At work we are building a new webserverfarm, we are replacing our current farm which consists of 11 SunFire V210 servers running Apache with 4 Sun T2000 servers with 8 cores and 8Gb of memory.

Currently we are building the new T2000 webservers in our staging environment because it all needs to be thoroughly tested and signed off before all this can actually go into the production environment.

At a certain point while we were load testing one of the T2000’s we saw the following on the commandline when running an uptime :

bash-3.00# uptime
3:53pm up 13 day(s), 23:22, 2 users, load average: 2546.46, 1461.60, 632.73

I was amazed by the load average figure i must say…. i mean i’ve seen servers running at load averages of 200 with stress tests but this i’ve never seen before : a load average of 2546!!!

Does anyone know of any occurances or stories about T2000’s with such high load average figures ?

I am curious how much these servers can actually take since i haven’t seen this one break at all, it was happy to continue to run at this load average and the CLI was still responsive. I’ve tried searching around for any testresults which show at which point a T2000 breaks but haven’t found any yet.  Any pointers ?


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s