PDA

View Full Version : Seagate 7200.11 failure rate


Olin Coles
14th October 2008, 10:47 AM
In the past year or more, I have been using the 500GB Seagate 7200.11 hard drive for many of my system builds. While this drive has offered impressive performance, likely part of the 32MB cache buffer, this drive has had more total failures than I can remember in a very long time. At the moment, I have a running total of 7 RMA's for this drive. This morning I serviced a system that appears like #8. Is anyone else having high failure rates with this drive?

Bruce Normann
14th October 2008, 11:20 AM
My 7200.11, ST3750330AS, 750GB has been running mostly 24/7 for about a year, no problems. Same series, with 250GB platters.

Olin Coles
14th October 2008, 11:24 AM
The platters seem to be fine on most of these failures; no noise or anything. The drive(s) simply stop being recognized by the BIOS.

Bruce Normann
14th October 2008, 11:38 AM
10% of the total reviews for this drive are for failure.
The "Enterprise" version, Seagate Barracuda ES.2 ST3500320NS 500GB has ~ 18% failure reviews.

FWIW.... this looks to be about double the nominal failure rate, among Newegg reviwers, anyway.

Olin Coles
14th October 2008, 11:41 AM
The model that keeps failing on my systems is the ST3500320AS. You're right though, even 10% seems abnormally high.

mystikmedia
14th October 2008, 01:08 PM
I have experienced the exact behavior you have mentioned before, even with SSDs. It seems to be related to the RAID card. I have experienced it with Areca cards. In my case, when it would happen, often just rebooting would result in it working.

Olin Coles
14th October 2008, 01:23 PM
All of my failures were in RAID-1 arrays for small servers.

Bruce Normann
14th October 2008, 01:31 PM
Yup, that's the model on Newegg, ST3500320AS...984 reviews, 101 failures.


The model that keeps failing on my systems is the ST3500320AS. You're right though, even 10% seems abnormally high.

Olin Coles
14th October 2008, 02:35 PM
Ouch. I probably shouldn't mention that I ordered four of these from NewEgg on Friday to replace the one's that have gone bad.

Miles Cheatham
14th October 2008, 03:24 PM
I had two of the Seagate Barracuda ES.2 ST3500320NS 500GB drives and one failed within the first three months of operation and the other has been running trouble free for the last two years or so. I also have two of both the 750GB and 1TB enterprise versions and they have worked flawlessly since I recveived them. I have heard numerous complaints about both the 7900.11 and ES version of the 500 GB drives.

Matt Williams
15th October 2008, 01:10 PM
If you're looking for another drive, I've heard good things about the Seagate ST3640323AS (http://www.newegg.com/Product/Product.aspx?Item=N82E16822148335)
Edit: I did some more checking and this drive doesn't have the greatest access times. Read and burst speeds are decent though.
http://www.hardforum.com/showthread.php?t=1342467

My personal drive of choice is the WD6400AAKS (http://www.newegg.com/Product/Product.aspx?Item=N82E16822136218). They're fast, and I haven't had any failures yet, although I don't know how they'd perform in a server environment.

Austin Downing
24th June 2010, 02:39 AM
I have 2x1.5 7200.11 in my personal server, and 1x1.5 7200.11 for a video editing rig, and they are going strong hopefully i will have 2 more 1.5 by the end of summer

Daryl Greene
25th November 2010, 02:30 AM
Slightly OT but, with a 2TB drive, would you format it as a single drive or break it into several virtual drives?

Daryl Greene
1st December 2010, 05:54 PM
Well, I guess no one has an opinion on my question?

Olle P
6th December 2010, 07:41 AM
Slightly OT but, with a 2TB drive, would you format it as a single drive or break it into several virtual drives?Depends on the use. If it's your only physical drive, then multiple partitions help keeping things tidy. (I use one partition for the OS, one for programs and one for file storage.)

Cheers
Olle

Daryl Greene
6th December 2010, 05:31 PM
The HDD is one of those 5900rpm Seagate drives. Seems to work decently. just don't know if it will slow it down after I get a large amount of data stored on it. Right now I only have about 27GB stored on it so it seems to run pretty quick.

I always heard that breaking a large drive up made accessing information stay at a faster rate. Sounds hokey to me, but who knows? Never played with the different possibilities to see.

Bruce Normann
6th December 2010, 09:16 PM
I just did two partitions: one for the OS and then the rest. The performance boost is real. I didn't test mine, I let other people do the testing....:D Then I just applied their knowledge. Make sure you choose a decent size for the OS partition. It gets real messy if you need to adjust the partition size. OTOH, the smaller you make it, the faster it is....

Honestly though, the best bang for the buck is a $90 SSD as the boot drive. You won't believe the difference.


The HDD is one of those 5900rpm Seagate drives. Seems to work decently. just don't know if it will slow it down after I get a large amount of data stored on it. Right now I only have about 27GB stored on it so it seems to run pretty quick.

I always heard that breaking a large drive up made accessing information stay at a faster rate. Sounds hokey to me, but who knows? Never played with the different possibilities to see.

Olle P
7th December 2010, 01:05 AM
I always heard that breaking a large drive up made accessing information stay at a faster rate. Sounds hokey to me, but who knows?There is, at least in theory, a real performance boost based on the fact that the reading head won't have to travel that far between the endpoints of the partition, thus somewhat reducing the read/write times.

The actual performance boost depends on two factors:
1) How much data fragmentation is there?
The performance boost compared to a bigger partition is somewhat proportional to the level of fragmentation. With frequent (weekly or more often) defragmentation runs this shouldn't be a problem in the first place.
2) How often does the head have to switch between partitions?
If it has to constantly alternate between partitions the result might be reduced rather than increased performance.

Cheers
Olle