Visit Open-E website
Results 1 to 3 of 3

Thread: Replication performance problems

  1. #1

    Default Replication performance problems

    We have two iSCSI SANS both running iSCSI R3 v5.0 and are trying to replicate a volume between the two.

    They are currently sat side by side but running through a VPN to simulate the architecture when one is moved off site.

    Running through the vpn we are only getting replication task throughput of around 4388kb/s which on a 2.5TB volume will take a while to complete.

    We are not using teamed NIC's; all replication traffic is through dedicated NIC's and all user storage traffic through separate NIC's.

    Any suggestions as to how we can improve the throughput --- could it be anything to do with using Block I/O rather than File I/O???

    Any suggestions gratefully received!

    Cheers,
    Carl

  2. #2

    Lightbulb

    For doing (synchronous) volume replication, you're going to need a REALLY fast VPN. The inexpensive one we can find won't go much faster than a T1 (which is fine, since that's the only connection we'll need), but we use (asynchronous) data replication (i.e. file-level replication), so 1.5 Megabits/sec is fine for us.

    I am thinking you would need a VERY fast VPN to get 40Megabytes/sec (i.e. 320 Megabits/sec), which is the default bandwidth set aside for volume replication.

    BTW, is that 4388kbits/s or 4388kbytes/s?

  3. #3

    Default

    Also check the RAID controller to make sure you have all the cache values set for performance as I have seen this cause slow replication speeds.
    All the best,

    Todd Maxwell


    Follow the red "E"
    Facebook | Twitter | YouTube

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •