Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
×
Data Storage Hardware IT

MS Researchers Call Moving Server Storage To SSDs a Bad Idea 292

An anonymous reader writes "As an IT administrator did you ever think of replacing disks by SSDs? Or using SSDs as an intermediate caching layer? A recent paper by Microsoft researchers provides detailed cost/benefit analysis for several real workloads. The conclusion is that, for a range of typical enterprise workloads, using SSDs makes no sense in the short to medium future. Their price needs to decrease by 3-3000 times for them to make sense. Note that this paper has nothing to do with laptop workloads, for which SSDs probably make more sense (due to SSDs' ruggedness)."
This discussion has been archived. No new comments can be posted.

MS Researchers Call Moving Server Storage To SSDs a Bad Idea

Comments Filter:
  • by nweaver ( 113078 ) on Wednesday April 08, 2009 @02:03PM (#27506543) Homepage

    This is an ACM article behind a paywall.

    How about a slashdot policy of not linking to articles behind paywalls?

  • by eples ( 239989 ) on Wednesday April 08, 2009 @02:04PM (#27506555)

    Their price needs to decrease by 3-3000 times for them to make sense.

    Hm. I was thinking the same thing about the ACM subscription.

  • by hxnwix ( 652290 ) on Wednesday April 08, 2009 @02:05PM (#27506567) Journal

    SSD is already cheaper per gig than some SAS drives. Also, 3-3000 times? What the hell sort of estimate is that?

  • What if... (Score:4, Insightful)

    by Thelasko ( 1196535 ) on Wednesday April 08, 2009 @02:06PM (#27506577) Journal
    they don't use NTFS?
  • by Larry Clotter ( 1527741 ) on Wednesday April 08, 2009 @02:08PM (#27506627)
    It's called "pulling numbers out of your ass".
  • 3 to 3000 percent? (Score:4, Insightful)

    by erroneus ( 253617 ) on Wednesday April 08, 2009 @02:10PM (#27506643) Homepage

    My goodness! They have really done their research in order to produce data as accurate as that!

    The fact was, they said the same thing when it came to magnetic tape versus magnetic disks. These days, hard drives are cheaper than tapes and will hold their data longer and more compatibly.

    Microsoft fears change that they do not control. If they don't control the changes, someone might write them out of the story.

  • I actually don't think times cheaper makes any sense.

    I hear it all the time, but it is meaningless.

    3000 times cheaper than what? The current price?

    If I am selling something that is now "twice as cheap" is that half the price?, double the discount?, twice as shoddily made?

  • Something's wrong (Score:3, Insightful)

    by drsmithy ( 35869 ) <drsmithy&gmail,com> on Wednesday April 08, 2009 @02:42PM (#27507139)
    They list the write IOPS of their "Enterprise SSD" drive as only ~350. That number seems like it's an order of magnitude too low, which would obviously skew the conclusions.
  • Funny, my other complaint is twice as slow.

    The problem I see is 3 times slower doesn't multiply anything by 3, it divides it.

    and intuitively slower of cheaper are not inverse, since we use statements like 5 dollars cheaper.

  • Re:I concur (Score:3, Insightful)

    by saiha ( 665337 ) on Wednesday April 08, 2009 @03:31PM (#27507971)

    It really depends on how much data you are talking about and your performance requirements. SSD gives a good medium between RAM and HD for both speed and cost.

  • by HTH NE1 ( 675604 ) on Wednesday April 08, 2009 @04:18PM (#27508817)

    They're not the only ones pulling numbers out of their ass. You also seem to be too, unless you're finding the absolutely most expensive drive in any given capacity class.

    Those were actual prices extracted from a Google Product Search. Actual prices being charged on the web. The search term was "solid-state drive". However, I did take the search results at face value; some of those drives may have come with superb service plans good for the duration of the copyright of anything recorded on them, or only sold to government contractors AFAIK.

    For example, the 80GB Intel X25-M runs around $380, so is better than any of the prices you pulled up.

    Which is still 47.5 times more expensive in GB/$ than current (albeit consumer level) 1 TB drives. So you've gotten closer to the 3-factor while I sought to find something close to the 3000-factor.

    Perhaps we have enough data to reverse engineer their bang-for-buck target value now?

  • by sexconker ( 1179573 ) on Wednesday April 08, 2009 @04:19PM (#27508825)

    And dollars is a unit, so you add and subtract.

    If you said something was 5 times something, the 5 would be a scalar (on the second something) and you would divide or multiply.

    It makes perfect sense if you just stop to think about what the words slower, faster, cheaper, etc. mean. They all measure something, find out what, and the logically appropriate operation will be obvious.

    5 times slower means something takes 5 times as long, and therefore runs at 1/5th the rate.

    5 times cheaper means something gives 5 times as much product/service/"value" for the same cost, and therefore is 1/5th the cost for the same product/service/"value".

  • Re:Read the Paper (Score:4, Insightful)

    by careysub ( 976506 ) on Wednesday April 08, 2009 @04:50PM (#27509369)

    It is informative, but there is one aspect that they omit from their analysis - the effect of device performance of the cost of the server farm needed to provide service.The whole analysis is based on the storage device cost only (there are good reasons for this, but limits the relevance of their analysis). If higher read rates of an SSD translate into higher server transaction rates then fewer servers are needed at possibily dramatic additional savings.

    Here is a specific scenario to make this concrete.

    You have a search engine application that accesses a relative static index (small parts refreshed daily maybe, all of it refreshed monthly). The ability to randomly read blocks determines how many queries per second your server can handle. The 17-fold speed advantage of the SSD over the Cheetah 15K is a huge win here. Of course you can set up a RAID 0+1 of Cheetah's but your server box only holds 4 data drives (out of 6, you mirror 2 more for redundant storage of the OS and application). So you need to buy four times as many servers using Cheetahs than SSDs, which use more than 4X the power and take up extra data center space (which is not free).

    Or you could stuff a dozen or more Cheetahs into a RAID chassis that costs several times more than one server box.

    Either way the cost of the Cheetahs themselves is trivial compared to the cost of the hardware required to actually make use of them.

  • by Sam36 ( 1065410 ) on Wednesday April 08, 2009 @04:55PM (#27509457)
    3000 Times cheaper than a peanut found from Obama's dung and sold on ebay.
  • "As an IT administrator did you ever think of replacing disks by SSDs? Or using SSDs as an intermediate caching layer?

    SSDs aren't big enough for some uses as mass storage but they could speed up things if used as a cache.

    Note that this paper has nothing to do with laptop workloads, for which SSDs probably make more sense (due to SSDs' ruggedness)."

    I think laptops are where SSDs can come into their own. There shouldn't as much need to large mass storage and SSDs extend battery life. Having said that, I replaced the 160GB HDD in my 1 1/2 year old laptop with a 320GB drive, the biggest I could find.

    Falcon

  • by afidel ( 530433 ) on Wednesday April 08, 2009 @05:46PM (#27510183)
    Today you just buy 2.5" drives with 2" platters, much more cost effective =) As an example HP 146GB 15K 2.5" has a full stoke latency of 4.85ms, nearly as fast as the average (short stroke) latency of a 144GB 10K 3.5" drive (3.9ms). No way the 2.5" cost 2x more =)

He has not acquired a fortune; the fortune has acquired him. -- Bion

Working...