Commercialization of User-friendly Performance Analysis & Profiling Tools to Enable High Performance Computing (HPC) Applications on the ARM Processor architecture.

Period of Performance: 02/21/2017 - 11/20/2017

$229K

Phase 1 SBIR

Recipient Firm

EP Analytics, Inc.
12121 Scripps Summit Drive Suite 130
San Diego, CA 92131
Firm POC
Principal Investigator

Abstract

The “ARM” microprocessor architecture, commonly used in cell phones and tablets, is emerging as a key technology for energy-efficient high performance computing (HPC). An easy-to-use and robust tool chain for understanding, porting and refactoring existing HPC codes is essential to unlocking the full potential of ARM in the HPC realm. General Statement of How this Problem is Being Addressed: EP Analytics will partner with ParaTools Inc. to commercialize an easy-to-use set of performance analysis and profiling tools to enable HPC applications on ARM processor-based systems. To accomplish this, EP Analytics is developing an open source binary instrumentation (BI) toolkit for ARM called “EPAX,” which will be integrated with the ParaTools’ “TAU Performance System” to provide a set of performance analysis tools for ARM with visual/graphical aids. What is to be done in Phase I: During Phase I we propose to develop a prototype component of EPAX (a BI based function timer tool) and integrate the prototype within the TAU Performance System. To demonstrate the feasibility of our approach, this will be tested on HPC applications to generate function timer profiles. Our Phase II efforts will focus on developing a fully functional release of EPAX and fully integrating it within TAU, “hardening” the tools for commercial release. Commercial Applications and Other Benefits: The ARM software development tools being created by EP Analytics have commercial applications across HPC, hyperscale data centers, and cloud and mobile computing. In the case of the latter, the potential exists to enable HPC “apps” on future tablets and smartphones, where today ARM is the predominant architecture. This research also has the potential to contribute to and advance the national leadership in “Exascale” supercomputing technology.