[Gluster-devel] Multiplexing - good news, bad news, and a plea for help

[Jeff Darcy]

> I wonder if we are spending more time in io-threads. Does setting
> idle-time in io-threads to 1 help with anything?
> It might be useful to add instrumentation subsequently to dump
> statistics (number of fops serviced, time spent in servicing) per
> thread. Having the same visibility for our epoll threads would also be
> useful.
> 
>  I think the answer is no - but does profiling yield anything useful
> at this point?

Plain profiling (perf record/report) shows approximately what I'd
expect if things were working properly - lots of time in lots of
routines, lookup fops especially noticeable but nothing dominant.
At this point I think it's less about raw execution time than
about what executes in which order.