Blogbench RAID benchmarks

Siju George at
Mon Jul 25 03:19:10 PDT 2011

On Fri, Jul 22, 2011 at 9:54 PM, Matthew Dillon
<dillon at> wrote:
>    Ok, after much experimentation I've figured out what is going on.
>    So, I am going to commit some experimental code to HAMMER which tries
>    to manage the locking conflicts between the frontend reader threads and
>    the backend flusher threads.  I am going to do this by creating a
>    pulse-width modulated time-domain multiplexer in HAMMER which tries
>    to 'slot in' reads and writes based on the number of inodes backlogged
>    in the flusher.

So Should i wait updating my Server's Development Version till this
code is considered stable?

Thanks :-)


More information about the Kernel mailing list