Sign in to follow this  
Davin

Western Digital Raptor Preview

Recommended Posts

Most people don't have a SATA controller now either. And for those of us using SCSI now, it's an expense that has already been covered. If someone released an ATA drive that could compete with SCSI, I would consider dumping SCSI, but this drive certainly isn't it.

As people have already mentioned, this drives is not really marketted at people who already have SCSI. It is not a replacement for SCSI, merely something which might bridge the vast performance gap between SCSI and 7200RPM IDE.

8-ball

Share this post


Link to post
Share on other sites

I have a question, why is it that the larger drives are scoring better marks? It seems to me that if you fill up these larger drives it will take longer to access the information stored on them. I have WD1200Jb drive with XP on it and it takes several hours to Speeddisk that drive, extremely slow and poor performance if you ask me. On my other drives, Quantum Atlas 10K (Ultra 160), Fujitsu's MAN3367 (Ultra 160) and Maxtor's 30 GB (ATA100) drive take no time at all to defrag the hard drives. Are the results better on those drives because they are smaller or is it that WINME runs Speeddisk faster?

I plan to buy the Rapter for the following reasons, my current HD is garbage (120Gb is too big, too slow), the price/performance should be better than SCSI drives, the Rapter scores better performance in what really matters, transfer rate, average read service time, bootup and server performance. I would like to know on what sytem this benchmark took place? According to THG, Serial ATA drives score better performance on the KT400 chipset motherboards than any other.

I don't care if the Rapter fails to perform as fast as the high end SCSI drives, it is faster than any current ATA 133 drives and the price should be just right. I would love to go out and buy the latest server board and put an Ultra 320 drive on it but the costs are far more than I really care to spend on it. I upgrade my system every six months and all I want to do is play games (really fast). Besides, Plextor hasn't improved their line of SCSI CDRoms for a while.

Share this post


Link to post
Share on other sites
Most people don't have a SATA controller now either. And for those of us using SCSI now, it's an expense that has already been covered. If someone released an ATA drive that could compete with SCSI, I would consider dumping SCSI, but this drive certainly isn't it.

As people have already mentioned, this drives is not really marketted at people who already have SCSI. It is not a replacement for SCSI, merely something which might bridge the vast performance gap between SCSI and 7200RPM IDE.

8-ball

Once again, how many servers have onboard SATA? You're going to end up buying a controller card anyway. If you have to put together a new system just to get onboard SATA, then upgrading to SCSI would be cheaper. The market for this drive is very small.

Share this post


Link to post
Share on other sites
Most people don't have a SATA controller now either. And for those of us using SCSI now, it's an expense that has already been covered. If someone released an ATA drive that could compete with SCSI, I would consider dumping SCSI, but this drive certainly isn't it.

As people have already mentioned, this drives is not really marketted at people who already have SCSI. It is not a replacement for SCSI, merely something which might bridge the vast performance gap between SCSI and 7200RPM IDE.

8-ball

Once again, how many servers have onboard SATA? You're going to end up buying a controller card anyway. If you have to put together a new system just to get onboard SATA, then upgrading to SCSI would be cheaper. The market for this drive is very small.

SATA adapters are cheaper than SCSI adapters. And the prices will drop even more when SATA becomes mainstream. Also SATA RAID adapters are cheaper than SCSI RAID adapters (I mean basic RAID0 or RAID1). And future motherboards will have native SATA.

Cheers,

Jan

Share this post


Link to post
Share on other sites

A 4port Promise SATA is $55. A dual channel U2W controller with cables and terminators is $79. An OEM Promise ATA133 controller is $36. Either way, that's nothing most companies would lose sleep over.

"And future motherboards will have native SATA."

Upgrading motherboards is no cheaper than upgrading to SCSI.

Share this post


Link to post
Share on other sites

balding_ape and Eugene,

Yes, the 10% rule is not so much a rule as it is a nice even number that sounds good and is close enough to illustrate a point :) Of course it cannot be applied universally to all benchmarks. And of course it is not exactly 10%. I mentioned that this was found using IOMeter testing and that it is likely closer to "the 7% rule" in the desktop benchmarks, a generalization that Eugene tentatively agrees with... But I made all of these disclaimers up front in my original post and asked you not to focus on the accuracy, but on the central message:

"Now this 10% rule will probably receive a lot of scrutiny -- and rightfully so -- but focus not on the accuracy of the 10%... maybe it's closer to 7%... whatever. My contention is that if the Raptor had capacity anywhere close to that of its competitors in a capacity dependent benchmark, it would had put forth a very competitive showing. Pound for pound, gigabyte for gigabyte -- WD should be proud of their midget on the desktop."

----------

The current SR testbed does not have any drive with multiple capacities, but the old one had a 45GB and 75GB 75GXP. If you look at the I/O meter results, the 45GB version swept the 3 indicies. So adding platters does not increase the server performance which the drive is supposed to be targetted at.

King Gremlin,

1. You chose one of the few drives that was clearly designed to (and advertised to) have significantly different actuator seek performance from the factory between capacities -- again, a factor I clearly warned that would invalidate such comparisons in affected drive families:

Extra platters and heads make the head stack and actuator heavier, sometimes penalizing larger capacity drives with slower seek performance (and giving smaller drives with fewer platters/heads an edge). So, it is not as clear cut as I originally made it out to be -- this 10% rule only holds if seek performance does not vary by platter count in the same drive family.

IBM's 60 and 75 GB 75GXP models were spec'd to seek slower due to that 4-5 platter and 8-10 head stack and actuator assembly. That's a lot of mass flying around. No wonder IBM couldn't make it seek as quickly as the 15-45 GB 1-3 platter models. So, the results are not surprising in the least.

2. Server performance tests run in IOMeter are capacity INDEPENDENT. Again, I covered this with a disclaimer in my original post:

Now, as for the lackluster showing in the server benchmarks... as I understand it, IOMeter results are tested on full unpartitioned drives (thereby rendering capacity insignificant), so the Raptor's poor showing there cannot be excused.

Share this post


Link to post
Share on other sites
A dual channel U2W controller with cables and terminators is $79.

Well, it's not really fair to compare the price of an 80MB/sec device with SATA. A 160MB/sec SCSI controller would be a better comparison.

Anyway, I think the target market will be for machines with SATA RAID controllers in a configuration that uses between 2 and 8 raptors per machine.

Sure, SATA RAID is rare now, but it'll be mainstream soon enough. I don't think WD is thinking this drive will be used to upgrade current systems.

If I were WD, I would target the OEM industry, not aftermarket enthusiasts. Think how many they could sell under a contract with Dell? Dell (like anyone else) wants to sell servers that are cheap, fast, and very reliable. If WD can provide the drives that meet that criteria, they'll sell a boatload. Given these drives are faster than ATA, cheaper than SCSI, and (by virtue of their 5yr warantee and 1.2 million hour MTBF) very reliable, I think they're on the right track.

Share this post


Link to post
Share on other sites

[q]1. You chose one of the few drives that was clearly designed to (and advertised to) have significantly different actuator seek performance from the factory between capacities -- again, a factor I clearly warned that would invalidate such comparisons in affected drive families:[/q]

As I said, those were chosen as they were the only example in the SR database. That excuse doesn't apply to the DM+9 drives though, what's the explanation there? I don't disagree with your premise, but there are too many loopholes and too many examples of it not working to try and pin some blanket rule like you are trying to apply.

[q]Well, it's not really fair to compare the price of an 80MB/sec device with SATA. A 160MB/sec SCSI controller would be a better comparison.[/q]

First of all, it's dual channel SCSI card, secondly, the one drive per cable limit for SATA completely negates the bandwidth advantage since we are dealing with 60MB/s drives here. SATA could have 4GB/s of bandwidth and it wouldn't make a difference.

Share this post


Link to post
Share on other sites

whoops, keep forgetting the quoting on this board is different than boards I'm normally on which also have an edit feature so I never have to use a preview button.

Share this post


Link to post
Share on other sites
As I said, those were chosen as they were the only example in the SR database.  That excuse doesn't apply to the DM+9 drives though, what's the explanation there?  I don't disagree with your premise, but there are too many loopholes and too many examples of it not working to try and pin some blanket rule like you are trying to apply.

Well, the DM+ 9 drives are a strange breed. Their polymorphic platter configurations and refusal to adhere to any set of specifications makes them an inconsistent source of data for any tester. I'll guess that the different capacities may be using different firmware versions and this is the source of the 60 GB/platter's surprisingly strong showing relative to the other two.

I'll point you to some other drives that work... WD's series of drives. The WD800JB and WD1200JB are the same family and show a nice trend with capacity. The IBM 120GXP, while a different gen than the 60GXP, shows a higher effect on capacity (would be expected considering the higher areal density in the subsequent gen). Atlas 10k3 to Atlas 10kIV. Same situation as the GXP's. Newer gen with higher D platters skews results, but you can say capacity plays a role in the performance increase.

Yes, the exact numbers will vary depending on the situation, but the overall premise has been around for years. It is a known and accepted fact that capacity affects performance, and a long accepted practice as well. Short-stroking server drives has been done by more than a few server admins over the years to increase performance.

Share this post


Link to post
Share on other sites
but remember that areal density does do a lot too.  WDC does not understand how to make 10,000 rom drives.  They got out of that business a long time ago.  I believe that the companies that still produce high rpm drives may make a better after many trial and error models.

36gb/platters is the norm for the current generation of 10k drives...

That's right, but we're up to 83gb per platter on 7200 rpm drives - that damn near alleviates (sp?) the issue as can be clearly seen from the benchmarks.

Obviously the ideal situation is a single 83gb platter (although they need it slightly smaller due to 10,000 rpm, so perhaps 60gb per platter?) @ 10,000 rpm.

I must say the noise (whine) annoys me - I can't STAND that whine! early 7200's had it also. - but the seeks being quieter than the 200gb model is impressive as was the heat.

All in all, it's a disapointment, the least it could have done was beat all other ide drives outright, then it would be a sure fire purchase for at least SOME enthusiasts.

Sorry WD - go back and do one with a much higher arial density (sp?)

I hope this "failure" doesn't turn them off though.

As for people nitpicking my post and saying "hey firmware might improve it" - sure if so, great - however I'm far far more skeptical the firmware can do much for it - it seems to be lacking in pure data xfer speeds - I'd say that's not a firmware issue.

You're forgetting, 10k RPM drives use smaller platters, therefore, they can store less data. I know someone on here did the math once upon a time, and determined that moving from a 3.5" platter (5400 and 7200 RPM) to a 3" platter (10k,) you lose more than 50% of the capacity. So 36GB/platter at 3" is about in line with the same density as an 80GB/platter 3.5" disk. Remember, it's not the amount of data per platter that determines speed, it's how densely it's packed. That's why the "small" 15k drives are so fast. They are packed just as dense as a large 7200RPM drive, they just have a significantly smaller platter.

Ignoring the spindle hole, a 3.5" disk has about 38.5 square inches of space per side, a 3" disk only has 28.3 square inches, and a 2.5" disk only has 19.6 square inches. (Area is pi * r ^ 2, right?)

As for the firmware, just the fact that ATA 7200 RPM drives can beat 10k RPM SCSI drives shows that firmware makes a difference. (Even the 'original' special edition WD with only a 40GB/platter design can best some 10k RPM SCSI drives at 36GB/platter.)

Share this post


Link to post
Share on other sites
Obviously the ideal situation is a single 83gb platter (although they need it slightly smaller due to 10,000 rpm, so perhaps 60gb per platter?) @ 10,000 rpm.

obviously there are also technical reasons as to why the areal densities on 10k drives are the way they are.

WD did a darned good job, IMO in terms of areal density on the raptor.

WTF are you talking about, 10,000 rpm disks have been @ 36gb per platter for ages, this like Ford implimenting a V8 motor when the others have already done it!?!?!?

Share this post


Link to post
Share on other sites
You're forgetting, 10k RPM drives use smaller platters, therefore, they can store less data.  I know someone on here did the math once upon a time, and determined that moving from a 3.5" platter (5400 and 7200 RPM) to a 3" platter (10k,) you lose more than 50% of the capacity

Less, I can beleive, 50% I can not

So 36GB/platter at 3" is about in line with the same density as an 80GB/platter 3.5" disk.  Remember, it's not the amount of data per platter that determines speed, it's how densely it's packed.  That's why the "small" 15k drives are so fast.  They are packed just as dense as a large 7200RPM drive, they just have a significantly smaller platter.

That's fine I can beleive that.

Ignoring the spindle hole, a 3.5" disk has about 38.5 square inches of space per side, a 3" disk only has 28.3 square inches, and a 2.5" disk only has 19.6 square inches.  (Area is pi * r ^ 2, right?)

As for the firmware, just the fact that ATA 7200 RPM drives can beat 10k RPM SCSI drives shows that firmware makes a difference.  (Even the 'original' special edition WD with only a 40GB/platter design can best some 10k RPM SCSI drives at 36GB/platter.)

Well as far as I'm concerned we have 36gb per platter 10,000 rpm disks NOW and you know what? We've had them for a while.

therefore this is nothing new, nothing new at all.

It's a 10,000rpm model with a SATA controller strapped on.

Now,.... lets dismiss the technical arguments, because if I do make a mistake you will all jump on it.

Let's go beyond the nit picking and the "who has the biggest brain" competition lets get to the real substance of my comment which makes it a disapointment.

This drive does NOT beat ALL other ATA drives currently released in all benchmarks.

Nor does this drive beat ALL other 7200 rpm drives for that matter.

To me, that's a failure. or at least a BIG disapointment.

Sure it might be good at some stuff.

But put your hand up anyone in this thread who thought when they read about this disk, that it would be THE fastest ATA style disk ever made in everything,.. cmon be honest... put your hand up if you thought that.

I think anyone who denies that would be kidding themselves.

Hence, I set myself an expectation, sure it might not be realistic? - perhaps so, however I personally was expecting more - logically it SHOULD be the best ATA drive bar none - it's not and hence, I'm one sad panda :(

sorry WD, not for me.

Share this post


Link to post
Share on other sites
Guest Eugene

An intellectual exercise for everyone:

If SR posts substantially better performance figures for a raptor, what would be the better explanation?

a) the beta drive's performance didn't have a lot to do with production drive performance

B) WD tuned the final drives performance to better perform on SR's tests

Share this post


Link to post
Share on other sites
An intellectual exercise for everyone:

If SR posts substantially better performance figures for a raptor, what would be the better explanation?

a) the beta drive's performance didn't have a lot to do with production drive performance

B) WD tuned the final drives performance to better perform on SR's tests

I take option c ) Eugene caves into 10,000$ of pressure on the 10,000 rpm review! ;)

Seriously though Eugene I (think) I speak for everyone here when I say the whole reason we come here is the quality of the reviews.

Your review criteria is FAR FAR too broad for some dodgy "tricks" to speed this drive up - you are cluey enough to spot an anomalie and investigate, the reviews will definately stress the drive properly, I have NO doubt if this drive speeds up, it's due to option A rather than B.

Even if they try B, it must cause (some?) of A I'd think at a guess.

- Scott

Share this post


Link to post
Share on other sites

Wow, you are disappointed. You said it in triplicate!

This drive does NOT beat ALL other ATA drives currently released in all benchmarks.

Was it supposed to? I read the WD press release, and I don't think they ever claimed that. In fact, it said:

The WD Raptor hard drive enables storage vendors and systems builders, from large to small, to minimize their customers' storage hardware costs, while not sacrificing reliability, data integrity or performance.

See? It's all about low cost, good reliability, and good performance. Smart business reasons. They never boasted that it would beat all other ATA drives in all benchmarks.

Sorry you were disappointed. Perhaps you should just get the fastest SCSI drive available and call it good. As for me, I may not use this drive in my personal workstation (which is already high-end SCSI), but I will probably use these as a low-cost alternative to SCSI in future departmental servers (where performance is generally not a problem anyway).

Share this post


Link to post
Share on other sites

OK, I -had- to register for this one:

I -can't wait- to start using these puppies in desktops.

In our K12 environment, we're lucky to have 8 year replacement cycles on PCs. 1.2 million hours MTBF and a 5 year warranty are an absolute godsend.

Some things are just more important than whether it takes 8 or 10 seconds to load that Word document.

Peace.

Share this post


Link to post
Share on other sites
OK, I -had- to register for this one:

I -can't wait- to start using these puppies in desktops.

