Design and Implementation of Parallel Debugger and Profiler for MPJ Express
Aleem Akhtar,Aamir Shafi,Mohsan Jameel
DOI: https://doi.org/10.48550/arXiv.1408.6347
2014-08-27
Abstract:MPJ Express is a messaging system that allows computational scientists to write and execute parallel Java applications on High Performance Computing (HPC) hardware. Despite its successful adoption in the Java HPC community, the MPJ Express software currently does not provide any support for debugging and profiling parallel applications and hence forces its users to rely on manual and tedious debugging/profiling methods. Support for such tools is essential to help application developers increase their overall productivity. To address this we have developed debugging and profiling tools for MPJ Express, which are the main topic of this paper. Key design goals for these tools include: 1) maintain compatibility with existing logging, debugging, and visualizing tools, 2) build these tools by extending existing debugging/profiling tools instead of reinventing the wheel. The first tool, named MPJDebug, builds on the open-source Eclipse Integrated Development Environment (IDE). It provides an Eclipse-based plugin developed using the Eclipse Plugin Development Environment (PDE). The default Eclipse debugger currently does not support debugging parallel applications running on a compute cluster. The second tool, named MPJProf, is a utility based on Tuning and Analysis Utility (TAU)-an open-source performance evaluation tool. Our goal here is to exploit TAU to profile Java applications parallelized using MPJ Express by generating profiles and traces, which can later be visualized using existing tools like paraprof and Jumpshot. Towards the end of the paper, we quantify the overhead of using MPJProf, which we found to be negligible in the profiling stage of parallel application development.
Distributed, Parallel, and Cluster Computing,Software Engineering