The metrics were parsed from the gc log with -XX:+PrintFLSStatistics=1 on. I created new YCSB workloads for it. I updated the doc a bit and attached it as PDF to this new JIRA:
https://issues.apache.org/jira/browse/HBASE-3455 <https://issues.apache.org/jira/browse/HBASE-3455>-Todd On Thu, Jan 20, 2011 at 9:48 AM, Ted Dunning <[email protected]> wrote: > Also, can you say what YCSB workload that used? > > On Thu, Jan 20, 2011 at 9:33 AM, Todd Lipcon <[email protected]> wrote: > > > I did some experiments to understand our full GC issues better last > night. > > Here are the results: > > http://people.apache.org/~todd/hbase-fragmentation/ > > > > <http://people.apache.org/~todd/hbase-fragmentation/>Basically my > > conclusion > > here is that (for these YCSB workloads) the memstore is way worse for > heap > > fragmentation than the block cache. > > > > Also we now have some tools and reference for comparison if we make any > > changes to memstore to try to combat this issue. > > > > -Todd > > -- > > Todd Lipcon > > Software Engineer, Cloudera > > > -- Todd Lipcon Software Engineer, Cloudera
