One of the users also brought up the issue that we don't have any metrics for entry processors [1]. Probably, makes sense to include this into IEP as well.
-Val On Mon, Nov 13, 2017 at 4:48 PM, Denis Magda <[email protected]> wrote: > Hi Anton, > > Thanks for brining this out! > > The memory metrics really need a serious overhaul. For instance we a > missing basic things like: > - Bytes used per data region. > - Bytes used per cache. > - Size of checkpointing buffer. > > Plus advanced memory metrics will not be harmful: > - Space occupied by indexes. > > The ticket for the memory metrics improved is anxiously waiting to be > solved: https://issues.apache.org/jira/browse/IGNITE-5796 < > https://issues.apache.org/jira/browse/IGNITE-5796> > > — > Denis > > On Nov 13, 2017, at 9:14 AM, Anton Vinogradov <[email protected]> wrote: > > > > Igniters, > > > > As you may know Ignite have a lot of JMX based metric, but to perform > more > > effective grid monitoring some new JMX metrics needs to be implemented. > > > > Here's the accumulated list of metrics I'd like to see at nearest Ignite > > version: > > Topology > > - Current topology version > > - Total server nodes count > > - Total client nodes count > > - Method to count nodes filtered by some node attribute > > - Method to count nodes grouped by some node attribute > > > > Communication SPI > > - Received messages count grouped by message type > > - Received messages count grouped by sender node > > - Sent messages count grouped by message type > > - Sent messages count grouped by receiver node > > > > Partitions allocation (for cache groups) > > - Total primary partitions count located on the current node > > - Total backup partitions count located on the current node > > - Min/max partition backups left in the cluster for cache group > > - Maybe some methods to show partitions map/partition distribution > > statistics in the cluster > > > > Jobs execution > > - Total jobs execution time (now job execution statistics gathered since > > node started and can't be used to calculate average execution time > between > > probes, implementation of this metric can solve this problem) > > > > Cache > > - Topology validation status for cache > > > > IEP-6 [1] was prepared and ready to be discussed. > > > > Please, feel free to suggect or decline some metric. > > > > Anton. > > > > [1] > > https://cwiki.apache.org/confluence/display/IGNITE/IEP- > 6%3A+Metrics+improvements > >
