bigbulus

Seagate 300GB 15K.5 benchmark result here. Is this normal?

Recommended Posts

I installed Seagate 300GB 15K.5 SAS in Dell Server with Dell PERC RAID 5/i.

Seagate P/N: ST3300655SS. Can somebody please tell me this is normal? SAS should be 300Mbps, but why this one is slower?

Result from HD Tach:

>

Result from ATTO:

>

Share this post


Link to post
Share on other sites
I installed Seagate 300GB 15K.5 SAS in Dell Server with Dell PERC RAID 5/i.

Seagate P/N: ST3300655SS. Can somebody please tell me this is normal? SAS should be 300Mbps, but why this one is slower?

Result from HD Tach:

>

Result from ATTO:

>

I'm not familiar with ATTO, but the HD Tach scores are about right. You seem to be conflating the interface speed of SAS with the hard drive speed. They're not the same.

Share this post


Link to post
Share on other sites

Hmm.. 256MB of onboard RAID Cache, along with the HDD's own cache... How can you be for sure, that the cache didtnt fool with your numbers?

But as Jeff pointed out... Interface speed, should not be confused with actual data transfer speed of the head to spindle rear/write operation.

Share this post


Link to post
Share on other sites

It is difficult to really benchmark complicated arrays with regular utilities and many times they show results that are wildly different as in this case.

HDTech graph seems in line with the drive speed as tasted on storagereview. Judging by the decay of the graph, it seems to me that HDTech is NOT influenced by the RAID cache.

ATTO on the other hand shows ridiculously high numbers indicating the cache is working just about right. Increasing the total length to 32MB, and queue depth to 10 might go beyond the cache and show how the drive itself is performing.

Share this post


Link to post
Share on other sites

yeah thats what I experience with my controller also for the ATTO scores.

I have a MegaRAID 320-2E w/ 128MB cache on it and it fools the heck out of benchmarks that dont access the drive directly.

My atto is out of this world like yours because I think ATTO only uses a 32MB test file or something...that entire 32MB fits into your cache on the controller.

I know SiSoft Sandra file system test ( if you wanted to get a nice file system performance eval ) makes a 2GB test file....so I know thats too big to fit in cache so it actually tests the performance quite well.

HD TACH on the other hand tests the drive directly bypassing the cache for the most part...cept in a few cases Ive noticed that the Burst test will get tested in my controllers cache and be really high.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now