K. K, Ankit Rathore, V. B, P. D. Shenoy, V. R., Vignesh T Prabhu
{"title":"Optimal Token Bucket Refilling for Tor network","authors":"K. K, Ankit Rathore, V. B, P. D. Shenoy, V. R., Vignesh T Prabhu","doi":"10.1109/CONECCT.2018.8482389","DOIUrl":null,"url":null,"abstract":"The infrastructure of the Tor network is completely dependent on the relays voluntarily contributing their bandwidth for the community. Although the network provides a very high level of anonymity to the end user, it is designed such that the anonymity comes at the cost of very high latency. Hence, improving the performance of Tor network is always a very important aspect of the Tor project. Token Bucket algorithm is used in relays to ensure that the packets sent by clients are not lost and the bandwidth limits configured by the relay administrator is respected. In Token bucket algorithm, an important aspect is the Token Bucket Refill Interval which decides on how often the tokens are added to the bucket. In the Tor network, a default value of 100ms is used. This value is vindictive for a network where the majority of the clients are web clients with small download size. But as the bulk clients increase, it is important to understand how the value of Token Bucket Refill Interval impacts the performance of the network. In this paper, we continue on the previous work done to find the optimum value for Token Bucket Refill Interval for different scenarios of Tor network. Based on the simulations, we have found that download size of the clients is inversely related to the Token Bucket Refill Interval for the overall performance of the network. A larger value of Token Bucket Refill Interval gives a much better performance in Tor network when download size is small. However, as the download size increases, lower values of Token Bucket Refill Interval show better performance in Tor network.","PeriodicalId":430389,"journal":{"name":"2018 IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-03-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CONECCT.2018.8482389","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
The infrastructure of the Tor network is completely dependent on the relays voluntarily contributing their bandwidth for the community. Although the network provides a very high level of anonymity to the end user, it is designed such that the anonymity comes at the cost of very high latency. Hence, improving the performance of Tor network is always a very important aspect of the Tor project. Token Bucket algorithm is used in relays to ensure that the packets sent by clients are not lost and the bandwidth limits configured by the relay administrator is respected. In Token bucket algorithm, an important aspect is the Token Bucket Refill Interval which decides on how often the tokens are added to the bucket. In the Tor network, a default value of 100ms is used. This value is vindictive for a network where the majority of the clients are web clients with small download size. But as the bulk clients increase, it is important to understand how the value of Token Bucket Refill Interval impacts the performance of the network. In this paper, we continue on the previous work done to find the optimum value for Token Bucket Refill Interval for different scenarios of Tor network. Based on the simulations, we have found that download size of the clients is inversely related to the Token Bucket Refill Interval for the overall performance of the network. A larger value of Token Bucket Refill Interval gives a much better performance in Tor network when download size is small. However, as the download size increases, lower values of Token Bucket Refill Interval show better performance in Tor network.