.com> wrote:
> Looking for some community feedback on the Citrix/NetScaler LB 
> solution
> - specifically the 12000 platform - stability, support, admin ease,
> availability of metrics for uptime/capacity planning/etc
>
> Also - as far as I can tell Netscaler has no support for 10G - anyone 
> have any other data or know when 10G support is coming?
> http://www.citrix.com/site/resources/dynamic/salesdocs/CitrixNetScaler
> Ap pDeliveryHardwarePlatforms.pdf looks like the 12000 is biggest box,
> the only option with dual cpu, and it does only G.
>
>
> Regards,
>
> Chuck Adkins
>
> 
>
> --------------------------------------------------------
> This message may contain confidential information and is intended for
specific recipients unless explicitly noted otherwise. If you have 
reason to believe you are not an intended recipient of this message,
please delete it and notify the sender. This message may not represent
the opinion of IntercontinentalExchange, Inc. (ICE), its subsidiaries or
affiliates, and does not constitute a contract or guarantee. Unencrypted
electronic mail is not secure and the recipient of this message is
expected to provide safeguards from viruses and pursue alternate means
of communication where privacy or a binding message is desired.
>
>
> _______________________________________________
> lb-l mailing list
> lb-l@vegan.net
> http://vegan.net/mailman/listinfo/lb-l
> Searchable Archive: http://vegan.net/lb/archive http://lbdigest.com
> Load Balancing Digest
>
_______________________________________________
lb-l mailing list
lb-l@vegan.net
http://vegan.net/mailman/listinfo/lb-l
Searchable Archive: http://vegan.net/lb/archive http://lbdigest.com Load
Balancing Digest
--------------------------------------------------------
This message may contain confidential information and is intended for
specific recipients unless explicitly noted otherwise. If you have
reason to believe you are not an intended recipient of this message, 
please delete it and notify the sender. This message may not represent
the opinion of IntercontinentalExchange, Inc. (ICE), its subsidiaries or
affiliates, and does not constitute a contract or guarantee. Unencrypted
electronic mail is not secure and the recipient of this message is
expected to provide safeguards from viruses and pursue alternate means
of communication where privacy or a binding message is desired.
_______________________________________________
lb-l mailing list
lb-l@vegan.net
http://vegan.net/mailman/listinfo/lb-l
Searchable Archive: http://vegan.net/lb/archive
http://lbdigest.com Load Balancing Digest
_______________________________________________ 
lb-l mailing list
lb-l@vegan.net
http://vegan.net/mailman/listinfo/lb-l
Searchable Archive: http://vegan.net/lb/archive
http://lbdigest.com Load Balancing Digest
 

_______________________________________________
lb-l mailing list
lb-l@vegan.net
http://vegan.net/mailman/listinfo/lb-l
Searchable Archive: http://vegan.net/lb/archive
http://lbdigest.com Load Balancing Digest
Received on Mon May 7 17:24:40 2007

This archive was generated by hypermail 2.1.8 : Mon May 07 2007 - 17:24:40 EDT