{"title":"一些调度器实现垃圾率的比例分化","authors":"Y. Lai, A. Chang","doi":"10.1109/ISCC.2005.138","DOIUrl":null,"url":null,"abstract":"The queueing delay or loss rate is usually used as the performance metric for the real-time multimedia applications. In this paper, to provide more suitable quality of services (QoS) requirements of some applications, we propose a new performance metric, junk rate, where junk is the packet which queueing delay exceeds its time threshold. The model of proportional junk rate differentiation is also constructed to provide the predictable and controllable ratio of junk rates for different classes. Three schedulers, namely, proportional junk rate scheduler with infinite memory, proportional junk rate scheduler with memory M, average junk distance scheduler, are proposed to achieve this model. Simulation results show that three schedulers actually yield this model well.","PeriodicalId":315855,"journal":{"name":"10th IEEE Symposium on Computers and Communications (ISCC'05)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Some schedulers to achieve proportional junk rate differentiation\",\"authors\":\"Y. Lai, A. Chang\",\"doi\":\"10.1109/ISCC.2005.138\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The queueing delay or loss rate is usually used as the performance metric for the real-time multimedia applications. In this paper, to provide more suitable quality of services (QoS) requirements of some applications, we propose a new performance metric, junk rate, where junk is the packet which queueing delay exceeds its time threshold. The model of proportional junk rate differentiation is also constructed to provide the predictable and controllable ratio of junk rates for different classes. Three schedulers, namely, proportional junk rate scheduler with infinite memory, proportional junk rate scheduler with memory M, average junk distance scheduler, are proposed to achieve this model. Simulation results show that three schedulers actually yield this model well.\",\"PeriodicalId\":315855,\"journal\":{\"name\":\"10th IEEE Symposium on Computers and Communications (ISCC'05)\",\"volume\":\"25 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2005-06-27\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"10th IEEE Symposium on Computers and Communications (ISCC'05)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ISCC.2005.138\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"10th IEEE Symposium on Computers and Communications (ISCC'05)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISCC.2005.138","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Some schedulers to achieve proportional junk rate differentiation
The queueing delay or loss rate is usually used as the performance metric for the real-time multimedia applications. In this paper, to provide more suitable quality of services (QoS) requirements of some applications, we propose a new performance metric, junk rate, where junk is the packet which queueing delay exceeds its time threshold. The model of proportional junk rate differentiation is also constructed to provide the predictable and controllable ratio of junk rates for different classes. Three schedulers, namely, proportional junk rate scheduler with infinite memory, proportional junk rate scheduler with memory M, average junk distance scheduler, are proposed to achieve this model. Simulation results show that three schedulers actually yield this model well.