Asynchronous PVM network computing
- Univ. of Texas, San Antonio, TX (United States)
The term asynchronous computing can be used to describe a scheme in which individual processors do useful work during the time normally spent waiting for a synchronized operation to complete. This concept has been widely used for several particular classes of application problems. Asynchronous PVM (APVM) is an enhancement to PVM that allows message passing to be performed asynchronously. It was developed to reduce the performance penalty that was incurred when a program was blocked waiting for messages. The synchronous PVM used a stop and wait protocol with a program sending a request to the PVM daemon and blocking until a response was received. With APVM, a program can send a message to the APVM daemon and continue processing. When the request has been accomplished by the APVM daemon, a response is sent back to the user program. When the message arrives at the user program, an interrupt is generated, and the message is read into the incoming message buffer. When a receiver expects a message, it can periodically check to see if the message has been received. If it has, the message is read. If it has not, processing continues and the input buffer is checked again at a later time.
- OSTI ID:
- 125621
- Report Number(s):
- CONF-950212-; CNN: Grant CCR-9102854; Grant CCR-9400719; Grant DUE-9250265; TRN: 95:005768-0166
- Resource Relation:
- Conference: 7. Society for Industrial and Applied Mathematics (SIAM) conference on parallel processing for scientific computing, San Francisco, CA (United States), 15-17 Feb 1995; Other Information: PBD: 1995; Related Information: Is Part Of Proceedings of the seventh SIAM conference on parallel processing for scientific computing; Bailey, D.H.; Bjorstad, P.E.; Gilbert, J.R. [eds.] [and others]; PB: 894 p.
- Country of Publication:
- United States
- Language:
- English
Similar Records
Portable system for simulating distributed parallel processors on shared memory machines
Adaptive Runtime Features For Distributed Graph Algorithms