Hello,

> Why is there this significant difference in single-task throughput
> between the two kinds of storage, knowing that both storage systems
> perform more or less similar at peak performance?

  I think the difference is because controllers on the XIV and other
storage boxes are tuned so they do not give their full power to one
I/O thread on one host. They are tuned to serve many hosts with many
I/O threads at the same time. Try to run 8 threads of mixed load
(read/write, sequential/random) and I can almost bet that SAN box will
outperform internal RAID controller by quite some margin.

> What should I do to increase the performance when doing single-process
> I/O to the XIV? Could it be related to the multipath system

  I do not think you can do much to improve single threaded sequential
read performance of XIV. IBM even presents this fact that there is not
much to tune as an advantage. According to them XIV is quite good in
autotuning itself.
  For single process performance SAN boxes are not good solution. As
you saw internal RAID controller can beat them for a fraction of a
price.

  By the way did you try to test random read performance and latency
of XIV? Looking into architecture (multiple hosts with 7200rpm SATA
drives interconnected via 1Gb Ethernet) it should not perform well in
this area.

  Mindaugas

_______________________________________________
rhelv5-list mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/rhelv5-list

Reply via email to