ParTools allows the interactive analysis of a C program execution profile and data dependencies to facilitate the discovery and selection of suitable parallelization candidates in a manual parallelization process. The flow does not assume any specific parallelization technique, thus it can be broadly applied.

The original (serial) C source is automatically annotated to trace the execution profile and data dependencies at run-time. The annotated program is then executed using a significant (but small) data set selected by the developer. The data collected is cross-referenced with the original source and can be interactively analyzed graphically to determine the best parallelization candidates and techniques.

Features

  • Automatic C source annotation for run-time tracing.
  • Automatic run-time data collection.
  • Interactive graphical analysis of execution profile and data dependencies.
  • Cross-reference with the original C source.
  • Can be used to support analysis for any parallelization technique.

Project Samples

Top-level view of an MJPEG encoder algorithm.

Project Activity

See All Activity >