Link Details

I'm not a big fan of JBoss, but this entry is written by a Terracota CTO. Not sure if he unbiased.

I think the fact that the URL is terracottatech.com makes that pretty clear =)

Hi Kupolov,

Not to defend here, because you are correct. I am not trying to hide the fact that I am CTO. But I am not sure how bias factors here. This was JBoss's own benchmark from their site. And we didn't run it...I was unclear in my blog posting but the customer ran both JBoss and Terracotta (having JBossians onsite and Terracotta onsite during the 2 test runs). It is repeatable outside the customers' 4 walls. Anyone can try it.

Everybody has a bias, I don't see a problem with that.

Probably would have been useful to spruce the article up a bit though... a nice graph or something?

will do. Thanks for the suggestion, BTW.

I'm interested to know what this "official test" is. As the tech lead on JBoss Cache, I can confirm that we have not released any official benchmark tests.

http://wiki.jboss.org/wiki/Wiki.jsp?page=WhatShouldWeExpectOfThePojoCachePerformance

Its on your site. Written by Ben Wang, I guess.

Cheers,

--Ari

That is specific to PojoCache. AFAIK you tried to bench TreeCache?

PojoCache uses fine-grained replication (thanks to JBoss AOP) to overstep the java serialization performance penalty.

Not sure I follow what you're saying. The link mentioned shows results for TreeCache. Are those valid or invalid?