You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@hbase.apache.org by Tim Sell <tr...@gmail.com> on 2011/02/09 19:42:57 UTC

Most useful metrics?

What do people find to be the most useful metrics for monitoring their
cluster? Both for performance and long term planning.

Presumably requests and disk-space are high up the list.

Is there any useful ones that aren't covered by the metrics package?

~Tims

Re: Most useful metrics?

Posted by Wayne <wa...@gmail.com>.
Compaction Queue size usually explains a lot. That along with load and disk
utilization are what I use the most. I am definitely interested in what
others use, especially metrics that indicate early for problems.

Thanks.

On Wed, Feb 9, 2011 at 1:42 PM, Tim Sell <tr...@gmail.com> wrote:

> What do people find to be the most useful metrics for monitoring their
> cluster? Both for performance and long term planning.
>
> Presumably requests and disk-space are high up the list.
>
> Is there any useful ones that aren't covered by the metrics package?
>
> ~Tims
>