English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

UNIPROCESSOR PERFORMANCE ANALYSIS OF A REPRESENTATIVE WORKLOAD OF SANDIA NATIONAL LABORATORIES' SCIENTIFIC APPLICATIONS Master of Science in Electrical Engineering New Mexico State University Las Cruces, New Mexico, 2005 Dr. Jeanine Cook, Chair Throughout the last decade computer performance analysis has become absolutely necessary to maximum performance of some workloads. Sandia National Laboratories (SNL) located in Albuquerque, New Mexico is no different in that to achieve maximum performance of large scientific, parallel workloads performance analysis is needed at the uni-processor level. A representative workload has been chosen as the basis of a computer performance study to determine optimal processor characteristics in order to better specify the next generation of supercomputers. Cube3, a finite element test problem developed at SNL is a representative workload of their scientific workloads. This workload has been studied at the uni-processor level to understand characteristics in the microarchitecture that will lead to the overall performance improvement at the multi-processor level. The goal of studying vthis workload at the uni-processor level is to build a performance prediction model that will be integrated into a multi-processor performance model which is currently being developed at SNL. Through the use of performance counters on the Itanium 2 microarchitecture, performance statistics are more » studied to determine bottlenecks in the microarchitecture and/or changes in the application code that will maximize performance. From source code analysis a performance degrading loop kernel was identified and through the use of compiler optimizations a performance gain of around 20% was achieved. « less

Uniprocessor Performance Analysis of a Representative Workload of Sandia National Laboratories' Scientific Applications.