Hamburglar

22k Rpm Sound Good?

Recommended Posts

New higher speed drives in the works. They aren't coming soon, so cool off, hotrod. They'll probably come with serial SCSI, as well as fibre channel and legacy SCSI.

I got this info from an unnamed source at EMC. Sure it's not specific, but it's enough to get the speculation mill a-turnin.

Talk amongst yourselves.

Share this post


Link to post
Share on other sites

22k sure is a big step. If I was a bettin man, I would've counted on 18k - 20k as the next progression. But hey, 22k is even better. I still think this is probably 24 months away though :(

Share this post


Link to post
Share on other sites

22KRpm is coming but it will only be available in the new 2.5 inch enterprise form factor as the platter diameter needs to be quite small so packing in a 3.5" form factor doesn't make sense.

SG

Share this post


Link to post
Share on other sites

I second Eugene. The advantage going from 15K to 22K imparts to the rotational latency of a disk is minimal. I don't think a lot of thought has gone into your excitement about these 'on their way' disks.

You only shave 0.6 ms off the average rotational latency. That's only 60% of the gain achieved from going from 10K to 15K. Compare the performance differences between the best 10K drives and the best 15K drives. In IOMeter where the only advantage is derived, it's only 36%. Remember that a good portion of that is due to seeking advantages not latency advantages. Now contrast this performance boost with the usual performance increase between generations within a drive family. Great...

Just trying to be sensible.

SAN_Guy,

22KRpm is coming but it will only be available in the new 2.5 inch enterprise form factor as the platter diameter needs to be quite small so packing in a 3.5" form factor doesn't make sense.

Of course. 15000 RPM drives are already confined to 2.5" platters. There's nothing new here. As for "packing in a 3.5" form factor doesn't makes sense," they'll pack it in a 3.5" form factor if it is still a standard, just as they did for 15000 RPM drives. As for 2.5" if it becomes a standard, as I hope it does, then, of course, we'll se drives in that form factor as well.

Do well.

Jonathan Guilbault.

Share this post


Link to post
Share on other sites
22k sure is a big step. If I was a bettin man, I would've counted on 18k - 20k as the next progression.

Well, all the speed bumps have been ~50%.

5400<7200<10000<15000<22000

So it makes sense.

The speed improvement of 22k over 15k will be comparable to the improvement of 15k over 10k. I wouldn't classify this with the trivial boost from P4 3.2 to 3.4.

Certainly, upcoming cutting edge HDs will all have 2.5" platters. I haven't heard any talk of a new enterprise form factor, so I doubt there will be a move away from 3.5" hh. Then again any 22k drive is at least a couple years away... who knows.

Share this post


Link to post
Share on other sites

Avg rotational latency of 15K drive: 2 ms

Avg rotational latency of infinitely fast drive: 0 ms

Affect that 15K to 22K will have: trivial

Share this post


Link to post
Share on other sites
The speed improvement of 22k over 15k will be comparable to the improvement of 15k over 10k. I wouldn't classify this with the trivial boost from P4 3.2 to 3.4.

Unless they improve more than just rotational latency over the existing 15K RPM drives, I would guess tha the improvement in performance will in fact be trivial. As Gilbo has noted, less than a millisecond is saved. That's close to nothing, an amount easily overcome by cache algorithms or other mechanical/electronic considerations. Just look at the WD Raptor, and how well it competes against 15K RPM drives at just 10K.

Share this post


Link to post
Share on other sites

Odds are, along with the rotational increase, the platters will be shrunken again which will decrease average seek as well. If areal density drops proportionally to what we have seen with other rotational increases, STR's will be record setting, and the decay will be pretty low, with very impressive minimal transfer rates from these drives. A 2" 22k RPM drive will almost eliminate the difference between beginning and end of platter performance differences that we are all accustomed to.

Share this post


Link to post
Share on other sites
Odds are, along with the rotational increase, the platters will be shrunken again which will decrease average seek as well.  If areal density drops proportionally to what we have seen with other rotational increases, STR's will be record setting, and the decay will be pretty low, with very impressive minimal transfer rates from these drives.  A 2" 22k RPM drive will almost eliminate the difference between beginning and end of platter performance differences that we are all accustomed to.

