Uploaded image for project: 'JDK'
  1. JDK
  2. JDK-8264651

Renaissance-Neo4jAnalytics is 45% faster on lucky compilation order

XMLWordPrintable

      I happened to notice that JmhNeo4jAnalytics is much faster with -XX:-TieredCompilation -XX:CICompilerCount=1 compared to default:
      12494.47 ms/op
      WIth -XX:-TieredCompilation -XX:CICompilerCount=1
      6830.58

      The perfasm profiles are so different I am not yet sure what is going on. With default it spends much more time in stubs.

            Unassigned Unassigned
            ecaspole Eric Caspole
            Votes:
            0 Vote for this issue
            Watchers:
            4 Start watching this issue

              Created:
              Updated: