is hammer for us

Mag Gam magawake at gmail.com
Wed Aug 12 03:52:54 PDT 2009


I was under the impression HAMMER was a parallel filesystem. sorry


On Wed, Aug 12, 2009 at 1:04 AM, Matthew
Dillon<dillon at apollo.backplane.com> wrote:
>
> :
> :The I/O bottleneck is coming from the disk subsystem and network. I
> :was wondering if HAMMER can do parallel filesystem implementation
> :similar to GPFS or Lustre.
> :
> :Also, the reads/writes are random access there is very little
> :sequential streaming, but the files are large.Each file is around 30GB
> :each
>
>    It can do master->multi_slave replication if that is what you mean.
>    I don't know how that might compare to GPFS or Lustre.  You are going
>    to have more choices in linux-land then in BSD-land, particularly if
>    you have a large array of drives.
>
>    If there is an I/O bottleneck from the disks due to random access
>    seeks the only solution is more spindles.
>
>    If you can stage the data in any way a large SSD (solid state drive)
>    might help.  e.g. one or more 256G SSDs for data staging eliminates
>    the seek bottlneck and probably also eliminates the need for large
>    amounts of ram in the machines.
>
>                                                -Matt
>
>





More information about the Users mailing list