Value of hard drive rpm

[Horsepower]

It's been quite a while since I got my first raptor, and I keep noticing there are still more 7200 rpm drives available than 10000 rpm. I though the whole line would rapidly change. Can someone explain? Thank you.

 

[ch33zw1z]

Dont know exactly why, but I would guess cost. Same reason SCSI never went big for home PCs.

 

[SuperNaruto]

Most likely cost.. thats why scsi stayed expensive, they're rock solid... you need better parts to prevent failure.. if the price of making better hdd rise for home, no one is going to buy it.. if you are going to buy it.. buy scsi, why bother making another market segment

http://www.anandtech.com/IT/showdoc.aspx?i=2859&p=5


The difference in reliability between typical SATA and real enterprise disks has been proven in a recent test by Seagate. Seagate exposed three groups of 300 desktop drives to high-duty-cycle sequential and random workloads. Enterprise disks list a slightly higher or similar failure rate than desktop drives, but that does not mean they are the same. Enterprise disks are tested for heavy duty highly random workloads and desktop drives are tested with desktop workloads. Seagate's tests revealed that desktop drives failed twice as often in the sequential server tests than with normal desktop use. When running random server or transactional workloads, SATA drives failed four times as often![²] In other words, it is not wise to use SATA drives for transactional database environments; you need real SCSI/SAS enterprise disks which are made to be used for the demanding server loads.

Even the so called "Nearline" (Seagate) or "Raid Edition" (RE, Western Digital) SATA drives which are made to operate in enterprise storage racks, and which are more reliable than desktop disks, are not made for the mission critical, random transactional applications. Their MTBF (Mean Time Between Failure) is still at least 20% lower than typical enterprise disks, and they will show the similar failure rates when used with highly random server workloads as desktop drives.

 

[Lonyo]

WD don't have a SCSI line, most other hard drive manufacturers do.
As SuperNaruto says, why make another market segment? Western Digital are the only ones to do so because they don't have a SCSI presence.

 

[Horsepower]

And I thought SATA was the upcoming technology that would replace scsi.

 

[ch33zw1z]

They may take some of the market from SCSI, such as small servers, high end workstations, some others i cna't think of....but SCSI isn't going anywhere for a while. You will see more Serial Attached SCSI (SAS) products being put into high performance boxes now..

 

[Roguestar]

Originally posted by: Horsepower
And I thought SATA was the upcoming technology that would replace scsi.

???

No, SATA is the replacement for PATA. SCSI is still superior and maybe be for a long time until we start using all of the SATA specification instead of ignoring it in favour of backward compatibility.

 

[willtriv]

SATA is really nothing more than pata with a few new buzz words (ncq?!) and a smaller wire. It an improvement but is certainly isnt an u320 raid5 controller. Sas is basically the same as sata is to pata. It makes the wire smaller I just wish there were more SAS controllers for pci-e x1 slots. I hate how servers always have different slots (even if they use the same architecture now!) All controllers are x4 or x8 and I have a server board that doesnt even have either of these:S

 

[willtriv]

o and to add to the whole 10000rpm thing:
higher rpm is better, but only marginally. 7200rpm is decent for desktop use and most users won't even notice the difference. Enough home PC's have raid 0,1,0+1/10 so they could use two 7200rpm in a striping setup and achieve more performace (and space) for the same cost as 1 10000rpm disk.

 

[ch33zw1z]

Originally posted by: willtriv
SATA is really nothing more than pata with a few new buzz words (ncq?!) and a smaller wire. It an improvement but is certainly isnt an u320 raid5 controller. Sas is basically the same as sata is to pata. It makes the wire smaller I just wish there were more SAS controllers for pci-e x1 slots. I hate how servers always have different slots (even if they use the same architecture now!) All controllers are x4 or x8 and I have a server board that doesnt even have either of these:S

SATA is not PATA, the handle data differently using different protocols(mainly data is sent serial as opposed to parallel)...same with SAS. This allows higher data throughput, causing higher performance.

 

[velis]

SATA is based on PATA standards. The packets are basically the same. The v1.1 spec says on the first page that the basis for it is ATAPI v5 standard.
The first difference that it introduces is the way data is transferred through cables (serial vs. parallel). Since all of the sudden humanity can make serial transfers more efficient than parallel ones, this results in more bandwidth. The second plus that it introduces are thinner / more manageable cables and the third is a handier power connector which also provides 3.3V power to the device, although this is currently unreliable since many manufacturers simply include 4pin --> sata power connector converter, which of course does not have the 3.3 voltage. Suma summarum, the standard was pushed purely because it offers more bandwidth (and has a much higher potential to even increase the difference in the near future). Neater cables are just a bonus.

SAS uses the same electrical signal transport protocols as SATA resulting in thinner cables and more bandwidth. It's packet protocols also support SATA (among other things) so AFAIK there should be no problems plugging a SATA drive into a SAS controller.

Edit - to answer to the OP: WD used the knowledge they had from their SCSI line in an enthusiast product. These drives are rather expensive to make and the market for them is very small. I guess the other major players didn't go for it because they don't see much profit in it. I doubt Raptor drive makes much direct profit for WD either, but it certainly is a good marketing tool, if nothing else.

Rotational latency:
5400 RPM 11 ms
7200 RPM 8.3 ms
10000 RPM 6 ms

This latency is for one full revolution, so you can halve it to get an "average" time the disk takes to reach the designated sector on the target track.

As you can see, the difference between 10000RPM drive and a 7200RPM drive is only 1.2 ms. This difference can be overcome by using a faster / more precise actuator which actually even shows on the newer 7200 drives vs the original 36GB Raptor. Given a small enough partition, they tend to be even faster than the Raptor despite the 1.2 ms average headstart that the Raptor has.

 

[Horsepower]

Of course, by dedault, my cmos is not set to pure SATA. That means I'm losing possible throughput to my DVD and raptor.