Abstract
High Performance Fortran Languages such as Vienna Fortran and High Performance Fortran (HPF) allow the programming of massively parallel machines at a relatively high level of abstraction, based on a user-specified distribution of data across the processors of the machine. In this paper we examine the parallelization of two advanced applications - which require irregular data and work distributions - in Vienna Fortran and identify the reasons why the current version of HPF does not provide adequate functionality for formulating efficient parallel versions of these codes.
| Original language | English |
|---|---|
| Pages (from-to) | 401-407 |
| Number of pages | 7 |
| Journal | Future Generation Computer Systems |
| Volume | 11 |
| Issue number | 4-5 |
| DOIs | |
| State | Published - Aug 1995 |
Keywords
- Data distribution
- Data parallel languages
- Fortran language
- Massively parallel machines
- Numerical computation
Fingerprint
Dive into the research topics of 'High Performance Fortran Languages: Advanced applications and their implementation'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver