That looks pretty normal to me. The JVM doesn't do a full GC until the old gen crosses a (dynamic, defaults to about 70% of heap) threshold.
On Thu, Jun 17, 2010 at 1:01 PM, Mike Subelsky <m...@subelsky.com> wrote: > I just started munin to monitor the performance of my cluster which is > working great. I noticed the attached pattern for JVM heap size - > lots of peaks and valleys, which looks like GC is firing pretty > frequently. I don't have a lot of experience with Java apps, and our > installation is relatively untuned (3 large EC2 instances each with 5G > heap memory with the commit log on an ephemeral drive and the data > drive on an EBS volume). Does that pattern look bad or is it > relatively normal? If it's not normal, any advice for what I should > change to get better performance? > > Big thanks to Jonathan and James for building these awesome plugins. > As a JMX novice this was a huge boost: > > http://github.com/jamesgolick/cassandra-munin-plugins > > thanks! > > -Mike > > -- > Mike Subelsky > oib.com // ignitebaltimore.com // subelsky.com > @subelsky > -- Jonathan Ellis Project Chair, Apache Cassandra co-founder of Riptano, the source for professional Cassandra support http://riptano.com
