Java runtime systems: Characterization and architectural implications

The Java Virtual Machine (JVM) is the cornerstone of Java technology and it s efficiency in executing the portable Java bytecodes is crucial for the su ccess of this technology. Interpretation, Just-in-Time (JIT) compilation, a nd hardware realization are well-known solutions for a JVM and previous res earch has proposed optimizations for each of these techniques. However, eac h technique has its pros and cons and may not be uniformly attractive for a ll hardware platforms, Instead, an understanding of the architectural impli cations of JVM implementations with real applications can be crucial to the development of enabling technologies for efficient Java runtime system dev elopment on a wide range of platforms. Toward this goal, this paper examine s architectural issues from both the hardware and JVM implementation perspe ctives. The paper starts by identifying the important execution characteris tics of Java applications from a bytecode perspective. It then explores the potential of a smart JIT compiler strategy that can dynamically interpret or compile based on associated costs and investigates the CPU and cache arc hitectural support that would benefit JVM implementations. We also study th e available parallelism during the different execution modes using applicat ions from the SPECjvm98 benchmarks. At the bytecode level. it is observed t hat less than 45 out of the 256 bytecodes constitute 90 percent of the dyna mic bytecode stream. Method sizes fall into a trinodal distribution with pe aks of 1, 9, and 26 bytecodes across all benchmarks. The architectural issu es explored in this study show that. when Java applications are executed wi th a JIT compiler, selective translation using good heuristics can improve performance, but the saving is only 10-15 percent at best. The instruction and data cache performance of Java applications are seen to be better than that of C/C++ applications except in the case of data cache performance in the JIT mode. Write misses resulting from installation of JIT compiler outp ut dominate the misses and deteriorate the data cache performance in JIT mo de. A study on the available parallelism shows that Java programs executed using JIT compilers have parallelism comparable to C/C++ programs for small window sizes, but falls behind when the window size is increased. Java pro grams executed using the interpreter have very little parallelism due to th e stack nature of the JVM instruction set, which is dominant in the interpr eted execution mode. In addition, this work gives revealing insights and ar chitectural proposals for designing an efficient Java runtime system.