6 messages in org.freebsd.freebsd-scsiRe: RAID performance/benchmarking| From | Sent On | Attachments |
|---|---|---|
| Brandon Huey | Apr 14, 1998 12:41 pm | |
| Andreas Klemm | Apr 14, 1998 2:12 pm | |
| Simon Shapiro | Apr 14, 1998 4:52 pm | |
| Justin T. Gibbs | Apr 14, 1998 5:54 pm | |
| Tom | Apr 14, 1998 6:24 pm | |
| Simon Shapiro | Apr 14, 1998 9:18 pm |
| Subject: | Re: RAID performance/benchmarking | |
|---|---|---|
| From: | Tom (to...@sdf.com) | |
| Date: | Apr 14, 1998 6:24:28 pm | |
| List: | org.freebsd.freebsd-scsi | |
On Tue, 14 Apr 1998, Brandon Huey wrote:
Dell PowerEdge 4200/300, 96MB RAM, 3 9GB Seagate Barracuda SCA drives DPT PM3334UW w/64MB cache
FreeBSD 2.2.6 DPT kernel options: DPTOPT DPT_MEASURE_PERFORMANCE DPT_FREELIST_IS_STACK
i created an array group with the dos util 'dptmgr' from default settings.
using IOZONE and Bonnie, i create a 128MB file for read/write benchmarking.
i am aware of the performance implications of RAID-5, but I am getting about 2MByte/sec max write transfer rate compared with 19MByte/sec on reads.
For a pci ultrawide controller, are these figures suspect? What kind of write performance hit should one expect w/RAID-5 ?
No, those figures are typical for RAID5 on the DPT PM334 when used with the default stripe size.
However, do you want to optimize your system for bonnie/iozone (sequentional and single-user)? If so, there are lot of things you can change. However, then random-access, multi-user things don't benchmark well!
I have a DPT PM334UW in a mail server with a 5 disk RAID5 array, and I'm constantly amazed about how well it performs real-world tasks.
Tom
To Unsubscribe: send mail to majo...@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message