Well my requests are fully random so I expected to see them distributed
quite evenly across all the regions.
How can I know which datanode was hosting the .META. table? I'm not running
the cluster currently so I cannot check this for the moment.

Thanks for your input!

Thibault Dory

On Fri, May 13, 2011 at 4:06 PM, Ted Yu <[email protected]> wrote:

> HBase, even the trunk version, doesn't balance load measured by read
> requests.
> Its balancer tries to put equal number of regions on each server - as shown
> in Figure 13.
> Balancing by request count is the goal for next generation balancer.
>
> Did you remember whether client3 was also hosting .META. table ?
>
> Thanks
>
> On Fri, May 13, 2011 at 4:06 AM, Thibault Dory <[email protected]
> >wrote:
>
> > Hello,
> >
> > I have written with a few other people a paper for the ACM Symposium
> > On Cloud Computing. This paper describes the methodology,
> > infrastructure and configuration used as well as the results obtained
> > for elasticity and scalability of three noSQL databases, of wich
> > HBase. The paper can be downloaded here :
> > http://www.nosqlbenchmarking.com/wp-content/uploads/2011/05/paper.pdf<
> >
> http://www.google.com/url?sa=D&q=http://www.nosqlbenchmarking.com/wp-content/uploads/2011/05/paper.pdf
> > >
> >
> >
> > Any feedback on the methodology used would be appreciated, we would
> > like to know if HBase is used in a "fair" way in those tests.
> >
> > We also encountered a problem with the distribution of requests among
> > region
> > servers. This problem is described in section 5.4.2 and any hints on how
> to
> > solve this problem would be appreciated. Please note that the request
> > generation is independent of the specific database layer and that we did
> > not
> > observe this problem for the two other databases.
> >
> > Regards,
> >
> > Thibault Dory
> >
>

Reply via email to