Storage Boards Cleanup
To make it easier to find information about HPE Storage products and solutions, we are doing spring cleaning. This includes consolidation of some older boards, and a simpler structure that more accurately reflects how people use HPE Storage.
Disk Arrays
cancel
Showing results for 
Search instead for 
Did you mean: 

EVA3000 performance

Darren Lauser_2
Occasional Advisor

EVA3000 performance

Hello,

We are seeing 17.5MB/s read performance on a 500GB RAID5 LUN on our EVA3000. We are barely using this array since we have only moved a handful of users to this system.

There are two disk cabinets and 16 spindles at 10K RPM and 300GB each, but we are only have 200GB of data on there so far.

We should be able to expect better data throughput than this ... shouldn't we?

Regards,
Darren

3 REPLIES
Ivan Ferreira
Honored Contributor

Re: EVA3000 performance

You can use IOMETER/IOZONE to test the performance, because we don't know the test you used, for example, one copy is a single threaded operation, you can run multiple copies to obtain more transfer rate.
Por que hacerlo dificil si es posible hacerlo facil? - Why do it the hard way, when you can do it the easy way?
Peter Mattei
Honored Contributor

Re: EVA3000 performance

The MB/s is just one metric of performance.
Usually you use that for sequential, large block IOs (>64kB).

For transactional IO (usually in the 2 to 8kB range) the IO/s are much more important and actually the limitaion.

So we need to know what IOs you are measuring.

The rule of thumb says that you can have ~100 IO per disk in an array (without cache help) So with your 16 disks you can expect approximately 1600 IO/s which with 8kB blocks wold result in 12.8MB/s.

So, if you are in the 8kB range your 17.5MB/s are very good!!

Cheers
Peter
I love storage
Darren Lauser_2
Occasional Advisor

Re: EVA3000 performance

Hello,

Thanks for your insights. I was able to get better throughput using multiple threads.

Regards,
Darren