Cost-Effective Parallel Computing
Abstract
Many academic papers imply that parallel computing is only worthwhile when applications achieve nearly linear speedup (i.e. execute nearly p times faster on p processors). This note shows that parallel computing is cost-effective whenever speedup exceeds costup-the parallel system cost divided by uniprocessor cost. Furthermore, when applications have large memory requirements (e.g. 512 megabytes) the costup--and hence speedup necessary to be cost-effective--can be much less than linear.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1994
- Accession Number
- ADA395472
Entities
People
- David A Wood
- Mark D. Hill
Organizations
- University of Wisconsin Madison Department of Computer Science