Why is read-performance so slow on EVA4000 from RHEL4?

[Jesper Sivertsen]

Hello all.

I have 2 EVA4000 with 40 disks each ( 20x300Gb and 20x146Gb ) all 10K fiber-disks.

Both have two disk groups. One with 20x300Gb and one with 20x146Gb.

The 2 EVAs are located at 2 different sites, but is connected and used for HA on our ESX environment.

Firmware on both EVA4000 systems is: CR0EB0xc3p-6100.

It is mainly used for Oracle-databases and VMware ESX-Server.

All Oracle database-servers in the production environment are HP DL585's. Test environment runs on DL385. They are all equipped with HP FC2242SR 4Gb HBA ( Emulex ) for fiber connectivity.

OS is RedHat EL 4.

The BIOS on both DL385 and the HBA's are all up to date, the same goes for drivers and other software involved.

In general I can only read about 40Mb/s and write 120 Mb/s.

I am measuring with linux dd. I use a 10Gb file, so I'm sure not to hit the read/write cache.

I have tried different tricks and tweaks in linux but noting helps. ( Readahead cache, elevator=none )

I have even tried to make a single lun on each EVA and stripe them in linux, but I get the same result. This tells me that the bottleneck may not be the EVA, so it could be the san-switch, the HBA, the HBA driver, etc.

I have read a lot of threads on the internet about this issue, but I have never found a solution.

I have even had a support-case with HP, where they checked all settings and hardware and everything was ok.

Does anyone have a successful story regarding read performance from EVA4000? Where could the bottleneck be, and what is causing it?

Regards Jesper.

[Uwe Zessin]

Short version: a bug in the controller firmware that affects some arrays - upgrade to XCS V6.200.

[Jesper Sivertsen]

Thanks i wil try that.

What read-performance could i expect?

Jesper

[Víctor Cespón]

With 20 disks you should be getting 200 MB/s read and 160 MB/s write in a vdisk in VRAID 5.

But, as the EVA is in continuous access with another EVA, the DR Group log can take space and I/Os on the disk group.

[Richard Tengdin]

Have you verified that the volumes are on different controllers (you are load balanced on the HSV controllers)?

I don't believe the Linux multipath drivers do any load balancing yet, they are pure failover drivers, and you may be trying to access all your volumes on the EVA out the same controller port.

You can see what is going on by monitoring port traffic on your SAN switches or by running Windows PerfMon on the CV server and pulling up the EVA port counters.

[Jesper Sivertsen]

Hello again.

I am now on XCS V6.200 with no big difference.

Switch-firmware is also upgraded.

I se anything from 30 to 60 Mb/s but nowhere near 200Mb/s.

I think the drivers that HP delivers for RHEL4 is only failover.

Even that if i only use 1 fibre at 2GB/s it shuld be ok to se 200Mb/s.

Any idears?

[Sivakumar MJ._1]

Jesper,

How is the EVA connected to the Linux.

What is the throughput you get in the SAN Switch

IF you have tried upgrading the firmware, check the compatability of the fimwares of SAN Switches with the EVA Controller firmware and the HBA..

Check the IO transfer rate from Linux Server to the Switch to the EVA..