Fast heterogeneous binary data interchange
- 7 November 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10975209,p. 90-101
- https://doi.org/10.1109/hcw.2000.843735
Abstract
As distributed applications have become more widely used, they more often need to leverage the computing power of a heterogeneous network of computer architectures. Modern communications libraries provide mechanisms that hide at least some of the complexities of binary data interchange among heterogeneous machines. However, these mechanisms may be cumbersome, requiring that communicating applications agree a-priori on precise message contents, or they may be inefficient, using both "up" and "down" translations for binary data. Finally, the semantics of many packages, particularly those which require applications to manually "pack" and "unpack" messages, result in multiple copies of message data, thereby reducing communication performance. This paper describes PBIO (Portable Binary Input/Output), a novel messaging middleware which offers applications significantly more flexibility in message exchange while providing an efficient implementation that offers high performance.Keywords
This publication has 8 references indexed in Scilit:
- Efficient layering for high speed communication: Fast Messages 2.xPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- DataExchange: High performance communications in distributed laboratoriesParallel Computing, 1998
- Supporting parallel applications on clusters of workstations: The Virtual Communication Machine‐based architectureCluster Computing, 1998
- The Nexus Approach to Integrating Multithreading and CommunicationJournal of Parallel and Distributed Computing, 1996
- VCODEPublished by Association for Computing Machinery (ACM) ,1996
- USCACM SIGCOMM Computer Communication Review, 1994
- Architectural considerations for a new generation of protocolsPublished by Association for Computing Machinery (ACM) ,1990
- Performance of Firefly RPCPublished by Association for Computing Machinery (ACM) ,1989