While reading a JFR dump, I came up with a VM Operation called GenCollectForAllocation
that is taking more time than other processes. Can anyone explain to me what this operation is?
This is the reason of why a safepoint was triggered (there are many of them). You can read it as : "a generational collector just had an allocation failure", as such a safepoint will be triggered so that the application is stopped (stop-the-world event).
It seems these are cumulative times, so a total sum of all times is added; considering that minor GC happens a lot and if your application is running for a long time - this is not really a problem. You also had 223
of such events, with the longest duration of 305 ms
, but on average (9.786 / 223
): 43ms
. That is a healthy figure, IMHO.