Dynamic Java Program Corpus Analysis

Abstract

Program corpus analysis is important in optimizaton of run-time systems. Conventional linguistic analysis is static in nature and can't reflect dynamic behaviors revealed by versatile object-orented programming languages. Pattern based run-time profiler is proposed and realized in this paper. Unlike conventional profiler or run-time visualization tools, representative program corpora accumulated and benchmarked not only show monolithic functions tha introduce excessive run-time overhead but also reflect their correlated code patterns. We proposed pattern-based analysis to address program run-time bottleneck in a sequence of method invocations. It will reveal more semantic meanings in performance bottleneck rendered by objcet-oriented programming system.