Sorry, I could not understand this sentence: """ When polling the metrics/snapshot, in my case when running short sub-second tasks, I have to poll even faster than the churn time of tasks to get an accurate picture of resource allocation. """ Could you explain more?
On Thu, Aug 6, 2015 at 6:32 PM, Hans van den Bogert <[email protected]> wrote: > Hi, > > I’m trying to plot resource usage and allocation. Basically what I want > can already be seen in the web UI (though not plotted). > 1) amount of allocated resources > 2) actual usage of allocated resources > 3) Max share of a framework / per framework > > Now the allocated resources can be gotten through the metrics page, > although using the metrics/snapshot does give me the problems associated > with polling. When polling the metrics/snapshot, in my case when running > short sub-second tasks, I have to poll even faster than the churn time of > tasks to get an accurate picture of resource allocation. > > If I use the (event) log files of master and slaves, I can’t see usage, > and retrieving/calculating max share per framework is going to be tough, I > guess. > > Does someone have a solution or idea to plotting the above three listed > metrics accurately. > > Regards, > > Hans van den Bogert -- Best Regards, Haosdent Huang
