Western Digital's Raptors in RAID-0: Are two drives better than one?
by Anand Lal Shimpi on July 1, 2004 12:00 PM EST- Posted in
- Storage
Pure Hard Disk Performance
To measure "pure" hard disk performance, we took a real world benchmark - in this case, the entire Winstone 2004 suite - and used Intel's IPEAK utility to capture a trace file of all of the IO operations that take place during a single run of Business Winstone 2004 and MCC Winstone 2004. We then use IPEAK to play back the trace, much like a timedemo, on each of the hard drives, which gives us a mean service time in milliseconds; in other words, the average time that each drive took to fulfill each IO operation.In order to make the data more understandable, we report the scores as an average number of IO operations per second so that higher scores translate into better performance.
Keep in mind that these performance scores are best only for comparing pure hard disk performance, and in no way do they reflect the actual real world performance impact of these hard drives.
For descriptions of what the Business and Multimedia Content Creation Winstone 2004 tests consist of, reference those benchmark pages.
While a score of 901 IO operations per second seems impressive, what's important to note is that two Raptors in RAID-0 only offers 20% more performance than a single Raptor in this purely disk-bound test. We have already seen how aggressive caching, pre-fetching and other such real world elements of modern day computers reduce the impact of hard drive performance on overall system performance. With only a 20% improvement in pure disk performance, this doesn't bode too well for RAID-0 offering much of a real world performance boost.
With data from audio editing as well as other more IO intensive tasks, our IPEAK recording of Content Creation Winstone 2004 shows much more potential for RAID-0, with the Raptor RAID-0 array outperforming a single drive by 38%.
Now, it's time to see how these numbers translate into real world performance.
Facebook
Twitter
RSS
127 Comments
View All Comments
Denial - Friday, July 2, 2004 - link
A$$ Masher,You seem to be the only person turned on by my system. Sorry, but I don't swing that way. You'll have a better chance at the internet cafes over in Chelsea.
TheCimmerian - Friday, July 2, 2004 - link
...Anyone have any thoughts on RAID0 for DV capture/editing/rendering?...masher - Friday, July 2, 2004 - link
#71, the statistics and applicability for MTBF and MTTF are a bit complex...so much so that most drive manufacturers themselves usually don't apply them properly (or intentionally mislead people).Technically, you're correct...RAID0 doesn't halve MTBF. However, for what the average user means by "chance of failure", a two-disk Raid0 array does indeed double your chance of a failure.
As to your comment of Raid-0 loading maps faster...true if its a large file (10MB+) and probably not discernably noticeable till you're in the 20-40MB range. For tiny files or heavily fragmented ones, it may even be slower.
Z80 - Friday, July 2, 2004 - link
This article was very informative. However, the statements that RAID0 cuts MTBF in half and that RAID1 doubles MTBF are statistically incorrect. Also, RAID0 does improve game performance especially when large game maps are involved (i.e., BF1942, BF Vietnam, Far Cry). It definitely provides an advantage in online FPS games by loading large maps faster and giving RAIDO equipped players an advantage in first choice for weapons and position. Your tests probably didn't measure map loading times.GokieKS - Friday, July 2, 2004 - link
"What's the difference between losing one 74gig Raptor in RAID-0 array or one 160gig stand-alone drive? THERE IS NO DIFFERENCE!"There is. The chance of you getting a HDD failure increases with every drive you add. A 2 disk RAID-0 array will have the same chance at failure as 2 independent non-RAIDed drives. The difference is, with the independent drives, you lose one drive's worth of data when it fails. With the RAID-0 array, you lose two.
~KS
sparky123321 - Friday, July 2, 2004 - link
I keep hearing about double the cost and the additional risk associated with a RAID-0 array.First off, double the price gets you DOUBLE the capacity of a single drive. It's a wash price wise. On top of that, you increase disk performance by up to 20+%. Normally, there tends to be a decrease in performance as capacity increases when comparing similar generation drives.
Secondly, with regard to risk. What's the difference between losing one 74gig Raptor in RAID-0 array or one 160gig stand-alone drive? THERE IS NO DIFFERENCE! If you don't have a recent backup, you've lost everything. Just spend an additional $90 to buy a backup 160gig 7200rpm IDE and use Acronis to do a complete disk mirror every week or two. If you lose a RAID-0 drive, you can just boot off of the backup drive and be up and running in a matter of minutes. Worst case, you've lost you're most recent work only.
Power to the Raptors and I think I'll stick with my RAID-0 array!!!!!!!
abocz - Friday, July 2, 2004 - link
I think #63 summed it up pretty well. For most real world usage the RAID0 setup never gets to shine because of the ratio of seek times/data transfers. Lower (7200) RPM drives will only compound the situation since their seek times are worse. Finally, add to this phenomenon the fact that the ratio of seeks will increase over time as fragmentation increases.Which begs the question of how well defraggers work in a RAID 0 setup? Anybody know?
Inferno - Friday, July 2, 2004 - link
Maybe you should try this test with some lesser drive. The raptors are kind of the be all end all drives. Maybe a pair of midrange Maxtors.binger - Friday, July 2, 2004 - link
#61, jvrobert is right in saying that the advantages of a raptor-raid0 are restricted to faster boot-up times and smoother handling of very large files, eg when it comes to dv.although i have heard similar statements before, whether a raid0-array of two 160gb samsung drives comes close to the performance of a single 74gb raptor drive, i don't know - i would surely appreciate being pointed to an appropriate review or at least a couple of significant benchmark results.
jvrobert - Friday, July 2, 2004 - link
Two points:First, it doesn't matter much what card you use for RAID 0. There's no parity calculation, so onboard hardware won't help much. Probably Windows striping, Intel RAID, VIA RAID, Highpoint RAID, etc.. are within 1 percent of eachother.
Second, this is a limited test that comes to an overgeneralized conclusion. As some have mentioned - these are raptors. I have a single raptor as my OS disk. Where RAID helps is with slower drives - you can get a "virtual" raptor of e.g. 360GB by buying 2 cheap, quiet, cool Samsung spinpoints.
Third (OK, 3 points) - it only tests games (which don't use much hard disk IO) and business (again, disk speed doesn't matter much). I'm getting into video now, and RAID 0 will certainly improve performance there. It will also help with load times of the OS and of large applications.
So the article comes to an over-general conclusion limited on a few quick tests.