skill-tree:pe:3:b
**This is an old revision of the document!**
Table of Contents
PE3-B Benchmarking
Background
- Benchmarking is the activity to measure performance reliably and to assess the obtained performance.
- A benchmark is the act of running a computer program, a set of programs, or other operations, in order to assess the relative performance of an object, normally by running a number of standard tests and trials against it.
Aim
- Provide a basic understanding of benchmarking or performance measurement, so one can quantify successes and failures and use that information to improve the application performance.
- Prepare a method of comparing the performance of various subsystems across different chip/system architectures.
- To assess speedups and efficiencies as the key measures for benchmarks of a parallel program.
- To differentiate between strong and weak scaling
- To assess the performance impact of certain features of current CPU architectures (temperature and dynamic CPU frequencies)
Outcomes
- Consider not only the speed of computational performance, but also the qualities of service, the total cost of ownership and facilities burden (space, power, and cooling).
- Differentiate types of benchmarks.
- Assess speedups and efficiencies as the key measures for benchmarks of a parallel program
- Benchmark the runtime behaviour of parallel programs, performing controlled experiments by providing varying HPC resources (e.g. 1, 2, 4, 8, … cores on shared memory systems or 1, 2, 4, 8, … nodes on distributed systems for the benchmarks)
Subskills
skill-tree/pe/3/b.1593109568.txt.gz · Last modified: 2020/06/25 20:26 by kai_h