{"title":"XBitTorrent: Localized latency-driven bittorrent","authors":"James Patrick A. Acang, Arlene A. Satuito","doi":"10.1109/HNICEM.2014.7016202","DOIUrl":null,"url":null,"abstract":"Peer-to-Peer systems was developed to compensate limitations of the internet and to reduce traffic among the nodes. In this way data transmission will be quicker and data distribution will be faster. One of the famous P2P distribution systems is BitTorrent. BitTorrent has gained reputation by its remarkable performance in distributing large files over the internet. Since BitTorrent and other P2P systems were usually established at the user level, they do not perceive information from the network. We investigated and demonstrated here an approach of improving the BitTorrent by integrating network information in the protocol. We consider here latency. This can be very useful for network applications since it provides a way for higher level protocols to infer states and problems within the network. High latency values would mean congestion and instability of a node, in which we can perform decisions to handle traffic effectively in the network. We integrated latency as an indicator in the selection of neighbors in the peerset. Also, we conducted experiments with several network scenarios to validate our approach. Results show that with our method, 86% reduction in the download time can be achieved, thus indicating highly significant improvement in performance.","PeriodicalId":309548,"journal":{"name":"2014 International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management (HNICEM)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2014-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2014 International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management (HNICEM)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/HNICEM.2014.7016202","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
Peer-to-Peer systems was developed to compensate limitations of the internet and to reduce traffic among the nodes. In this way data transmission will be quicker and data distribution will be faster. One of the famous P2P distribution systems is BitTorrent. BitTorrent has gained reputation by its remarkable performance in distributing large files over the internet. Since BitTorrent and other P2P systems were usually established at the user level, they do not perceive information from the network. We investigated and demonstrated here an approach of improving the BitTorrent by integrating network information in the protocol. We consider here latency. This can be very useful for network applications since it provides a way for higher level protocols to infer states and problems within the network. High latency values would mean congestion and instability of a node, in which we can perform decisions to handle traffic effectively in the network. We integrated latency as an indicator in the selection of neighbors in the peerset. Also, we conducted experiments with several network scenarios to validate our approach. Results show that with our method, 86% reduction in the download time can be achieved, thus indicating highly significant improvement in performance.