Percolator throughput is not stable

this is the statistic I'm seeing from marvel in my dev cluster.

I run a new process around 11:00 and as you see throughput fluctuates and
eventully stuck at less than 10k/s which is worse than the throughput in
it's early phase.

It seems very weired to me, I looked up resource usage(cpu, mem) nothing
seems problematic.

Can you guess any reason for this?

You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
To view this discussion on the web visit
For more options, visit