Sign in with
Sign up | Sign in

Intel X25-M Mainstream Flash SSD (80 GB)

The SSD Workload Performance Analysis
By

Throughput Testing

Remember the test procedure: the first three runs consist of throughput testing, followed by I/O testing (you'll find the I/O results below), while the last three runs were only about throughput. Clearly, the minimum read throughput crashed from 224.9 MB/s to as little as 42.8 MB/s when we executed the IOMeter benchmark on the X25-M.

On the third cycle, the performance dropped even more, and it took until the 6th cycle for throughput to recover. Clearly the X25-M with the initial 8160 firmware takes quite a performance hit. Throughput may still reach the maximum, but it crashed to only 33.9 MB/s minimum during a h2benchw throughput test.

While read performance was still okay due to constant high maximum and average results, the sequential write performance results were really horrible. When we only execute one full IOMeter cycle, the sequential write throughput drops from 65-80 MB/s to only 3.1-3.8 MB/s. Although the maximum write throughput recovers a bit, the minimum numbers remain at an extremely low level of only 2.0-3.8 MB/s, which is far less than a conventional hard drive can provide as minimum. Even the average results are disappointing.

At this point we have to mention Intel’s 8820 firmware update, which introduces a significant performance over these results.

I/O Testing

While the throughput numbers suffer significantly with every I/O benchmark cycle, the I/O performance remains at a rather high level in the case of the database benchmark. Still, there is a noticeable drop that recovers a bit in the third run.

The results are different for the fileserver benchmark pattern, which reads and writes much larger blocks than the database benchmark. Here, I/O performance increases from 605 to 1,763 I/O operations per second after the first run, and reaches 915 I/O operations per second in the third run. Clearly, the intensive throughput benchmarking seems to have a positive effect on this I/O benchmark pattern. Still, a performance increase of 2.9x followed by a 50% drop isn’t really what you would consider ideal.

Web server performance doesn’t suffer much from our torture test cycles, as it is based 100% on reading small random blocks, which flash SSDs can handle very well at all times. Still, I/O performance decreases from 11,154 to 9,000 I/O operations per second, which represents a 20% drop.

Workstation performance also decreases, but not dramatically. Let’s look at the results with Intel’s new 8820 firmware now.

Ask a Category Expert

Create a new thread in the Reviews comments forum about this subject

Example: Notebook, Android, SSD hard drive

Display all 23 comments.
This thread is closed for comments
  • 1 Hide
    mrubermonkey , April 27, 2009 7:16 AM
    Props to you guys for this review.
  • 3 Hide
    Tindytim , April 27, 2009 7:47 AM
    I really wish we were able to Thumb up or Thumb down articles. This one would get a large thumbs up from me.
  • 1 Hide
    chookman , April 27, 2009 7:58 AM
    Thumbs up on this one too from me. Although if anyone desires more in depth understanding of the problem i found the PC Perspective articles better in that aspect.

    Intel Firmware change
    http://www.pcper.com/article.php?aid=691

    OCZ Indilinx Chip
    http://www.pcper.com/article.php?aid=670
  • -9 Hide
    chookman , April 27, 2009 7:58 AM
    Thumbs up on this one too from me. Although if anyone desires more in depth understanding of the problem i found the PC Perspective articles better in that aspect.

    Intel Firmware change
    http://www.pcper.com/article.php?aid=691

    OCZ Indilinx Chip
    http://www.pcper.com/article.php?aid=670
  • -8 Hide
    chookman , April 27, 2009 7:59 AM
    Thumbs up on this one too from me. Although if anyone desires more in depth understanding of the problem i found the PC Perspective articles better in that aspect.

    Intel Firmware change
    http://www.pcper.com/article.php?aid=691

    OCZ Indilinx Chip
    http://www.pcper.com/article.php?aid=670
  • 5 Hide
    empstar , April 27, 2009 8:55 AM
    why don't show the HDD read / write data here together with SDD?
  • -4 Hide
    lire210 , April 27, 2009 10:56 AM
    this will not matter at all to servers until the price gooes down. the speed does not make sence in the 4+ bucks a gig. normal hd are safer just in the fact that you can put 7 hd and put it in a raid one fails o well. plus 800 bucks for 64gb o my god
  • 4 Hide
    snotling , April 27, 2009 11:29 AM
    lire210this will not matter at all to servers until the price gooes down. the speed does not make sence in the 4+ bucks a gig. normal hd are safer just in the fact that you can put 7 hd and put it in a raid one fails o well. plus 800 bucks for 64gb o my god

