Hitachi Deskstar 7K1000 in RAID 0: Is Two Terabytes really better than One?
by Gary Key on April 19, 2007 12:15 AM EST- Posted in
- Storage
Hard Disk Performance: HDTune 2.53
Our first screenshot is the Hitachi drive with Automatic Acoustic Management (AAM) and Native Command Queuing (NCQ) turned on in RAID 0 operation with the second screenshot showing the same configuration with a single drive. The RAID 0 setup shows a 98% increase in the minimum transfer rate and a 41% increase in the average sustained transfer rate. The maximum transfer rate also increases 41% although the burst rate is slightly lower. We believe HD Tune has a bug as the RAID 0 results indicate access times improving slightly when in reality they should not change or perhaps even increase slightly. Our HD Tach results bear this out so we will ignore the reported improvements in RAID 0 access times with this program.
The third screenshot is the RAID 0 results with AAM off and NCQ on. The transfer rates are basically the same with the only difference being a further reduction in access times compared to our first RAID 0 test result. We noticed the same drop in access times when turning off AAM during our single drive testing in the first article. Although the access times are improved we believe the benefits in the acoustics department outweigh the slight performance improvements we noticed in our application tests. We found that enabling AAM usually did not alter the test results more than 1% and at times the scores were a tie.
Hard Disk Performance: HD Tach 3.0
We are also including HD Tach results for review. Once again the order of the screenshots is the same as in our HDTune results. In this benchmark we see a 50% increase in the average read rates and a 52% increase in burst speeds over the single drive configuration with access times remaining the same in the AAM on /NCQ on tests. Our RAID 0 results with AAM off show an improvement in access times and sustained transfer rates while burst rates decline slightly compared to our other RAID 0 test results. While these numbers are impressive, we will have to see if they translate directly to improved application scores.
Our first screenshot is the Hitachi drive with Automatic Acoustic Management (AAM) and Native Command Queuing (NCQ) turned on in RAID 0 operation with the second screenshot showing the same configuration with a single drive. The RAID 0 setup shows a 98% increase in the minimum transfer rate and a 41% increase in the average sustained transfer rate. The maximum transfer rate also increases 41% although the burst rate is slightly lower. We believe HD Tune has a bug as the RAID 0 results indicate access times improving slightly when in reality they should not change or perhaps even increase slightly. Our HD Tach results bear this out so we will ignore the reported improvements in RAID 0 access times with this program.
The third screenshot is the RAID 0 results with AAM off and NCQ on. The transfer rates are basically the same with the only difference being a further reduction in access times compared to our first RAID 0 test result. We noticed the same drop in access times when turning off AAM during our single drive testing in the first article. Although the access times are improved we believe the benefits in the acoustics department outweigh the slight performance improvements we noticed in our application tests. We found that enabling AAM usually did not alter the test results more than 1% and at times the scores were a tie.
Hard Disk Performance: HD Tach 3.0
We are also including HD Tach results for review. Once again the order of the screenshots is the same as in our HDTune results. In this benchmark we see a 50% increase in the average read rates and a 52% increase in burst speeds over the single drive configuration with access times remaining the same in the AAM on /NCQ on tests. Our RAID 0 results with AAM off show an improvement in access times and sustained transfer rates while burst rates decline slightly compared to our other RAID 0 test results. While these numbers are impressive, we will have to see if they translate directly to improved application scores.
Facebook
Twitter
RSS
48 Comments
View All Comments
MadAd - Saturday, April 21, 2007 - link
<quote>Sorry but saving even 5 secs out of 10, 5 times a day is not work the extra money to me</quote>I take it you dont play battlefield 2 then. Having just two or three seconds advantage on each mapchange can mean the difference between a round flying a jet or helo, or a few seconds later watching everyone else fly off and being left with a humvee (if you are very lucky) or nothing at all.
Of course not all games have this problem however with bf2 when there are 32 players a side and only 2 jets each then its the quickest in that gets first picks, and if it takes running raid 0 just to pick up that extra second or two, then so be it.
ShadowdogKGB - Saturday, April 21, 2007 - link
My four little Hitachi 80gigs in R0 will load the single player Daging Oilfields in 18-20 seconds. Hows that for real world performance. Or maybe somebody from the church of the anti-raid can explain that away for me. My point of contention from this article is that the author went out of his way to denigrate the concept of raid. And another point is that you don't buy a 1 terabyte hard drives just to put them in raid, and especially put them in raid 0. These babies are for storage. You're definitely are not going to want 2 Terabytes of data sitting on a fragile Raid 0. No, this article is just plain skewed. Now there's gonna be a bunch of knuckle heads pointing to these benchmarks and saying "See? See? I told you so!" Yeah, HL2 Lost Coast. That's not even a real game. And The Sims2? Oh please. Yeah that's real world performance figures right there. Bleh. I'm no programmer or mathematician but I could have done a more decent article on this subject than this amateur.Axbattler - Friday, April 20, 2007 - link
I do not buy the 'extra money' argument that much (**). It's not like performance is the only (*possible) gain from striping two drives. The second drive get you extra capacity, and as long as people choosing to go RAID-0 are using the extra space, then they are not paying a financial premium over buying two drives and running them separately (unless they need to purchase a RAID controller). To me, the main cost from going RAID-0, is the added risk in case of failure.* Though I am in the school of thought that RAID-0, do not provide significant performance boost in the majority of the cases, I do find gains more often than penalties (from overhead).
Regarding from the article results, I am not surprised by the game loading results. I do, however suspect that the performance benefit of RAID-0 may be more noticeable in XP boot up time however (whether that is important enough, I'll shrug to it. Not my cash).
The file copy result make me wonder if there is not a bottleneck elsewhere though.
7.55 *1024 / 100 = 77.312 MB/sec on average. That's the transfer rate of a single drive.
** I do make an exception to people stripping Raptor's. I can't think of many desktop users who have enough 'performance sensive' applications (OS, apps, games - as opposed to multimedia files for instance) they use regularly - so much that they would need a second Raptor in the same rig. I do suspect that those users are really going for the bragging right rather than the 'free performance'.
Lastly, I wonder if RAM Disks, in their current form, are really faster the fastest SCSI drives at loading games. I seem to remember benchies of i-RAM some time ago showing it to edge the 150GB Raptor by not that much.
Griswold - Thursday, April 19, 2007 - link
Absolutely agree. The only winner is the storage industry.gramboh - Thursday, April 19, 2007 - link
No kidding, been waiting for the Seagate for a while. It will also be nice to see 250GB platters (x4) on a 1TB drive. I'm running 2x 7200.10 500's right now and am happy with them. I'd like 1TB to come out to push drive costs down so I use a few for external back-up.BoberFett - Thursday, April 19, 2007 - link
Anybody who stripes drives of this size is asking to lose a lifetime's worth of data. Even assuming it's data than can be reassembled such as ripped or downloaded music and movies, the time required to reassemble that data is pretty significant.goinginstyle - Thursday, April 19, 2007 - link
Glad to see you guys still telling it like it is with RAID 0. I am just waiting on the comments to come in from people who swear it lets them operate their systems at light speed. This drive seems to be really nice but I will wait for the Seagate 1TB to come out before making an upgrade decision. When is it coming out?Gary Key - Thursday, April 19, 2007 - link
The Seagate 1TB drives are due out in four to six weeks according to the last information we had from them.