30ms pauses are on the low side of normal for a 800MB young gen under
parnew. We're not going to be able to get rid of those, although it
looks like you're seeing objects in the young gen die *very* quickly,
so cutting that to say 200MB might give you shorter (and more
frequent) pauses for the you
MAX_HEAP_SIZE="9G"
HEAP_NEWSIZE="800M"
2 socket quad core
44GB RAM
Cassandra 1.0.7
[edward@cdbla120 cassandra]$ free -g
total used free sharedbuffers cached
Mem:43 34 8 0 0 25
-/+ buffers/cache: 8
We put a bunch of effort into reducing GC pauses in 1.0, which worked
great on our synthetic benchmarks but I don't have much data on how
well that's working in practice, on real-world workloads.
If you're running a Cassandra app in production or testing, I'd
appreciate it if you could send me [ei
