On Thu, May 29, 2014 at 2:41 AM, Askhat Asanaliev <[email protected]> wrote:
> Alex, Yigit,
>
> Please take a look at the following message when you have time. This is
> about monitoring statistics for CHOP. I would like to clarify the main
> points before starting the implementation. Just to be sure that there is no
> confusion or I am not missing any points.
>
> (1) The metrics we want to take: cpu, ram, net, disk.

Definitely.  Can we elaborate a bit on what sort of metrics we'll
intend to gather for each system module?

> (2) We take all these from the runner instances only, i.e. from the hosts
> where the runners are executed. We don’t measure the stats on the
> coordinator instance (i.e. the host where the web app runs).

Agreed

> (3) The stats data is taken on a runner instance but stored (accumulated)
> on the coordinator instance.

Agreed, this scales really well historically

> (4) The data is pushed (from a runner to the coordinator) only after the
> testing is completed. Thus we don’t want to affect the metric data by
> communication load between the runners and the coordinator.

Again, a very good idea IMHO.

> (5) The chart with these stats is displayed on the iteration view.

Sounds good.

>
> Please let me know if all these are correct or may be we need to add/remove
> points.
>
> The next actions:
> 1. Do some research for utilities to gather the stats.
> 2. Setup and integrate a utility in our running environment.
>
> Regards,
> Askat

Reply via email to