Research On Bandwidth Estimating In IP Networks

With the rapid expansion of the Internet,Internet has becomes a part of our lives that doesn't missing,the number of Internet users become more and more.At the same time,with the complexity and uncertain of the Internet,so many questions has appears,so people call for higher performance of Internet.The network's reliability and stability etc.has been concerned by network's administrator or users. So investigating networks,grappling the running states and usage of resource of networks is significant meanings for improving performance of networks and providing better applying services.Internet consists of many types of network,a structure like a school only has management rights of itself.Because some factors like commerce,it is unshared for internal performance data.So the technique based on monitoring internal nodes is difficult to implement in the "real world".The technique estimating end-to-end performance and behavior isn't restricted for accessing internal source and security strategy of internal nodes.End-to-end networks performance estimating is the need for performance estimating the Internet and application system.With the scale extending of network and application's using more and more, people have a difference request for quality of service,and have a more and more recognition for available bandwidth estimating.An effective means for understanding the character of network's behavior and actual application is estimate end-to-end available bandwidth of a path.What the Internet service provider and users interest in is how can they get the source and service that networks provide rapidly.Bandwidth is an important metric of the state of network,it has a widely application in load balance,congestion control,access control,server's selecting and security etc.So,the veracity of bandwidth measuring is very important to know the state of network and increase the usage of network resource.In a packet network,the terms "bandwidth" or "throughput" often characterize the amount of data that the network can transfer per unit of time.Bandwidth estimation is of interest to users wishing to optimize end-to-end transport performance,overlay network routing,and peer-to-peer file distribution,Techniques for accurate bandwidth estimation are also important for traffic engineering and capacity planning support.Existing bandwidth estimation tools measure one or more of three related metrics:capacity,available bandwidth,and bulk transfer capacity(BTC).Currently available bandwidth estimation tools employ a variety of strategies to measure these metrics.In this thesis,we first summarize the measuring of network and expatiated the meaning of network,measurement techniques and tools. Then we review the recent bandwidth estimation literature focusing on underlying techniques and methodologies as well as open source bandwidth measurement tools.Finally,in this thesis we make experime -nt use Kperf2.0 to prove TCP window size and buffer size is very important for end-to-end bandwidth estimating.Although for TOPP,we improved the technique with two methods,to make TOPP can get orientation of bottleneck link,as well as the capacity.