IO bechmark

  • I am testing SSD's with SQLIO to find R/W performance. how we can determine whether it is good or not based on the numbers returned by SQLIO.

  • I just started working with SQLIO in the past couple months. I am no expert but found this article to be very helpful:

    http://blogs.technet.com/b/josebda/archive/2013/03/28/sqlio-powershell-and-storage-performance-measuring-iops-throughput-and-latency-for-both-local-disks-and-smb-file-shares.aspx

    "I cant stress enough the importance of switching from a sequential files mindset to set-based thinking. After you make the switch, you can spend your time tuning and optimizing your queries instead of maintaining lengthy, poor-performing code."

    -- Itzik Ben-Gan 2001

  • thanks Alan for your response. I have completed all the test runs with different block size, results see to be good (I don't know ;-))

    ex for 8k block tested on 2GB file

    IOs/sec: 18475.63

    MBs/sec: 144.34

    Can I call these as better or ok or worst. I am trying to find would there any benchmark we can say this would before better for OLTP. thanks

  • mxy (5/7/2015)


    thanks Alan for your response. I have completed all the test runs with different block size, results see to be good (I don't know ;-))

    ex for 8k block tested on 2GB file

    IOs/sec: 18475.63

    MBs/sec: 144.34

    Can I call these as better or ok or worst. I am trying to find would there any benchmark we can say this would before better for OLTP. thanks

    I'm not the best resource on this, I'm still learning SQLIO and have never run it against SSDs. I'm going to reach out to some other folks and see if anyone can give you more insight on this.

    "I cant stress enough the importance of switching from a sequential files mindset to set-based thinking. After you make the switch, you can spend your time tuning and optimizing your queries instead of maintaining lengthy, poor-performing code."

    -- Itzik Ben-Gan 2001

Viewing 4 posts - 1 through 3 (of 3 total)

You must be logged in to reply to this topic. Login to reply