Robust, High-Speed Network Design for Large-Scale Multiprocessing
As multiprocessor system size scales upward, two important aspects of multiprocessor systems will generally get worse rather than better: (1) interprocessor communication latency will increase and (2) the probability that some component in the system will fail will increase. These problems ca...
| Main Author: | |
|---|---|
| Language: | en_US |
| Published: |
2004
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/6790 |
| _version_ | 1811082439068483584 |
|---|---|
| author | DeHon, Andre |
| author_facet | DeHon, Andre |
| author_sort | DeHon, Andre |
| collection | MIT |
| description | As multiprocessor system size scales upward, two important aspects of multiprocessor systems will generally get worse rather than better: (1) interprocessor communication latency will increase and (2) the probability that some component in the system will fail will increase. These problems can prevent us from realizing the potential benefits of large-scale multiprocessing. In this report we consider the problem of designing networks which simultaneously minimize communication latency while maximizing fault tolerance. Using a synergy of techniques including connection topologies, routing protocols, signalling techniques, and packaging technologies we assemble integrated, system-level solutions to this network design problem. |
| first_indexed | 2024-09-23T12:03:26Z |
| id | mit-1721.1/6790 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T12:03:26Z |
| publishDate | 2004 |
| record_format | dspace |
| spelling | mit-1721.1/67902019-04-12T08:32:18Z Robust, High-Speed Network Design for Large-Scale Multiprocessing DeHon, Andre fault tolerance multipath network latency smultiprocessing transit As multiprocessor system size scales upward, two important aspects of multiprocessor systems will generally get worse rather than better: (1) interprocessor communication latency will increase and (2) the probability that some component in the system will fail will increase. These problems can prevent us from realizing the potential benefits of large-scale multiprocessing. In this report we consider the problem of designing networks which simultaneously minimize communication latency while maximizing fault tolerance. Using a synergy of techniques including connection topologies, routing protocols, signalling techniques, and packaging technologies we assemble integrated, system-level solutions to this network design problem. 2004-10-20T19:55:06Z 2004-10-20T19:55:06Z 1993-09-01 AITR-1445 http://hdl.handle.net/1721.1/6790 en_US AITR-1445 205 p. 1915331 bytes 10054055 bytes application/octet-stream application/pdf application/octet-stream application/pdf |
| spellingShingle | fault tolerance multipath network latency smultiprocessing transit DeHon, Andre Robust, High-Speed Network Design for Large-Scale Multiprocessing |
| title | Robust, High-Speed Network Design for Large-Scale Multiprocessing |
| title_full | Robust, High-Speed Network Design for Large-Scale Multiprocessing |
| title_fullStr | Robust, High-Speed Network Design for Large-Scale Multiprocessing |
| title_full_unstemmed | Robust, High-Speed Network Design for Large-Scale Multiprocessing |
| title_short | Robust, High-Speed Network Design for Large-Scale Multiprocessing |
| title_sort | robust high speed network design for large scale multiprocessing |
| topic | fault tolerance multipath network latency smultiprocessing transit |
| url | http://hdl.handle.net/1721.1/6790 |
| work_keys_str_mv | AT dehonandre robusthighspeednetworkdesignforlargescalemultiprocessing |