EP ANALYTICS, INC.. — Department of Energy SBIR Phase I: 07b

Description

The Arm microprocessor architecture has emerged as a legitimate alternative and competitor to predominant x86-based supercomputing architectures. An easy‐to‐use and robust tool chain for understanding, porting and refactoring existing HPC codes (e.g., commercial, academic, and DOE Office of Science codes) is essential to unlocking the full potential of Arm in the HPC realm. EP Analytics is developing and commercializing PerfPal, a set of performance analysis, modeling and prediction tools that can be deployed to 1) design future Arm-based architectures for High-Performance Computing (HPC) applications, and 2) identify potential optimizations on application codes that can improve performance on existing and upcoming Arm-based systems. PerfPal relies on raw performance data extracted through direct analysis of the production binary, through a technique known as Binary Instrumentation (BI), to provide an in-depth view of application performance. Performance modeling/prediction component of PerfPal can utilize performance data collected on existing systems to project performance on future hardware design points. An ancillary product of the proposed effort is EP Analytics’ Arm-based BI toolkit, EPAX, which will be released under open-source license to benefit the overall Arm software ecosystem. PerfPal relies on a suite of low-level BI-based performance analysis tools that provide raw information about how HPC codes interact with and utilize specific performance-critical sub-components of supercomputing systems. During Phase I, we propose to develop foundational tools that will enable PerfPal to present high-level performance analysis to the end-users. In addition, a new tool will be designed that will facilitate bringing SST (Structural Simulation Toolkit), a popular co-design framework, to the Arm architecture. The Arm software development tools being created by EP Analytics have commercial applications across high performance computing, “hyper-scale” (Internet) data centers and cloud computing. This research also has the potential to contribute to and advance the national leadership in “Exascale” supercomputing technology.