Asynchronous procedures for parallel processing
Virtually all the algorithms now being used for network simulation, tracking, and optimization are synchronous. They cannot be broken into parallel processes for concurrent execution without inserting a number of synchronization points. The first process to reach a synchronization point must wait for the others to catch up. The resulting delays and communication overheads often grow rapidly with the number of processors. When this happens, large numbers of processors can be used with advantage only if the synchronous algorithm is replaced with an asynchronous procedure. Asynchronous procedures tend not to be intuitive, and are difficult to construct and to analyze. To reduce these disadvantages the authors develop useful concepts concerning asynchronous procedures, discuss a model after which they may be patterned, introduce some new convergence results, and present some illustrations of how to construct asynchronous procedures. Finally, they discuss an important application area namely contingency-constrained optimum power flows. 24 references.
- Research Organization:
- Carnegie-Mellon Univ., Pittsburgh, PA
- OSTI ID:
- 5199157
- Journal Information:
- IEEE Trans. Power Appar. Syst.; (United States), Journal Name: IEEE Trans. Power Appar. Syst.; (United States) Vol. 11; ISSN IEPSA
- Country of Publication:
- United States
- Language:
- English
Similar Records
The APRAM: A model for asynchronous parallel computation
Parallel Asynchronous Distributed-Memory Maximal Independent Set Algorithm with Work Ordering