Hey Mathias, What can be inferred from the high capacity values?Should we be worried? What should we do to change it ?
Thanks Aditya On Tue, Jun 23, 2015 at 5:54 PM, Nathan Leung <[email protected]> wrote: > Also to clarify, unless you change the sample frequency the counts in the > ui are not precise. Note that they are all multiples of 20. > On Jun 23, 2015 7:16 AM, "Matthias J. Sax" <[email protected]> > wrote: > >> I don't see any in-balance. The value of "Executed" is 440/460 for each >> bolt. Thus each bolt processed about the same number of tuples. >> >> Shuffle grouping does a round robin distribution and balances the number >> of tuples sent to each receiver. >> >> I you refer to the values "capactiy", "execute latency", or "process >> latency", shuffle grouping cannot balance those. Furthermore, Storm does >> not give any support to balance them. You would need to implement a >> "CustomStreamGrouping" or use direct-grouping to take care of load >> balancing with regard to those metrics. >> >> >> -Matthias >> >> >> >> On 06/23/2015 11:42 AM, bhargav sarvepalli wrote: >> > >> > >> > I'm leading a spout with 30 executors into this bolt which has 90 >> > executors. Despite using shuffle grouping , the load seems to be >> > unbalance.Attached is a screenshot showing the same. Would anyone happen >> > to know why this is happening or how this can be solved? >> > >> > >> > >> > >> >>
