Influences, challenges, and strategies for automotive HPC benchmarking and performance improvement
| Autor: | Myron Ginsberg |
|---|---|
| Rok vydání: | 1999 |
| Předmět: |
Computer Networks and Communications
Process (engineering) business.industry Computer science Automotive industry Benchmarking computer.software_genre Computer Graphics and Computer-Aided Design Theoretical Computer Science HPC Challenge Benchmark Artificial Intelligence Hardware and Architecture Systems engineering Benchmark (computing) Operating system The Internet Performance improvement business Massively parallel computer Software Lead time |
| Zdroj: | Parallel Computing. 25:1459-1476 |
| ISSN: | 0167-8191 |
| DOI: | 10.1016/s0167-8191(99)00059-9 |
| Popis: | Automotive HPC benchmarking needs have not been adequately met by existing benchmarks in the open literature. In this paper we outline what steps should be taken to improve the performance assessment process. HPC performance improvement is essential to meet auto industry needs to reduce lead time between concept design and production. Furthermore, we identify some of the trends which will influence future performance and identify some of the technological limitations. Topics discussed include: creating an auto industry HPC benchmark suite; tradeoffs between modest vs. massive parallelism, use of virtual reality technology; web-based computing including computational grids and utilization of Java for future automotive large-scale computing applications. An extensive list of references is provided including many Internet sites. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect