Alewife (multiprocessor)
Alewife was a cache coherent multiprocessor developed in the early 1990s by a group led by Anant Agarwal at the Massachusetts Institute of Technology.[1][2] It was based on a network of up to 512 processing nodes, each of which used the Sparcle computer architecture,[3] which was formed by modifying a Sun Microsystems SPARC CPU to include the APRIL techniques for fast context switches.[4]
The Alewife project was one of two predecessors cited by the creators of the popular Beowulf cluster multiprocessor.[5]
References
- Agarwal, A.; Chaiken, D.; Johnson, K.; Kranz, D.; Kubiatowicz, J.; Kurihara, K.; Lim, B. H.; Maa, G.; et al. (1991), The MIT Alewife Machine: A Large-Scale Distributed-Memory Multiprocessor, Tech. Report TM-454, Massachusetts Institute of Technology.
- Agarwal, A.; Bianchini, R.; Chaiken, D.; Chong, F. T.; Johnson, K. L.; Kranz, D.; Kubiatowicz, J. D.; Lim, Beng-Hong; et al. (1999), "The MIT Alewife Machine", Proceedings of the IEEE, 87 (3): 430–444, doi:10.1109/5.747864.
- Agarwal, Anant; Kubiatowicz, John; Kranz, David; Lim, Beng-Hong; Yeung, Donald; D'Souza, Godfrey; Parkin, Mike (1993), "Sparcle: An Evolutionary Processor Design for Large-Scale Multiprocessors", IEEE Micro, 13 (3): 48–61, doi:10.1109/40.216748, S2CID 14678370.
- Agarwal, A.; Lim, B.-H.; Kranz, D.; Kubiatowicz, J. (1990), "APRIL: a processor architecture for multiprocessing", Proc. 17th Annual International Symposium on Computer Architecture (ISCA 1990), pp. 104–114, doi:10.1109/ISCA.1990.134498.
- Sterling, Thomas; Becker, Donald J.; Savarese, Daniel; Dorband, John E.; Ranawake, Udaya A.; Packer, Charles V. (1995), "Beowulf: A parallel workstation for scientific computing", Proc. 24th Int. Conf. Parallel Processing, vol. I, pp. 11–14.
External links
- MIT Alewife Project
- Kubiatowicz, John. "The Alewife-1000 CMMU: Addressing the multiprocessor communications gap" (PDF). hotchips.org. Archived from the original (PDF) on April 23, 2016. Retrieved Feb 5, 2020.
This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.