{"title":"Quality of service provision in noncooperative networks: heterogenous preferences, multi-dimensional QoS vectors, and burstiness","authors":"Kihong Park, Meera Sitharam, Shaogang Chen","doi":"10.1145/288994.289022","DOIUrl":null,"url":null,"abstract":"This paper studies the quality of service (QoS) provision problem in noncooperative networks where applications or users are selfish and routers implement generalized processor sharing (GPS)-based packet scheduling. First, we formulate a model of QoS provision in nonc~ operative networks where users are given the freedom to choose both the service classes and traffic volume allocated, and heterogeneousQoS preferences are captured by individual utility functions. We present a comprehensive analysis of the noncooperative multi-class QoS provision game, giving a complete characterization of Nash equilibria and their existence criteria, and show under what conditions they are Pareto and system optimal We show that, in general, Nash equilibria need not exist, and when they do exist, they need not be Pareto nor system optimalHowever, we show that for certain “resource-plentiful” systems, the world indeed can be nice with Nash equilibria, Pareto optima, and system optima collapsing into a single class. Second, we study the problem of fac-ditating efiective QoS in systems with multi-dimensional QoS vectors containing ●Supporteclinpart by NSF grants.4NI-9714707, ESS-9806741, and grants from PRF and Sprint. tSupported in part by NSF grant CCR-9409809Present address: Dept. of Computer and InformationScienceand Engineering,Universityof Florida, Gainesville,FL 32611 $Supported in part by NSF grant -4NI-9714iOiPennission10makedigitalorhardcopiesof allorpartof thisworkfor personal orclassroomuseisgrantedwithoutfm pro~idedthatcopies arenotmadeordistriiuledforprotitorcomrnemial adwomgeandthat copiesbearthisnoticeandthefullcitationonthefirstpage.To copy otherwise, torepublish, topostonsen,ersortoredism-bute tolists, requires priorspecificpermissionand~or a fea ICE 9S CharlestonSC LISA Copyright ACM 19981-581 13-076-7/98/10...S5.00 Chen$ both meanand burstiness-related QoS measures. We extend the game-theoretic analysis to multi-dimensional QoS vector games and show under what conditions the aforementioned results carry over. Motivated by the same context, we study the impact of burstiness under multiple QoS measures on the properties of the induced QoS levels rendered by the service classes in the system. We show that under bursty traffic conditions, it is, in general, impossible for a service class to deliver quality of service superior in both meanand burstiness-related QoS measures.","PeriodicalId":147821,"journal":{"name":"International Conference on Information and Computation Economies","volume":"42 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1998-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"54","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Conference on Information and Computation Economies","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/288994.289022","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 54
Abstract
This paper studies the quality of service (QoS) provision problem in noncooperative networks where applications or users are selfish and routers implement generalized processor sharing (GPS)-based packet scheduling. First, we formulate a model of QoS provision in nonc~ operative networks where users are given the freedom to choose both the service classes and traffic volume allocated, and heterogeneousQoS preferences are captured by individual utility functions. We present a comprehensive analysis of the noncooperative multi-class QoS provision game, giving a complete characterization of Nash equilibria and their existence criteria, and show under what conditions they are Pareto and system optimal We show that, in general, Nash equilibria need not exist, and when they do exist, they need not be Pareto nor system optimalHowever, we show that for certain “resource-plentiful” systems, the world indeed can be nice with Nash equilibria, Pareto optima, and system optima collapsing into a single class. Second, we study the problem of fac-ditating efiective QoS in systems with multi-dimensional QoS vectors containing ●Supporteclinpart by NSF grants.4NI-9714707, ESS-9806741, and grants from PRF and Sprint. tSupported in part by NSF grant CCR-9409809Present address: Dept. of Computer and InformationScienceand Engineering,Universityof Florida, Gainesville,FL 32611 $Supported in part by NSF grant -4NI-9714iOiPennission10makedigitalorhardcopiesof allorpartof thisworkfor personal orclassroomuseisgrantedwithoutfm pro~idedthatcopies arenotmadeordistriiuledforprotitorcomrnemial adwomgeandthat copiesbearthisnoticeandthefullcitationonthefirstpage.To copy otherwise, torepublish, topostonsen,ersortoredism-bute tolists, requires priorspecificpermissionand~or a fea ICE 9S CharlestonSC LISA Copyright ACM 19981-581 13-076-7/98/10...S5.00 Chen$ both meanand burstiness-related QoS measures. We extend the game-theoretic analysis to multi-dimensional QoS vector games and show under what conditions the aforementioned results carry over. Motivated by the same context, we study the impact of burstiness under multiple QoS measures on the properties of the induced QoS levels rendered by the service classes in the system. We show that under bursty traffic conditions, it is, in general, impossible for a service class to deliver quality of service superior in both meanand burstiness-related QoS measures.