The gain in access time due to the size of the platter will be partially offset by the fact that the drives will hold significantly less data -- half as much per platter, if we're lucky. And STR isn't all that significant anymore in the majority of applications. I think all these changes add up to, at best, a small incremental gain.

Share this post


Link to post
Share on other sites

Sure, if you automatically assume worst case scenario for all performance factors like you are the difference may be minimal. Reality tells the rest of us that that isn't likely to happen. When comparing 10k and 15k drives from the same generation and company, for example Maxtor, the difference is over 20% in the workstation benchmarks, and just under 30% for the server benchmarks. That's not minimal to me, and if those percentages can be duplicated, and there's little reason to believe they can't (same 50% rotational increase), then that's plenty for most people to buy into it.

Share this post


Link to post
Share on other sites

Faster is good, but denser is better.

There's a lot more performance to be gained by packing the data in than spinning it faster.

Share this post


Link to post
Share on other sites

0ms latency is only 4ms faster than 4ms latency. 4ms is almost no time at all.

Share this post


Link to post
Share on other sites
Sure, if you automatically assume worst case scenario for all performance factors like you are the difference may be minimal.  Reality tells the rest of us that that isn't likely to happen.  When comparing 10k and 15k drives from the same generation and company, for example Maxtor, the difference is over 20% in the workstation benchmarks, and just under 30% for the server benchmarks.  That's not minimal to me, and if those percentages can be duplicated, and there's little reason to believe they can't (same 50% rotational increase),  then that's plenty for most people to buy into it.

You're right to say that I chose areas that would prove my case -- as you said "worst case scenarios." I would prefer to think of it as "typical SR user patterns."

Look, my point is that I don't think that 22K will bring any more increase in performance than a typical new generation of drive does, except perhaps in disk-intensive server situations (databases, for example). For all my tasks (which include gaming, Photoshop Elements, and then just typical "office" apps), I can't tell the difference between my X15-36LP and my Fujitsu MAJ drives.

Even in disk-intensive apps, you will not see a great a difference as between 10K and 15K drives, because the overall reductions are lesser. The fastest 10K drive's access time is 7.6ms, and the fastest 15K is 5.6ms. If we assume a fairly similar scale-down (after accounting for rotational latency), the fastest 22K drive will be about 4.2ms access. 7.6 to 5.6 is about 23% reduction, where going to 4.2 is only an additional 18% from the 7.6 baseline (said another way, you lose 2ms from 10K to 15K, where you lose 1.4ms from 15K to 22K). The higher you push the RPMs, the less impact increased RPMs have on performance. However, the impact of increasing capacity remains the same, assuming the capacity continues apace (not a particularly good assumption now, I'll grant).

How much that translates into performance will depend entirely on the application, of course.

Share this post


Link to post
Share on other sites

KingGremlin,

Odds are, along with the rotational increase, the platters will be shrunken again which will decrease average seek as well. If areal density drops proportionally to what we have seen with other rotational increases, STR's will be record setting, and the decay will be pretty low, with very impressive minimal transfer rates from these drives. A 2" 22k RPM drive will almost eliminate the difference between beginning and end of platter performance differences that we are all accustomed to.

None of these advantages have anything to do with moving to a 22K spindle speed. Your argument is silly --15K drives with identical platters will perform nearly identically to this 22K drive you describe. So what's the point? We have reached the point of diminishing returns.

the difference is over 20% in the workstation benchmarks, and just under 30% for the server benchmarks. That's not minimal to me, and if those percentages can be duplicated, and there's little reason to believe they can't (same 50% rotational increase), then that's plenty for most people to buy into it.

You haven't thought this out at all.

The differences between the fastest 10K and 15K drives in non-server benchmarks are inconsequential.

