Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In with Google

Our Sites

covecube.com 
community.covecube.com 
blog.covecube.com 
wiki.covecube.com 
bitflock.com 
stablebit.com 

Poll

No poll attached to this discussion.

In order to better support our growing community we've set up a new more powerful forum.


The new forum is at: http://community.covecube.com


The new forum is running IP.Board and will be our primary forum from now on.


This forum is being retired, but will remain online indefinitely in order to preserve its contents. This forum is now read only.


Thank you,

Write / Read Performance of Pool Drive Vs Non-pool Drive

edited December 2011 in DrivePool
I have just installed the M3 beta, my first taste of a Drive-pooler for Vali.  First impressions look good, almost looks like an official MS addon.  No problems with the basic tasks I have done so far, however I did notice that when copying to a duplicated pool drive, the copy performance drops from around 120MBps (when copying straight drive A to drive B), versus ~40-50Mbps from Drive A -> Pool Drive.

I figure at the point of copying, DrivePool is real time copying to both drives simultaneously, and hence is bottlenecks the performance.  Can anyone confirm if this is indeed the case and if these speeds seem normal. Also if this is the case I presume read speeds should be unaffected, or does DP read from both devices to verify data integrity?

Comments

  • I have also just noticed that when the DrivePool has only 1 disk in it, to move data from a non-pooled to a pooled folder on the same drive results in a 50% performance drop (50Mbps) compared to the HDD max speed.  This must be because it is reading from one part and writing to another of the same disk, so not very efficient.  Under a normal windows situation this would be isntantanous (as its moving the files) as the files would not actually be moved, instead only the indexes to the file.  Is there any way to overcome this as for large amounts of data this is a long process?

    I am not sure what the knock on effect would be in a multi drive pool?
  • Resident Guru
    At least as of M3 build 1433, the only way to overcome this is to *manually* move the data into the pool's folders (i.e. by stopping the services and directly accessing the pool folders on the physical drives rather than using the shares).

    In a multi-drive pool, if you did this you would also have to make sure you maintained DrivePool's duplication structure (if/as applicable) or upon restarting the services DrivePool would probably complain about the pool's health.

    Basically, hard-hat zone.
  • I just had the same thought, as its easy to find the correct hidden directory, however I am concerned about maintaining data integrity, in addition to not knowing the right service to shut down.

    Maybe this is a thought for the final stage before release  on improving performance / efficiency.  For now I am happy to wait  :)
Sign In or Register to comment.