Re: I/O benchmarking

Do you have any data from the filer during the benchmark run?

-Blake

On Mar 21, 2010, at 4:51 PM, "Suresh Rajagopalan" <SRajagopalan@williamoneil.com

wrote:

Here are some numbers from crystaldiskmark. On crystaldiskmark the max file size is 1000Mb. The host has 64G of RAM and 8 six-core processors.

1.   DL785G6, 100Mb file local disk
   
   

a. Seq Read 216.5MB/sec Write 78.4MB/sec

b. Random 512k Read 58.14MB/sec Write 301.9MB/sec

c. Random 4k Read 26.2MB/sec Write 41.3MB/ sec

2.   DL785G6 100MB file on Filer LUN (NTFS)
   
   

a. Seq Read 175.1MB/sec Write 100.7MB/sec

b. Random 512k Read 103.7MB/sec Write 71.44MB/ sec

c. Random 4k Read 15.7MB/sec Write 7.6MB/sec

3.   DL785G6 1000MB file on local disk
   
   

a. Seq Read 236.8MB/sec Write 92.7MB/sec

b. Random 512k Read 49.71MB/sec Write 217.2MB/sec

c. Random 4k Read 1.33MB/sec Write 20.63

4.   DL785G6 1000Mb file on filer LUN (NTFS)
   
   

a. Seq Read 164MB/sec Write 98.8MB/sec

b. Random 512k Read 101.1MB/sec Write 63.2MB/sec

c. Random 4k Read 13.9MB/sec Write 7.8MB/sec

Suresh

From: Timothy Naple [mailto:tnaple@BERKCOM.com] Sent: Friday, March 19, 2010 9:21 PM To: Suresh Rajagopalan Cc: Toasters List Subject: RE: I/O benchmarking

Suresh,

Performance benchmarking is a science that involves many variables. I am not familiar with CrystalDiskMark but I just downloaded the source for 3.0 RC2 and will have a look to see how applicable it could be to a filer vs local disk comparison. Can you add some more details about your configuration? (any options you run with the test, specs/model of the server including controller/RAID card(s), OS on the server, disk model in the server, disks in the filer, model of the filer, ONTAP rev, etc). A lot of detail is going to be required to make any headway or recommendations for a valid test.

Thank you,

Tim

From: owner-toasters@mathworks.com [mailto:owner- toasters@mathworks.com] On Behalf Of Suresh Rajagopalan Sent: Friday, March 19, 2010 8:55 PM To: Toasters List Subject: I/O benchmarking

I’m using the free tool Crystaldiskmark to do some I/O comparison b etween local disk and our filers. On at least one system (SAN conne cted), the local disk (6 disks in RAID1) consistently comes out ahea d in both read and write. Filer is lightly loaded, and this is on a 56 disk aggregate. I’m kind of stumped on this one, and would like to know if:

a) Are there any other commonly used benchmarks which I can try with the filers?

b) This is on a 2G FC SAN. How much improvement can I expect with 4G or 8G?

Thanks

Suresh