In our K12 environment, we're lucky to have 8 year replacement cycles on PCs.  1.2 million hours MTBF and a 5 year warranty are an absolute godsend.

Some things are just more important than whether it takes 8 or 10 seconds to load that Word document.

Hmm, for some reason I never thought of using this drive strictly for reliability. That's a good point you bring up.

BTW, welcome aboard!

Share this post


Link to post
Share on other sites

I personally don't go snooping around MTBF data. Is the rating for this drive really all that exciting?

Really, I don't think that this drive is going to be any more reliable than Western Digital's existing products.

Share this post


Link to post
Share on other sites
WTF are you talking about, 10,000 rpm disks have been @ 36gb per platter for ages, this like Ford implimenting a V8 motor when the others have already done it!?!?!?

what ARE YOU talking about?

the maxtor 10k iv uses 36gb disks, and has become availible just in the past 30 days. see the frontpage.

ages? i think not.

36gb is the current generation (2002-2003), 18 was the last (2001-2002), and 9 was the previous (2000-2001). 36gb/platter is CURRENT technology, and definitly not "ages" old.

what makes you think that WD, who has been out of the 10k technology for however long, can produce better 10k platters than the current 10k manufactures?

You're forgetting, 10k RPM drives use smaller platters, therefore, they can store less data. I know someone on here did the math once upon a time, and determined that moving from a 3.5" platter (5400 and 7200 RPM) to a 3" platter (10k,) you lose more than 50% of the capacity

Less, I can beleive, 50% I can not

you lose 36% just going from 3.5 disks to 3 inch disks. (not including inside radius of the spindle and the spindle speed)

Share this post


Link to post
Share on other sites
An intellectual exercise for everyone:

If SR posts substantially better performance figures for a raptor, what would be the better explanation?

a) the beta drive's performance didn't have a lot to do with production drive performance

B) WD tuned the final drives performance to better perform on SR's tests

"Better? That depends on the definition of better." -The Two Bills (Clinton and Gates)

You know, Eugene, you have mastered the art of generating forum traffic for SR by dropping little conversation topic nuggets like this one :)

My answer? Definitely both. You can't have one without the other. Option B, while more sensationalistic , is probably common practice in the HD industry. I would imagine HD mfr's routinely use benchmarks like WinBench, IOMeter, and IPEAK to tune their firmware for desktop and server applications. I mean, how else are you going to do it? So, if they are able to improve performance from beta to production, it will be with the help of benchmarking tools like the ones you use at SR. It is only logical that the drive will perform better on SR benchmarks the 2nd time around.

Share this post


Link to post
Share on other sites
If SR posts substantially better performance figures for a raptor, what would be the better explanation?

Are you suggesting something here, Eugene? :) Perhaps you have a new sample from WD already?

Share this post


Link to post
Share on other sites
Guest Eugene
Are you suggesting something here, Eugene? :) Perhaps you have a new sample from WD already?

Well, to be perfectly honest, one of the main reasons we were hesitant to publish was the somewhat stubborn character we see from some readers. Many, for example, already seem to have passed final judgment on the Raptor and dismissed it due to this article. We feared that some readers would refuse to accept substantially improved figures as representative of the family's performance because they'd fallaciously assume that WD's final was merely a reaction to our beta figures.

We went ahead and published, however, because we got wind that there were several other sites that were going to publish reviews. In addition to all the petty matters of pride, etc etc, we were concerned that other sites might not give the drive a fair shake with some rather limited benchmark suites and thus wanted to ensure that our thorough take was out there first.

That said, Anand has published his review (without a p- in front of that). He did a better job than I would have hoped, although IPEAK SPT (which he implies he's used) can be a tricky beast and not properly record/playback only what one intends unless one is very careful and thorough. Unfortunately he hasn't disclosed any AnalyzeTrace results for his traces and my ability to scrutinize them is pretty limited.

The Raptor fared a bit worse in his tests than ours. He used an SI 3112 -based controller, one that I believe has some issues when paired with these beta Raptors. That's why we went ahead with the Promise controller despite strong pushes from Seagate and HyperMicro to standardize TB3's SATA extensions on the SI chipset.

Regards,

Eugene

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this