Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

Could anyone here explain to me what is implied by this term? (I've seen the same thing mentioned with the 3 terms).

At first when I read about it, for some reason I understood that it was some way of splitting the bytes across the platters of the disk, which sounded like a good idea and obviously doesn't make sense, because that wouldn't cut disk size in half (and disk are probably already splitting bytes across platters)...

The best I've come to understand is that basically instead of creating one partition for the whole size of the disk, you create 2 partitions, and use only one of them, either the one in the "center" or the one in the "rim" of the platters, and since one of the two is faster (people didn't seem to agree on which one was faster), that makes everything better.

Am I understanding this correctly?
Has anyone tried this with their drives and had a good outcome?

Thanks!

share|improve this question
1  
Not since the days of Commodore computers have I seen this strategy used. Funny how old ideas keep coming back. :) –  John Gardeniers Jul 6 '09 at 21:47
add comment

4 Answers

up vote 3 down vote accepted

Short-stroking is basically what you found. You specifically use only the last few tracks on each platter of your hard disk. I have heard of this, but haven't looked at it in a while.

Looking at new articles, as well as from memory, the details about this are a mixed bag, mainly bad from my perspective.

  • Extremely reduced capacity for the drive because short-stroking only shows a benefit with very small "drive" sizes.
  • Roughly 40% better random seek times.
  • Slightly faster bulk transfer rates.
  • Requires specialized software that doesn't seem to be widely available.
  • When used in a RAID 0 array (as most articles recommend) this uses a lot of power for relatively small drive sizes.

I have previously recommended against ideas like this, as just buying larger, faster disks is cheaper in the long run, unless you don't pay for your electricity. The time savings may help in a database server with very little memory, but I can think of no other situation.

In general reading from the outside sectors of the platters is faster, as more sectors pass under the heads per second at 7500 RPM (or whatever) than towards the middle. Also, the heads rest on the outside of the drive when resting so, making a partition only near the center of the drive could actually give you worse seek speeds.

share|improve this answer
1  
Today it is a hardly sensible approach - it was very important some years ago, but today... just get an SSD. If you short stroke you do that bcause you need IOPS more than speed, and SSD are totally killing hard discs. Double your IOP with SSD - good. 450 -> 900. SSD = 50.000+ ;) Ouch. –  TomTom Mar 29 '12 at 6:03
add comment

In the beginning each "track" of a disk platter had the same amount of 512 byte sectors - meaning the density was highest towards the center. This was as I understand it pretty early on improved by making each track have a variable amount of sectors to increase efficiency and have about the same density over the entire platter (ZBR).

Hence, the further out on the platter data lies, the faster it can be read and written to as the raw throughput will be higher.

So yes, partitioning only the outside half of a disk would definitely increase overall performance.

Is it worth it? No idea. It's usually employed for high-end 15krpm drives in critical environments. Today I'd say the modern drive controllers in these situations can handle this intelligently enough without specifically "short-stroking" drives.

I'd be curious to know if this is used "at factory" as well, like producing smaller-sized high-speed drives with perhaps more than one platter that only internally uses the most outer tracks to get a performance advantage?

share|improve this answer
add comment

Short stroking and Half stroking are both a kind of partial stroking. Basically, all three means the hard disk head does not need to move from the center to the outer-most edge (full stroke).

I've got a short stroke setup, where I have a 500GB disk, with 2 partitions. The C partition is 30GB, and the remaining space into the D letter partition.

It's obvious that it's MUCH responsive, as the head basically stays in the out-most edge, where the transfer speed is at its maximum, and the head does not need to seek much.

FYI, in the outmost edge, the data density is the highest, as one rotation have a much larger area than the inner most circle.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.