{"title":"浅谈为SDN网络提供低成本流量监控","authors":"Haythem Yahyaoui, M. Zhani","doi":"10.1109/CloudNet51028.2020.9335797","DOIUrl":null,"url":null,"abstract":"Traffic monitoring at the flow or even at the packet level has recently gained momentum with the emergence of critical and high-precision network applications like telesurgery, teleportation, and video gaming. However, achieving such fine-grained, continuous, and high-frequency monitoring is particularly challenging as it may result in a high monitoring traffic load on the network consuming significant amounts of bandwidth (referred to as monitoring cost), especially when this traffic has to cross several hops to reach the collecting point. Another challenge is to ensure that the statistics reporting delay ( i.e., the time needed to retrieve the statistics) does not exceed a certain threshold in order to analyze the statistics in a timely manner. In this paper, we address the problem of minimizing the monitoring cost while satisfying the flows' reporting delays by carefully selecting the switch reporting statistics of each flow in the network and taking into consideration the bandwidth available for monitoring and the capacity of the switches. Specifically, we formulate the problem of switch-to-flow selection as an integer linear program and put forward a heuristic algorithm to cope with large-scale instances where the number of flows and switches are significant. Through extensive simulations, we show that the proposed algorithm outperforms two existing monitoring strategies in terms of monitoring cost and reporting delay and provides near-optimal solution with minimal computation time.","PeriodicalId":156419,"journal":{"name":"2020 IEEE 9th International Conference on Cloud Networking (CloudNet)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":"{\"title\":\"On Providing Low-cost Flow Monitoring for SDN Networks\",\"authors\":\"Haythem Yahyaoui, M. Zhani\",\"doi\":\"10.1109/CloudNet51028.2020.9335797\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Traffic monitoring at the flow or even at the packet level has recently gained momentum with the emergence of critical and high-precision network applications like telesurgery, teleportation, and video gaming. However, achieving such fine-grained, continuous, and high-frequency monitoring is particularly challenging as it may result in a high monitoring traffic load on the network consuming significant amounts of bandwidth (referred to as monitoring cost), especially when this traffic has to cross several hops to reach the collecting point. Another challenge is to ensure that the statistics reporting delay ( i.e., the time needed to retrieve the statistics) does not exceed a certain threshold in order to analyze the statistics in a timely manner. In this paper, we address the problem of minimizing the monitoring cost while satisfying the flows' reporting delays by carefully selecting the switch reporting statistics of each flow in the network and taking into consideration the bandwidth available for monitoring and the capacity of the switches. Specifically, we formulate the problem of switch-to-flow selection as an integer linear program and put forward a heuristic algorithm to cope with large-scale instances where the number of flows and switches are significant. Through extensive simulations, we show that the proposed algorithm outperforms two existing monitoring strategies in terms of monitoring cost and reporting delay and provides near-optimal solution with minimal computation time.\",\"PeriodicalId\":156419,\"journal\":{\"name\":\"2020 IEEE 9th International Conference on Cloud Networking (CloudNet)\",\"volume\":\"33 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-11-09\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"6\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2020 IEEE 9th International Conference on Cloud Networking (CloudNet)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/CloudNet51028.2020.9335797\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 IEEE 9th International Conference on Cloud Networking (CloudNet)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CloudNet51028.2020.9335797","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
On Providing Low-cost Flow Monitoring for SDN Networks
Traffic monitoring at the flow or even at the packet level has recently gained momentum with the emergence of critical and high-precision network applications like telesurgery, teleportation, and video gaming. However, achieving such fine-grained, continuous, and high-frequency monitoring is particularly challenging as it may result in a high monitoring traffic load on the network consuming significant amounts of bandwidth (referred to as monitoring cost), especially when this traffic has to cross several hops to reach the collecting point. Another challenge is to ensure that the statistics reporting delay ( i.e., the time needed to retrieve the statistics) does not exceed a certain threshold in order to analyze the statistics in a timely manner. In this paper, we address the problem of minimizing the monitoring cost while satisfying the flows' reporting delays by carefully selecting the switch reporting statistics of each flow in the network and taking into consideration the bandwidth available for monitoring and the capacity of the switches. Specifically, we formulate the problem of switch-to-flow selection as an integer linear program and put forward a heuristic algorithm to cope with large-scale instances where the number of flows and switches are significant. Through extensive simulations, we show that the proposed algorithm outperforms two existing monitoring strategies in terms of monitoring cost and reporting delay and provides near-optimal solution with minimal computation time.