Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Analysis of performance accelerator running ETMSP. Final report

Technical Report ·
OSTI ID:10104996
; ; ; ; ;  [1]
  1. Performance Processors, Inc., Palo Alto, CA (United States)

This project focused on predicting the performance of ETMSP, EPRI`s extended transient midterm stability program, running on PACE{trademark}/64, a 64-processor multiprocessing environment developed by the contractor, Performance Processors, Inc (PPI). PACE`s strength lies in solving large sparse-matrix problems quickly, exploiting parallel processing, PPI modified ETMSP 2P2 to run on a uniprocessor emulation of PACE, then measured parallel performance of ETMSP using five significant industry cases. Parallel ETMSP achieves breakthrough core speedups of approximately 29x from parallelism alone, well above the 4--8x range typical of prior work. Furthermore, the speedups on components of ETMSP suggest that PACE has the potential to accelerate other power-system applications with a sparse-matrix core. The overall purpose of this project was to provide fast transient analysis in support of on-line Dynamic Security Analysis (DSA). In 10 minutes, ETMSP`s core running on PACE/64 can evaluate 50 contingencies averaging 10 simulated seconds each for a 5000-bus power system with full generator and load modeling. In contrast, a 1993 workstation can process 1-2 contingencies in 10 minutes. Thus, ETMSP/PACE represents a considerable advance in the speed of transient stability analysis and a viable basis for an on-line DSA.

Research Organization:
Electric Power Research Inst., Palo Alto, CA (United States); Performance Processors, Inc., Palo Alto, CA (United States)
Sponsoring Organization:
Electric Power Research Inst., Palo Alto, CA (United States)
OSTI ID:
10104996
Report Number(s):
EPRI-TR--102856; ON: UN94003127
Country of Publication:
United States
Language:
English

Similar Records

Prototype on-line dynamic security assessment (DSA) program: User`s guide. Final report
Technical Report · Mon May 01 00:00:00 EDT 1995 · OSTI ID:87036

A Work-Efficient Parallel Sparse Matrix-Sparse Vector Multiplication Algorithm
Journal Article · Mon Jul 03 00:00:00 EDT 2017 · Proceedings - IEEE International Parallel and Distributed Processing Symposium (IPDPS) · OSTI ID:1525227

Vectorization of a multiprocessor multifrontal code
Journal Article · Sat Dec 31 23:00:00 EST 1988 · International Journal of Supercomputer Application; (USA) · OSTI ID:5422748