Since Fall 2004, the HCS Lab's primary emphasis of research with UPC is the investigation of key concepts and the development of a comprehensive suite of new research tools that directly support the performance analysis and optimization of global-address-space programs (UPC and SHMEM programs in particular) on complex HPC architectures and systems. This effort includes the study of existing and emerging performance analysis theory and tools, current and future HPC architectures, usability and user productivity preferences and methods, and the UPC and SHMEM programming models. The results of our in-depth investigation featured in the first year of this project are providing the foundation for the recently begun second year where the comprehensive high-level design serves as the basis for implementation, evaluation, and refinement of the PPW tools for UPC and SHMEM computing.
PPW heavily relies on the GASP performance tool interface, a joint work of the HCS lab with LNBL / UC Berkeley. More information on the GASP tool interface can be found at the GASP website.
PPW's work flow design has been influenced by various existing performance tools, including TAU, KOJAK, HPC Toolkit, CrayPat / Cray Apprentice^2, and valgrind / KCacheGrind.
PPW integrates with the Jumpshot trace visualization tool for visualizing trace data, and also integrates with the Vampir tool via an OTF export.
Finally, PPW takes advantage of the excellent timer and hardware counters provided by the GASNet tools and PAPI libraries.