{"title":"一类离散事件系统的大规模并行仿真","authors":"P. Vakili, Levent Mollamustafaoglu, Yu-Chi Ho","doi":"10.1109/FMPC.1992.234886","DOIUrl":null,"url":null,"abstract":"The authors describe a novel approach for the simulation of discrete-event systems on massively parallel computers. In spite of considerable partial parallelism that exists in discrete-event systems, the simulation of a single discrete-event system is intrinsically asynchronous and highly data dependent, and its implementation on massively parallel SIMD (single-instruction multiple-data) computers is particularly difficult. The authors propose to simulate several such systems simultaneously, and in parallel. These are variants of a nominal system with different system parameter values or operating policies. A single clock mechanism is used that drives all the variants in parallel, synchronizes their trajectories, and is the basis of the SIMD implementation. The authors describe a general discrete-event simulator that is developed for implementation on the SIMD MasPar MP-1 computer, and they address the computational issues related to their approach in this context.<<ETX>>","PeriodicalId":117789,"journal":{"name":"[Proceedings 1992] The Fourth Symposium on the Frontiers of Massively Parallel Computation","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1992-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"24","resultStr":"{\"title\":\"Massively parallel simulation of a class of discrete event systems\",\"authors\":\"P. Vakili, Levent Mollamustafaoglu, Yu-Chi Ho\",\"doi\":\"10.1109/FMPC.1992.234886\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The authors describe a novel approach for the simulation of discrete-event systems on massively parallel computers. In spite of considerable partial parallelism that exists in discrete-event systems, the simulation of a single discrete-event system is intrinsically asynchronous and highly data dependent, and its implementation on massively parallel SIMD (single-instruction multiple-data) computers is particularly difficult. The authors propose to simulate several such systems simultaneously, and in parallel. These are variants of a nominal system with different system parameter values or operating policies. A single clock mechanism is used that drives all the variants in parallel, synchronizes their trajectories, and is the basis of the SIMD implementation. The authors describe a general discrete-event simulator that is developed for implementation on the SIMD MasPar MP-1 computer, and they address the computational issues related to their approach in this context.<<ETX>>\",\"PeriodicalId\":117789,\"journal\":{\"name\":\"[Proceedings 1992] The Fourth Symposium on the Frontiers of Massively Parallel Computation\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1992-10-19\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"24\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"[Proceedings 1992] The Fourth Symposium on the Frontiers of Massively Parallel Computation\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/FMPC.1992.234886\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"[Proceedings 1992] The Fourth Symposium on the Frontiers of Massively Parallel Computation","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/FMPC.1992.234886","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Massively parallel simulation of a class of discrete event systems
The authors describe a novel approach for the simulation of discrete-event systems on massively parallel computers. In spite of considerable partial parallelism that exists in discrete-event systems, the simulation of a single discrete-event system is intrinsically asynchronous and highly data dependent, and its implementation on massively parallel SIMD (single-instruction multiple-data) computers is particularly difficult. The authors propose to simulate several such systems simultaneously, and in parallel. These are variants of a nominal system with different system parameter values or operating policies. A single clock mechanism is used that drives all the variants in parallel, synchronizes their trajectories, and is the basis of the SIMD implementation. The authors describe a general discrete-event simulator that is developed for implementation on the SIMD MasPar MP-1 computer, and they address the computational issues related to their approach in this context.<>