Comments for Network Performance Testing

Comment on Performance harness project by Network Performance Testing » What happened to the Performance Harness Project?

Network Performance Testing » What happened to the Performance Harness Project? — Wed, 30 Dec 2009 18:14:18 +0000

[...] of you have emailed asking what has happened to the Performance Harness Project I started writing about a few months ago. I wanted to give you all an [...]

Comment on WAN optimization testing by Alan

Alan — Thu, 01 Oct 2009 17:26:39 +0000

@steve
Yes, different TCP options on different OS make a big/huge difference in WAN testing.

Comment on Performance harness project by Network Performance Testing » Documenting the harness

Network Performance Testing » Documenting the harness — Tue, 29 Sep 2009 07:50:11 +0000

[...] I spent some time documenting the performance harness I’m building. Not that exciting – I’m really more interested in building tests and experimenting [...]

Comment on BreakingPoint Elite evaluation by steve

steve — Tue, 15 Sep 2009 05:57:45 +0000

There are a lot of different metrics that you could list for the query you have – and it really is dependent on configuration and options. We saw all sorts of numbers in our evaluation, and worked with the BreakingPoint folks to optimize our settings.

As for pricing, I think that’s something to approach the BreakingPoint folks about as well. I think they have general information in various places on the web.

Comment on BreakingPoint Elite evaluation by sandeep

sandeep — Mon, 14 Sep 2009 11:33:33 +0000

any update on the breaking point systems traffic generation tools’ performacne numbers?

say, per card/blade, how much performance are you seeing?

also, it would be nice to get information about the pricing details of this tool

Comment on TesLA Standards posted by Jim Zupon

Jim Zupon — Wed, 02 Sep 2009 20:21:24 +0000

Steve,

Thanks for keeping this going. Have you seen the Spirent Blog? It has some good general discussion on the importance of performance testing. Here is a link to one of the latest http://www.spirent.com/Blog/Broadband/2009-08-25%20Why%20Test%20Realism%20Matters.aspx

Thanks.

Comment on Evolution of performance testers by sathish

sathish — Wed, 01 Jul 2009 12:12:33 +0000

When it comes to software testing I have some very strong opinions about the background one should bring to the table. Not necessarily the educational background, as I’ve seen first hand how non-technical degree holders can hold their own, but really more of an emphasis on the experience they bring to the table. I was asked the other day what makes performance testing seemingly difficult or separate from more the more traditional functional based testing? What skill set is needed, what allows some to do it well and others not so well. While this is not intended to be prejudicial or discouraging I’ll defer to my favorite quote, “It is what it is!” made popular by a contestant on a cooking reality TV show(Top Chef) I hate to admit I was addicted too even though I can hardly make toast.

Comment on WAN optimization testing by Roo

Roo — Wed, 01 Jul 2009 07:47:47 +0000

You make some good points about the failure of the network equipment vendors to understand the requirements of their customers. I’ve used a HP NetworkTester in the past and the default content it returned for HTTP responses wasn’t even HTTP. The data returned wouldn’t exercise a DPI box and if you then uploaded your own real content the performance slowed to a crawl. There HDDs in the boxes were 2.5 inch laptop drives and the biggest problem we had with those were failures every six months or so. Since repair meant sending the boxes to Germany it wasn’t the best bit of kit I’ve worked with.

Comment on WAN optimization testing by steve

steve — Tue, 02 Jun 2009 21:32:45 +0000

Alan,

Can you elaborate on what you mean by the behavior of client and server machines? I know that a lot of vendors pre-package functionality into some protocols that allow you to tune how clients and servers behave. In particular for HTTP, most vendors allow you to change the headers and details in the packets.

I assume you might be talking about the actual TCP level behavior?

Comment on WAN optimization testing by Alan

Alan — Tue, 02 Jun 2009 18:13:06 +0000

Yes, I fully agree with you; particularly for the file-size limitation. It’s really frustrated that the vendors could not solve it for years. This will be the niche for new players to break into the market.
In term of content; I’m not that concerned; but I do concern about different types of client/server machine behaviors. Every combination of client/server pairs may have slightly different behavior and there is not test equipment can even closely to emulate it!