Using Hardware Performance Monitors to Understand the Behavior of Java Applications

Modern Java programs, such as middleware and application servers, include many complex software components. Improving the performance of these Java applications requires a better understanding of the interactions between the application, virtual machine, operating system, and architecture. Hardware performance monitors, which are available on most modern processors, provide facilities to obtain detailed performance measurements of long-running applications in real time. However, interpreting the data collected using hardware performance monitors is difficult because of the low-level nature of the data. We have developed a system, consisting of two components, to alleviate the difficulty of interpreting results obtained using hardware performance monitors. The first component is an enhanced VM that generates traces of hardware performance monitor values while executing Java programs. This enhanced VM generates a separate trace for each Java thread and CPU combination and thus provides accurate results in a multithreaded and multiprocessor environment. The second component is a tool that allows users to interactively explore the traces using a graphical interface. We implemented our tools in the context of Jikes RVM, an open source Java VM, and evaluated it on a POWER4 multiprocessor. We demonstrate that our system is effective in uncovering as yet unknown performance characteristics and is a first step in exploring the reasons behind observed behavior of a Java program.

[1]  Jeffrey S. Vetter,et al.  Scalable Analysis Techniques for Microprocessor Performance Counter Metrics , 2002, ACM/IEEE SC 2002 Conference (SC'02).

[2]  Luiz De Rose.  The Hardware Performance Monitor Toolkit , 2001, Euro-Par.

[3]  Oscar Naim,et al.  Dynamic instrumentation of threaded applications , 1999, PPoPP '99.

[4]  Vivek Sarkar,et al.  The Jalape ~ no Dynamic Optimizing Compiler for Java TM , 1999 .

[5]  Lizy Kurian John,et al.  Workload characterization of multithreaded java servers , 2001, 2001 IEEE International Symposium on Performance Analysis of Systems and Software. ISPASS..

[6]  Barton P. Miller,et al.  The Paradyn Parallel Performance Measurement Tool , 1995, Computer.

[7]  D.A. Reed,et al.  Scalable performance analysis: the Pablo performance analysis environment , 1993, Proceedings of Scalable Parallel Libraries Conference.

[8]  Jeffrey C. Mogul,et al.  The effect of context switches on cache performance , 1991, ASPLOS IV.

[9]  Erik Hagersten,et al.  Memory system behavior of Java-based middleware , 2003, The Ninth International Symposium on High-Performance Computer Architecture, 2003. HPCA-9 2003. Proceedings..

[10]  Daniel A. Reed,et al.  SvPablo: A multi-language architecture-independent performance analysis system , 1999, Proceedings of the 1999 International Conference on Parallel Processing.

[11]  Robert J. Fowler,et al.  HPCVIEW: A Tool for Top-down Analysis of Node Performance , 2002, The Journal of Supercomputing.

[12]  David Grove,et al.  Adaptive online context-sensitive inlining , 2003, International Symposium on Code Generation and Optimization, 2003. CGO 2003..

[13]  Sheng Liang,et al.  Java Virtual Machine Profiler Interface , 2000, IBM Syst. J..

[14]  Bowen Alpern,et al.  Implementing jalapeño in Java , 1999, OOPSLA '99.

[15]  Perry Cheng,et al.  Oil and water? High performance garbage collection in Java with MMTk , 2004, Proceedings. 26th International Conference on Software Engineering.

[16]  Mikko H. Lipasti,et al.  An architectural evaluation of Java TPC-W , 2001, Proceedings HPCA Seventh International Symposium on High-Performance Computer Architecture.

[17]  Stephen J. Fink,et al.  The Jalapeño virtual machine , 2000, IBM Syst. J..

[18]  S. Turner,et al.  Performance Analysis Using the MIPS R10000 Performance Counters , 1996, Proceedings of the 1996 ACM/IEEE Conference on Supercomputing.

[19]  Lieven Eeckhout,et al.  How java programs interact with virtual machines at the microarchitectural level , 2003, OOPSLA 2003.

[20]  Lance M. Berc,et al.  Continuous profiling: where have all the cycles gone? , 1997, ACM Trans. Comput. Syst..

[21]  Laurie J. Hendren,et al.  Dynamic metrics for java , 2003, OOPSLA '03.

[22]  James R. Larus,et al.  Exploiting hardware performance counters with flow and context sensitive profiling , 1997, PLDI '97.

[23]  William Gropp,et al.  From Trace Generation to Visualization: A Performance Framework for Distributed Parallel Systems , 2000, ACM/IEEE SC 2000 Conference (SC'00).

[24]  William Gropp,et al.  Toward Scalable Performance Visualization with Jumpshot , 1999, Int. J. High Perform. Comput. Appl..

[25]  Lizy K. John,et al.  Workload Characterization of Java Server Applications on Two PowerPC Processors , 2002 .

[26]  Matthew Arnold,et al.  Online feedback-directed optimization of Java , 2002, OOPSLA '02.

[27]  Jack J. Dongarra,et al.  A Scalable Cross-Platform Infrastructure for Application Performance Tuning Using Hardware Counters , 2000, ACM/IEEE SC 2000 Conference (SC'00).