On Wed, Apr 1, 2009 at 11:11 AM, Damien POBEL <[email protected]> wrote:
> > This article is the second the so called "eZ Publish Knowledge > > Series", more shall come, stay tuned ! > > > > Any feedback is of course greatly appreciated ! > Thanks for this article, this is a very interesting (and extreme) > benchmark :) Can you tell us what kind of hardware and what kind of site > (standard ezwebin templates ?) you use for it ? CPU: Quad xeon 1.86 GHz HDD: FUJITSU MBA3147RC, RAID 1 RAM: 4 GB OS: Virtualized debian, kernel 2.6.20-14 eZ publish: vanilla eZ flow I think it could be interesting to make another test during a longer > time (for instance 15 minutes) and where the content cache is cleared > every minute, this test could be a more realistic use case and the > average requests/sec number would be very interesting ! Yes, we can do that indeed. Note that the benchmark on the article was made by clearing the contentcache (using ezcontentcache.php) for the tested node 10 times during the bench. We do intend to sync it correctly (e.g. integrate the cache clear / create operation in the jmeter benchmark). -- Bertrand
-- Sdk-public mailing list [email protected] http://lists.ez.no/mailman/listinfo/sdk-public
