Portable and accurate collection of calling-context-sensitive bytecode metrics for the Java virtual machine

Abstract

Calling-context profiles and dynamic metrics at the bytecode level are important for profiling, workload characterization, program comprehension, and reverse engineering. Prevailing tools for collecting calling-context profiles or dynamic bytecode metrics often provide only incomplete information or suffer from limited compatibility with standard JVMs. However, completeness and accuracy of the profiles is essential for tasks such as workload characterization, and compatibility with standard JVMs is important to ensure that complex workloads can be executed. In this paper, we present the design and implementation of JP2, a new tool that profiles both the inter-and intra-procedural control flow of workloads on standard JVMs. JP2 produces calling-context profiles preserving callsite information, as well as execution statistics at the level of individual basic blocks of code. JP2 is complemented with scripts that compute various dynamic bytecode metrics from the profiles. As a case-study and tutorial on the use of JP2, we use it for crossprofiling for an embedded Java processor.

Efficient and Scalable Collection of Dynamic Metrics Using MapReduce

“…Hassoun et al [19] in their empirical validation study of dynamic coupling metrics considered the scalability of models as a threat to validity. Sarimbekov et al [28] measured the calling context sensitive Java bytecode metrics from XML file (containing CCT profile) created using JP2. They mentioned that the time and space consumption of such an analysis depends largely on the XQuery processor used, and hence it becomes crucial to use a processor that is proficient in streaming large input document (like the XML-based calling-context-tree profiles) since their size may exceed the main memory limits.…”

A comprehensive toolchain for workload characterization across JVM languages

“…Among the most heavyweight of our tools is JP2, which produces calling context trees; this incurs an overhead factor of roughly 100 [25]. However, this cost is amortized in that many different metrics are computed (as queries) over its output.…”

JVM-hosted languages

“…The first profiler we use is a modified version of JP2 2.1 [23] to collect information about the bytecode instructions and methods used by each benchmark. The original version of JP2 records method and basic block execution counts; our version also records the bytecode instructions within each basic block.…”

“…Our work relies in part on earlier incarnations of their tools (e.g. JP2 [23]) and has the same motivation of JVM-based cross-language comparison.…”