    Servers are actually where these drives makes the most sense, and you can RAID SSDs just like you RAID HDs... and about being "safe" well I would put my money on the technology with no moving parts!
  • 6 Hide
    xsamitt , April 27, 2009 12:27 PM
    I keep waiting for a reviews of new monitors.it seems we keep getting almost the same kind of topic every week lately.
  • 0 Hide
    rubix_1011 , April 27, 2009 1:37 PM
    I have to agree with most people on this one...a flashback to the really good reviews Tom's did in the past. A very relevant review for most people with some good data and testing steps.
  • 0 Hide
    arkadi , April 27, 2009 1:39 PM
    Well it like it was written specially for me. Grate job! One of the best articles i encounter in long time.
    I posted a question on storage configuration just few day ago, with no reply btw, with exactly the same thing on my mind. Hope we can go deeper with that, and talk how we can take an advantage of what we learned here in different situations with diffident storage/system configuration's, combining SSD with regular drives, raid configurations, moving tmp/swap some programs etc.. to different drives etc...
  • -4 Hide
    arkadi , April 27, 2009 1:40 PM
    Well it like it was written specially for me. Grate job! One of the best articles i encounter in long time.
    I posted a question on storage configuration just few day ago, with no reply btw, with exactly the same thing on my mind. Hope we can go deeper with that, and talk how we can take an advantage of what we learned here in different situations with diffident storage/system configuration's, combining SSD with regular drives, raid configurations, moving tmp/swap some programs etc.. to different drives etc...
  • 1 Hide
    tipoo , April 27, 2009 1:45 PM
    Nice review. So will you guys be reviewing the 1TB PCI-Express OCZ SSD? :-)
  • 0 Hide
    Anonymous , April 27, 2009 2:50 PM
    So if your SSD starts to slow down, does a reformat get rid of that issue?
  • 0 Hide
    yourhighness , April 27, 2009 4:22 PM
    You guys had any problems with that Power Supply?

    I bought an OCZ Elite Extreme 800W PSU last year an dhad 3 of them die in 6 months....The RMA process was long and slow, but they eventually made it right by upgrading me to a PC Power and cooling PSU.
  • 2 Hide
    Area51 , April 27, 2009 4:49 PM
    lire210this will not matter at all to servers until the price gooes down. the speed does not make sence in the 4+ bucks a gig. normal hd are safer just in the fact that you can put 7 hd and put it in a raid one fails o well. plus 800 bucks for 64gb o my god

    You need to have a better understanding of the problem before you make statements like this...

    1. The number of HD required to get the same IOPS will put the total price to be higher than SSD's
    2. The power requirement is much higher for the number of HD that you need to get the same performance as a single Intel SLC SSD. This can drop the power consumption of a server significantly down from regular HD, and increase it’s performance.
    3. Remember that in many servers it’s not the capacity that matters, It’s the performance that is more important.
    4. Since SSD's can do a very nice job of simultaneous read, and write running multiple jobs is not derogated by the SSD's as it is with regular HD's.
    5. SMART command in SSD’s can give you a predictive failure analysis, something that you cannot do with HD’s, HD’s can only show you if the Drive is good or bad. This is very important since data can be copied before the drive goes bad.
    6. MTBF of SSD’s (at lease Intel’s) are much higher than HD’s. 2M hrs. vs 1.2M hrs on HD’s


  • 2 Hide
    krazyderek , April 27, 2009 11:06 PM
    wheres the OCZ vertex ?????
  • 0 Hide
    krazyderek , April 27, 2009 11:08 PM
    lire210this will not matter at all to servers until the price gooes down. the speed does not make sence in the 4+ bucks a gig. normal hd are safer just in the fact that you can put 7 hd and put it in a raid one fails o well. plus 800 bucks for 64gb o my god

    have you seen how much SAS drives are??
  • 0 Hide
    hustler539 , April 28, 2009 11:07 AM
    Good to see more review sites doing articles on these performance degradations, helps us make a better informed decision when purchasing new hardware.

    Here's another great article which gives you a good in depth look at it
    http://www.anandtech.com/storage/showdoc.aspx?i=3531&p=4
    a must read if you are thinking of upgrading to some of those new shiny ssds
  • -1 Hide
    hustler539 , April 28, 2009 11:08 AM
    krazyderekwheres the OCZ vertex ?????



    ^ a review of that is in the article as well
Display more comments