Design of JFluid: a profiling technology and tool based on dynamic bytecode instrumentation

  • Authors:

  • Mikhail Dmitriev

  • Affiliations:

  • Sun Microsystems Laboratories, Mountain View, CA

  • Venue:
  • Design of JFluid: a profiling technology and tool based on dynamic bytecode instrumentation
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

Instrumentation-based profiling has many advantages and one serious disadvantage: usually high performance overhead. This overhead can be substantially reduced if only a small part of the target application (for example, one that has previously been identified as a performance bottleneck) is instrumented, while the rest of the application code runs at full speed. Such an approach can also beat scalability issues caused by a high volume of profiling information generated by instrumented code running on behalf of multiple threads. The value of such a profiling technology would increase further if the code could be instrumented and de-instrumented as many times as needed at run time. In this report we describe in detail the design of an experimental profiling system called JFluid, which includes a modified Java HotSpot™ VM and a GUI tool, and addresses both of the above issues. Our JVM™ supports arbitrary on-the-fly modifications to running Java methods, and can connect with a profiling tool at any moment, without any startup time preparation. Our tool collects, processes and presents profiling data on-line. To perform CPU profiling, it instruments a group of methods defined as an arbitrary "root" method plus all methods that it calls (a call subgraph). It appears that static determination of all methods in a call subgraph is difficult in the presence of virtual methods, but fortunately, with dynamic code hotswapping available, two schemes of dynamic call subgraph revelation and instrumentation can be suggested. Measurements that we obtained when performing full and partial program profiling using both schemes show that the overhead can be reduced substantially using this technique, and that one of the schemes generally results in a smaller number of instrumented methods and better performance, especially for large applications.