Parallel Nonlinear Optimization: Limitations, Opportunities, and Challenges
Abstract
The availability and power of parallel computers is having a significant impact on how large-scale problems are solved in all areas of numerical computation, and is likely to have an even larger impact in the future. This paper attempts to give some indication of how the consideration of parallel computation is affecting, and is likely to affect, the field of nonlinear optimization. It does not attempt to survey the research that has been done in parallel nonlinear optimization. Rather it presents a set of examples, mainly from our own research, that is intended to illustrate many of the limitations, opportunities, and challenges inherent in incorporating parallelism into the field of nonlinear optimization. These examples include parallel methods for small to medium size unconstrained optimization problems, parallel methods for large block bordered systems of nonlinear equations, and parallel methods for both small and large-scale global optimization problems. Our overall conclusions are mixed. For generic, small to medium size problems, the consideration of parallelism does not appear to be leading to major algorithmic innovations. For many classes of large-scale problems, however, the consideration of parallelism appears to be creating opportunities for the development of interesting new methods that may be advantageous for parallel and possibly even sequential computation. In addition, a number of large-scale parallel optimization algorithms exhibit irregular coarse-grain structure, which leads to interesting computer science challenges in their implementation.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1994
- Accession Number
- ADA461917
Entities
People
- Robert B. Schnabel
Organizations
- University of Colorado Boulder