DL 380 G5 RAID 5 Slow

[Patrick Zhang]

I met with the same problem with Chris.
The server is HP DL380G5, with 8*146GB 10Krpm SAS HDs (Raid5 using P400-256MB).
And the average write speed is 30MB/s as I tested. I've used a P400-512MB to have a try, almost the same write speed.
Any manners to improve the performance will be appreciated.

[Chris Camm]

I have put the latest firmware CD in and it found no updates needed.

Still running tests using pc's and servers on the same network switch.

Any other suggestions welcomed.

[Patrick Zhang]

The problem was solved!
We upgraded the Raid contoller from 256MB cache to 512MB with the battery. And adjust the cache percent, 50% for read and 50% for write. The current writing speed is about 95 MBps and the former is 30 MBps.
Maybe you could have a try like this.

[ViceSquad]

This is interesting because I have the same configuration (P400 256mb) and found the RAID5 painfully slow. In fact, when I did performance tests of RAID5 vs RAID10 the same operation that took 11 minutes on the RAID10 took more than 40 minutes on the RAID5.

Will installing the battery cache help here or do I need to upgrade the memory to 512mb as suggested by the recent poster?

[Chris Camm]

Update - I have booted the production server with the latest firmware CD and it found no updates where needed. Since the reboot:

Writing files to the Server 600Mbs
Reading files from the server 212Mbs.

Now I must be really missing something here because RAID 5 is supposed to be quick on Reading files from the server and not writing to it.

I changed the cache ratio several times and went through all the settings to see if it made a difference and it has made no significant improvement. There is no one in the office other than me writing to the server so this is about as quick as its going to get. I even enabled the Physical Drive Write Cache State.

I am know not sure if the problem is with the NIC or the Array now.

Has anyone managed to get good read speed on the servers? Writing even to the RAID 5 is fine its the same as writing to the RAID 0+1.

Thanks for any help

[Patrick Terlisten]

Hello,

are you using NIC teaming? If yes, please disable the teaming for a test. Maybe the teaming is play you a raw deal (Transmit / Receive Loadbalancing).

The interesting fact is the read peformance: A RAID 5 with a lower read performance, then write performance, is really spooky. :)

Best regards,
Patrick

[Chris Camm]

I have disabled to Team on a spare server and just enabled one NIC. I get the same throughput.

I think this is NIC related though. I have tried all the Drivers apart from the Broadcom driver as I couldnt install it for some reason.

I will try to download and install again.

Thanks

[Erik Ahlström]

Hi Chris!

I have installed a Hp DL360 G5 p400 256 mb controller using firmware 2.08 which also is painfully slow!!

I have installed W2k3 x64 SQL server 2005 x64
I have 6 gb ram installed and a Raid5 with 4 discs. all discs are SAS 146 GB 10krpm

Im aware that one logical disk raid5 is not the way to go but we need the discspace and budget forced us to this machine.

To unpack a *.bak file ( rar lowest pack rate ) packed file size 6 gb unpack 40 gb ( one database ) took about 3½ h. tested the same file at home with an areca 1220 controller raid5 8 discs ( though its sataI discs 250 gbs ) and made it in 14 minutes.

from what I can see that you have done so far you havent got any speed yet.

Is this a wellknown issue with this controller ?

I will keep googling and post here if I find anything usefull.

Kind regards Erik

[Antony Scerri]

Hi

I was just about to put this into the business forum when I found this post, so im going to post it both forums.

I have a DL380 G5 with the Smart Array P400. New to this hardware sestup, so if you need more info let me know. Its running Windows 2003 64bit Enterprise (32Gb RAM). The current driver and firmwares versions are as follows:

SA P400 Win Driver : 6.8.0.64
SA P400 firmware : 4.06
DG146ABAB4 firmware : HPD7

I have 8 DG146ABAB4 drives installed. 3 Logical disk setup 2 logical drives mirrored, and a 3rd RAID 5. The controllers accelerator status is:

ACCELERATOR STATUS:
Logical Drive Disable Map: 0x00000000
Read Cache Size: 52 MBytes
Posted Write Size: 156 MBytes
Disable Flag: 0x00
Status: 0x00000001
Disable Code: 0x0000
Total Memory Size: 256 MBytes
Battery Count: 1
Battery Status: 0x0001
Parity Read Errors: 0000
Parity Write Errors: 0000
Error Log: N/A
Failed Batteries: 0x0000
Board Present: Yes
Accelerator Failure Map: 0x00000000
Max Error Log Entries: 12
NVRAM Load Status: 0x00
Memory Size Shift Factor: 0x0a
Non Battery Backed Memory: 0 MBytes
Memory State: 0x00

The drives dont have their indivudal write caches enabled.

So the odd behaviour im seeing and I just wanted to check if this was normal, as it seems really odd to me. I wrote a simple program just to check disk speed in writting data to the disk after noticing what i thought was slow app disk writes. The program create a 1GB file writting blocks of bytes at a fixed size per run. If the block size is close to 64KB (2^16 bytes) I get speeds up to 1300MB/s. If it writes blocks below that threshold its drop, so at 10KB block size its achieving 600MB/s (still generating a 1GB file). This I can understand and im ok with as it requires more writes).

However pushing the block size up to 200KB I only get 220MB/s max sometimes as slow as 90MB/s. Going above this its consitently slower. The average seems to be about 120MB/s when im writing blocks of 1MB at a time.

My question why on earth does it achieve such slow rates when pushing larger blocks of data at it.

Tony