Wednesday, May 6, 2015

hardware - Incredibly low disk performance on HP ProLiant DL385 G7




As a test of the Opteron processor family, I bought a HP DL385 G7 6128 with HP Smart Array P410i Controller - no memory.

The machine has 20GB ram
2x146GB 15k rpm SAS + 2x250GB SATA2, both in Raid 1 configurations.
I run Vmware ESXi 4.1.



Problem:
Even with one virtual machine only, tried Linux 2.6/Windows server 2008/Windows 7, the VMs' feel really sluggish. With windows 7, the vmware converter installation even timed out. Tried both SATA and SAS disks and SATA disks are nearly unsusable, while SAS disks feels extremely slow.

I can't see a lot of disk activity in the infrastructure client, but I haven't been looking for causes or even tried diagnostics because I have a feeling that it's either because of the cheap raid controller - or simply because of the lack of memory for it.

Despite the problems, I continued and installed a virtual machine that serves a key function, so it's not easy to take it down and run diagnostics.
Would very much like to know what you guys have to say of it, is it more likely to be a problem with the controller/disks or is it low performance because of budget components?

Thanks in advance,


Answer



The HP Smart Array P410 is a fine controller, but you will get poor performance out of it if you don't have the battery-backed or flash-backed cache units installed. The cache makes a tremendous difference in that writes are buffered by the cache memory before being committed to disk. You get the write confirmation to the application without having to incur the latency of the physical disk drives.



Here's a 4GB dd on a similarly-spec'd system (DL380 G7 with 24GB RAM and a p410 with 2 x SAS disks and 1GB Flash-Backed Write Cache). The RAM helps a lot in a test like this, but you get the idea...



[root@xxxx /]# dd if=/dev/zero of=somefile bs=1M count=4096
4096+0 records in

4096+0 records out
4294967296 bytes (4.3 GB) copied, 3.70558 seconds, 1.2 GB/s


But realistically, your write performance with two SAS drives in a RAID 1 on that controller with the appropriate cache should be between a sustained 130-170 megabytes/second. A quick iozone test on the above server configuration shows:



[root@xxxx /]# iozone -t1 -i0 -i1 -r1m -s16g
Write
Avg throughput per process = 166499.47 KB/sec
Rewrite:

Avg throughput per process = 177147.75 KB/sec


Since you're using ESXi, you can't run online firmware updates. You should download the Current Smart Update Firmware DVD, burn it to disk and make sure your system is patched to a relatively recent level.



Here are the controller's quickspecs:
http://h18004.www1.hp.com/products/quickspecs/13201_na/13201_na.html



You will want to purchase one of the following, ranging from $350-$600 US:

512MB BBWC


512MB Flash Backed Write Cache

1G Flash Backed Write Cache



To answer your question, the cache solution will help the most. Additional disks won't make much of a difference until you handle the caching situation.



*Note for other users. If you have cache memory on recent HP controllers with up-to-date firmware, there is a write cache override available if you have RAM on the controller but no battery unit. It's slightly risky, but can be an intermediate step in testing what performance would be like on the way to buying a battery or flash unit.


No comments:

Post a Comment

linux - How to SSH to ec2 instance in VPC private subnet via NAT server

I have created a VPC in aws with a public subnet and a private subnet. The private subnet does not have direct access to external network. S...