New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
A quick question about server performance
I have two dual L5630 servers. I'm running 20 Windows 2008 instances on each server. I wonder if the I/O will not be an issue, will other factors be ok for running 40 windows 2008 instances on a dual L5639 server and roughly get the same performance like the instances are run on two servers? Thanks
Comments
It depends on several factors, the most important being simultanious read/ write speeds which is much higher in ssd than sas/sata
Well, you are pointing out the IO as a potential problem and didn't include the HDDs configuration in your post.
I'm saying if I/O will not be the issue, will other factors do the same job? eg. will the dual L5639 CPU be able to handle 40 windows 2008 instances etc..
Oh, I've misread this. Anyway, it would be possible to run 40 instances, in my eyes, however don't expect a good performance.
sorry my bad, didn't express it clearly.
so I guess the performance will be worst than running 40 instances on 2 separate dual L5630 servers?
For sure it would be worse. If you run 2 seperate nodes, running 20 machines each will have twice the computing power.
I thought 5639 could provide double performance or near to double of what L5520/5630 can provide.
Thanks for your answers.
Well, 5639 is 6 cores, the other one is 4 cores. However if you use 2 nodes with l5630, it would most likely outperform the l5639
Each L5639 CPU is 6 cores/12 threads (60W max TDP), whereas each L5630 CPU is 4 cores/8 threads (40W max TDP). Everything else about the CPUs are the same (clock rate, architecture, etc).
thanks, I've made up my mind I'm not gonna buy a L5639 to replace two L5630.
Cheers