Western Digital's New VelociRaptor VR200M: 10K RPM at 450GB and 600GB
by Anand Lal Shimpi on April 6, 2010 8:00 AM EST- Posted in
- Storage
Final Words
In the days before SSDs, the VelociRaptor was the drive that raised the cost per GB ceiling. These days, the 600GB drive almost seems like a bargain. Have a look:
| Cost Comparison of Modern HDDs/SSDs | |||||||
| Drive | Capacity | Price | Cost per Gigabyte | ||||
| Western Digital VelociRaptor VR200M | 600GB | $329 | $0.548 | ||||
| Western Digital VelociRaptor VR200M | 450GB | $299 | $0.664 | ||||
| Western Digital VelociRaptor VR150M | 300GB | $199 | $0.663 | ||||
| Western Digital Caviar Black WD1002FAEX | 1TB | $120 | $0.120 | ||||
| Seagate Barracuda XT ST32000641AS | 2TB | $250 | $0.125 | ||||
| Intel X25-M G2 | 160GB | $414 | $2.588 | ||||
| Intel X25-M G2 | 80GB | $220 | $2.750 | ||||
Western Digital's pricing picks up where the VR150M left off and drops the cost per gigabyte significantly for the 600GB drive. At $329 vs. $299, I'm not sure there's even a reason to consider the 450GB offering. That extra $30 buys you 150GB at $0.20 per GB. Now obviously compared to a high end 7200 RPM drive, you are paying a price premium for the VelociRaptor. Based on our tests I'd expect to see a 5 - 10% increase in overall system performance compared to a current generation, 7200RPM drive. If you have particularly random workloads, the performance gap can can grow to be something much higher in the 15 - 20% range (or beyond if you're truly I/O bound).
Our AnandTech Bench gaming workload does make the argument that if you're primarily interested in using this drive for games, you might be better served by a larger 3.5" drive. Game installs are pretty big these days and when playing games you're mostly performing sequential reads off the disk, which wastes much of the benefit of the 10K RPM spindle speed. It's only if you're planning on having other apps running in the background that hit the disk while you game that you could benefit from the VelociRaptor.
The rest of the tests make it very clear. As far as hard drives go, you can't beat the random read/write performance of the new VelociRaptor. For applications that absolutely demand to be run on a physical disk, this is your best bet.
The problem is once you take into account solid state storage. The new VelociRaptor boasts a 4KB random write speed of 1.9MB/s. Intel's X25-M G2 is amost 20x faster. The new VelociRaptor averages 178 IOPS in our typical Bench workload, Intel's X25-M can push nearly 800 IOPS in the same test.
While you are getting much more storage for your dollar with the VelociRaptor, a higher performance alternative would be to combine a good SSD with a 1TB drive. Using the SSD for your OS and apps, and the TB drive for all of your music, photos, videos and games. It's this sort of configuration that I use in my personal desktop (except I have two 1TB drives in RAID-1).
If you can't go the SSD route but still need the performance, WD has retaken the crown with the new VelociRaptor. If you can make it work however, you may be happier with an X25-M and a WD Caviar Black instead.
Facebook
Twitter
RSS
77 Comments
View All Comments
Makaveli - Tuesday, April 6, 2010 - link
I really think they should have gone with 300GB's platters dropped the 450GB model and release just a single platter 300GB's model and 2 platter 600GB model. Sell the 300GB model for $199 and the 600GB model for $299.I'm already using Intel SSD + 1TB Black for storage so I won't be buying one, raptors are dead to me!
efeman - Tuesday, April 6, 2010 - link
I have two VR150's in RAID-0. Is there any chance you could compare performance of the new VR's to that (or two VR300's, of course).Hacp - Tuesday, April 6, 2010 - link
Dear Anand,It would be helpful if you had some random latency tests, because that is what makes mechanical drives so horrible. Also, would be helpful if you did some short stroked benchmarks with this drive. Finally, I would like to compare it to an SSD drive. I know you are short on time but it would really make the review more interesting. Keep up the good work.
Anand Lal Shimpi - Tuesday, April 6, 2010 - link
The random tests actually give you latency, just represented in MB/s instead of ms.For example:
4KB random write test, 3 outstanding IOs:
VR200M got 1.9MB/s average write speed
That's 1945.6 KB/s (1.9MB/s * 1024KB/MB), which is 486.4 IOPS (1945.6KB/s / 4KB/IOP). That gives us IOs per second, or if we take the inverse we get seconds per IO: 2.05ms. Now since we've got 3 outstanding IOs that's 3 x 2.05ms or 6.16ms.
Latency is represented, just in the form of MB/s :)
Take care,
Anand
AstroGuardian - Tuesday, April 6, 2010 - link
Anand, please if you feel like telling us; you mentioned you have 1 x SSD and 2 x 1Tb drives in you personal computer in RAID1. What kind of 1Tb drives did you put there?Just a curiosity. I would like to know what your choice was...
Cheers
Anand Lal Shimpi - Tuesday, April 6, 2010 - link
I've got a pair of old Hitachi 1TB HDT721010SLA360 drives in my machine. I just used them because I had them laying around with no other purpose :)Take care,
Anand
pjconoso - Tuesday, April 6, 2010 - link
Are you actually using them? :) I could use a couple of those - 'wouldn't mind if they're slow as hell, lol.AstroGuardian - Wednesday, April 7, 2010 - link
Naaah dude, those drives are just great. We have plenty of those here in Europe. I have seen many defective Seagates, Spinpoints, few WDs but i have never seen a dead Hitachi (except for one of mine which i sent to death myself one angry morning...).Great choice Anand, but i still kind a think you chose those drives for a reason (what could that be lol) and not just because they were laying around. Haha....
strikeback03 - Thursday, April 8, 2010 - link
We have had one Hitachi, a Maxtor, a few Seagates, and a bunch of Samsung Drives die here at work.AstroGuardian - Thursday, April 8, 2010 - link
Haha, there you go. The Samsungs die every day. I have no idea what their issues are, especially those F3 - RAID Class drives. About the Maxtor i think it's time has come to pass away don't you think? Hehe