I have already discussed the advantage of 15K drives over 10K drives in server benchmarks in my previous post. I will repeat the relevant information:

1. Average Seek time is faster on 15K relative to 10K drives drives by a full ms (seek, not access). This is equal to the advantage 15K drives hold over 10K drives in rotational latency --also 1 ms. The increased spindle speed is contributing only 50% of the positional advantage 15K drives enjoy over 10K drives in server benchmarks.

2. That means the move from 10K to 15K rotational speeds offered an 18% performance boost in IOMeter. 36% total was achieved by lowering the seek times, an unrelated change. The move from 15K to 22K only offers 60% of the reduction in rotational latency that the move from 10K to 15K offered.

3. This requires an 11% performance advantage in situations which benefit the most from positional performance. 11% at a tremendous cost. 22K rotational speeds are significantly more difficult to design for relative to 15K than 15K drives were relative to 10K. Imagine the price premium.

Frankly, I doubt the usefulness of a 22K RPM disk drive and I certainly doubt its practicality. Virtually equal positional performance could be achieved at such a dramatically lower cost from simply shrinking the platters, without increasing the spindle speed, that I doubt anyone will bother to create such a drive. If a hard drive manufacturer did, competitors would be able to offer approximately equal performance at such a price advantage by using equal platters at 15K that the drive would almost certainly be a failure from a profitability stand point.

Frankly, I think this is simply one of these posts that have appeared semiregularly over the course of this forum since its birth, predicting the next speed-class soon. And like most of those posts, I think this one is pure BS. Even if its not, for the reasons I have cited, the 22K speed-class are likely to be utterly uninteresting.

I just thought:

Maybe the prohibitive cost of such drives will reduce the oppurtunity enthusiasts with too much money to burn have to create RAID 0 arrays, slowing the inundation of this forum with such posts... Now that is a tangible benefit that we can all appreciate.

Do well.

Jonathan Guilbault.

Share this post


Link to post
Share on other sites
0ms latency is only 4ms faster than 4ms latency.  4ms is almost no time at all.

When almost everything in a computer is electronic and measured in microseconds, 1 ms is an eternity. If there wasn't considerable room for performance increases in hard drives, we wouldn't have RAM to supplement it. Try running identical computers, one with a 15k drive and one with a SS drive and tell us those 4ms don't make a difference.

Share this post


Link to post
Share on other sites
None of these advantages have anything to do with moving to a 22K spindle speed. Your argument is silly --15K drives with identical platters will perform nearly identically to this 22K drive you describe. So what's the point? We have reached the point of diminishing returns.

What's silly is your assumption that spindle speed is the only thing that will change with a new generation of drives. When has that ever been the case? Do you honestly believe that hard drive engineers are going to go through all the work and research necessary for a product like this that will preform 4 or 5% faster than 15k drives?

Your diminishing returns comment is pretty stupid as well. What field of computing performance is increasing at a faster rate than it was 5 or 10 years ago? Why don't we just stop releasing new CPUs's, video cards, and optical burners because they aren't increasing in performance at the same rate they used to.

Share this post


Link to post
Share on other sites
I'm ready.

LOL. Best answer ever. ;)

Share this post


Link to post
Share on other sites

KingGremlin,

I don't think you read my post.

I said,

Virtually equal positional performance could be achieved at such a dramatically lower cost from simply shrinking the platters, without increasing the spindle speed, that I doubt anyone will bother to create such a drive. If a hard drive manufacturer did, competitors would be able to offer approximately equal performance at such a price advantage by using equal platters at 15K that the drive would almost certainly be a failure from a profitability stand point.

I think that says everything that matters.

Do well.

Jonathan Guilbault.

Share this post


Link to post
Share on other sites

Have you been following the news recently? What you said won't happen has already started to happen with Seagate and Fujitsu announcing 2.5" 10K drives and mainly promoting the faster access times the smaller drives will provide. 15k drives will naturally follow suit eventually. One of the posts above already stated that the new 22k drives will most likely be 2.5" form factor. Why just shrink the drives, when you can shrink them and speed them up? It's a no brainer.

Share this post


Link to post
Share on other sites
Why just shrink the drives, when you can shrink them and speed them up?  It's a no brainer.

return on investment? profitability? both for the manufacturer and consumer

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