Jump to content

What is your parity check speed?


gtroyp

Recommended Posts

I was wondering, with new systems, faster SATA and faster drives what parity check speeds were like for other people. In order to gauge whether or not my system was running good bad or indifferent, I thought I would ask around.

 

For the record: my last parity check today, running 6.b6 on the system below,  Duration: 8 hours, 40 minutes, 1 second. Average speed: 96.2 MB/sec

Link to comment

Might be interesting data ... but quite frankly it's almost useless without also knowing the specifics of the system.    For example, if you built a nice new system with a bunch of 6TB WD Reds, but also had a couple of older, slower drives in it, those drives would limit the parity check speed, so it'd be FAR slower than if they weren't there.

 

Link to comment

I call shenanigans on the 150+ MB/sec result.

 

I'd be interested in seeing what an array using ONLY the new 1.2TB/platter 6TB WD Reds would do ... I would not be surprised to see them exceed that on the outer cylinders -- although they certainly wouldn't average that for the entire check.    The 1TB/platter WD Red Pro's (which run at 7200 rpm) should also have speeds above that on the outer cylinders.

 

... we're probably not far away (1-2 years) from having drives that can actually average that  :)

 

If a 22TB system has enough capacity for you, you could actually get FAR better than 150MB/s speeds ==>  just build a 22TB system using 24 of these:

http://www.newegg.com/Product/Product.aspx?Item=N82E16820147356

 

You would, of course, have to be a bit less price sensitive than the average UnRAID builder  8)

 

 

Link to comment

I have a 2 x Hitachi and 2 x Toshiba 3T drives on old hardware (AMD 740G chipset). The array is typically right around 150Mbps on a parity check. With the faster 3T+ drive sizes I actually expected more results in that speed range.

 

My system starts in that range, but doesn't average nearly that -- as I noted above the average is 111+ MB/s.    What is your system's average for the entire check?

 

I assume you're using 7200rpm drives, since there aren't any 1TB/platter drives shown in the platter capacity database for Hitachi drives (even the 6TB Helium unit uses 860GB platters).    This would give them sustained rates slightly faster than the WD Reds ... but they'd still slow down a good bit on the inner cylinders.

 

Link to comment

Yes, 5-1/2 hours to parity check with 3 x 3T data drives.

2 x TOSHIBA DT01ACA300

2 x Hitachi HDS723030BLE640

 

The Toshiba drives zero'd during the preclear at an average speed of 148Mbps.

 

Please post the syslog snippet for the parity check. Thanks.

 

Unfortunately I recently rebooted the server after the first of the month and I'm leaving to go away all weekend tonight. I could type a copy of any number I felt like anyways, so if you don't believe it I really don't care.

 

 

Link to comment

Yes, 5-1/2 hours to parity check with 3 x 3T data drives.

2 x TOSHIBA DT01ACA300

2 x Hitachi HDS723030BLE640

 

The Toshiba drives zero'd during the preclear at an average speed of 148Mbps.

 

That's an excellent average ... but easy to believe.  The Toshiba drives are 1TB/platter units, so these should be about 16% faster than the Hitachi's (which are only 860GB/platter units).

 

It's a bit surprising that the parity checks are actually faster than the pre-clears ... but that may simply be a function of the Pre-Clear code vs. UnRAID's parity check logic. 

 

 

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...