Pub Date : 2004-05-10DOI: 10.1109/ISPAN.2004.1300453
Tao Li, Yang Yu, Peng Li, J. Xu, Xirong Ma, Yulu Yang
The connection topology of interconnection networks and the routing algorithm often dominate system performance in MPCs. RDT has already been proved to be one of the best interconnection networks used in such systems. It supports a smaller diameter and degree than most at the size of one thousand to tens of thousands nodes. The floating vector routing algorithm is very effective for transferring messages, which also shows good performance. In this paper, a descending vector routing algorithm is proposed. We also proposed a practical torus assignment named RDT(2,2,1)/a and the deadlock-free routing algorithm. RDT(2,2,1)/a mainly focuses on the MPC that has about one thousand nodes. With a simplified simulation and comparison, the results show that RDT(2,2,1)/a has a better performance at the size of one thousand nodes.
{"title":"Practical routing and torus assignment for RDT","authors":"Tao Li, Yang Yu, Peng Li, J. Xu, Xirong Ma, Yulu Yang","doi":"10.1109/ISPAN.2004.1300453","DOIUrl":"https://doi.org/10.1109/ISPAN.2004.1300453","url":null,"abstract":"The connection topology of interconnection networks and the routing algorithm often dominate system performance in MPCs. RDT has already been proved to be one of the best interconnection networks used in such systems. It supports a smaller diameter and degree than most at the size of one thousand to tens of thousands nodes. The floating vector routing algorithm is very effective for transferring messages, which also shows good performance. In this paper, a descending vector routing algorithm is proposed. We also proposed a practical torus assignment named RDT(2,2,1)/a and the deadlock-free routing algorithm. RDT(2,2,1)/a mainly focuses on the MPC that has about one thousand nodes. With a simplified simulation and comparison, the results show that RDT(2,2,1)/a has a better performance at the size of one thousand nodes.","PeriodicalId":198404,"journal":{"name":"7th International Symposium on Parallel Architectures, Algorithms and Networks, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125821208","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2004-05-10DOI: 10.1109/ISPAN.2004.1300448
L. Ni
Summary form only given. In the design of parallel algorithms, especially those involve various collective communication patterns, such as various types of one-to-all, all-to-one, and all-to-all communications, the underlying communication technology plays a vital role in determining the communication efficiency of these algorithms. We have repeatedly investigated the same problem, e.g., broadcast, due to different switching technologies or evaluation criteria. The author first gives an overview of the evolution of interconnection networks from various angles, such as switching technologies, network topologies, distance coverage, network bandwidth, and communication reliability. He then illustrates the difference in the design of some communication algorithms for different networks including the classical point-to-point networks, wormhole networks, system area networks, P2P networks, grid networks, ad hoc networks, and sensor networks. The importance of various performance metrics is also addressed.
{"title":"Evolution of interconnection networks: the impact on the of parallel algorithms","authors":"L. Ni","doi":"10.1109/ISPAN.2004.1300448","DOIUrl":"https://doi.org/10.1109/ISPAN.2004.1300448","url":null,"abstract":"Summary form only given. In the design of parallel algorithms, especially those involve various collective communication patterns, such as various types of one-to-all, all-to-one, and all-to-all communications, the underlying communication technology plays a vital role in determining the communication efficiency of these algorithms. We have repeatedly investigated the same problem, e.g., broadcast, due to different switching technologies or evaluation criteria. The author first gives an overview of the evolution of interconnection networks from various angles, such as switching technologies, network topologies, distance coverage, network bandwidth, and communication reliability. He then illustrates the difference in the design of some communication algorithms for different networks including the classical point-to-point networks, wormhole networks, system area networks, P2P networks, grid networks, ad hoc networks, and sensor networks. The importance of various performance metrics is also addressed.","PeriodicalId":198404,"journal":{"name":"7th International Symposium on Parallel Architectures, Algorithms and Networks, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129928707","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.1109/ispan.2004.1300505
A. Arif Rahman
This paper considers research activities into how much insulation resistance influences data communication via a xDSL system. In Indonesia, insulation resistance has been the only important electrical parameter of copper cable/networks for a long time. Before 2000, knowing whether the cable is good or bad depended on the value of insulation resistance. There is no other measurement for the cable condition. It has been said that if the insulation resistance is more than 1,000 M/spl Omega/ for the existing cable, the cable is good. If less than 1,000 M/spl Omega/, the cable is bad. There is no limitation on insulation resistance value for POTS and data communication. After 2000, several questions arose for electrical parameters for data transmission based on modem systems, especially the x-DSL system: (a) is the insulation resistance an important parameter for an x-DSL system? (b) What is the limitation of insulation resistance value for an x-DSL system? (c) What is the real electrical parameter that has most influence on a x-DSL system? Research needs to analyze the insulation resistance parameter for x-DSL performance and data transmission.
{"title":"Research of capability of existing network for broadband applications: simulation with x-DSL technology on Indonesian condition","authors":"A. Arif Rahman","doi":"10.1109/ispan.2004.1300505","DOIUrl":"https://doi.org/10.1109/ispan.2004.1300505","url":null,"abstract":"This paper considers research activities into how much insulation resistance influences data communication via a xDSL system. In Indonesia, insulation resistance has been the only important electrical parameter of copper cable/networks for a long time. Before 2000, knowing whether the cable is good or bad depended on the value of insulation resistance. There is no other measurement for the cable condition. It has been said that if the insulation resistance is more than 1,000 M/spl Omega/ for the existing cable, the cable is good. If less than 1,000 M/spl Omega/, the cable is bad. There is no limitation on insulation resistance value for POTS and data communication. After 2000, several questions arose for electrical parameters for data transmission based on modem systems, especially the x-DSL system: (a) is the insulation resistance an important parameter for an x-DSL system? (b) What is the limitation of insulation resistance value for an x-DSL system? (c) What is the real electrical parameter that has most influence on a x-DSL system? Research needs to analyze the insulation resistance parameter for x-DSL performance and data transmission.","PeriodicalId":198404,"journal":{"name":"7th International Symposium on Parallel Architectures, Algorithms and Networks, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133651867","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.1109/ispan.2004.1300527
Liu Fang-Ai, Chen Ya-Wen
{"title":"Wavelength assignment of parallel FFT communication pattern on a class of regular optical WDM networks","authors":"Liu Fang-Ai, Chen Ya-Wen","doi":"10.1109/ispan.2004.1300527","DOIUrl":"https://doi.org/10.1109/ispan.2004.1300527","url":null,"abstract":"","PeriodicalId":198404,"journal":{"name":"7th International Symposium on Parallel Architectures, Algorithms and Networks, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116813541","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：