Netest: a tool to measure the maximum burst size, available bandwidth and achievable throughput

Distinguishing available bandwidth and achievable throughput is essential for improving network application performance. Achievable throughput is the throughput considering a number of factors such as network protocol, host speed, network path, and TCP buffer space, whereas available bandwidth only considers the network path. Without understanding this difference, trying to improve network application performance is like the "blind men feeling the elephant" problem (J. Gemmill, 2001). We define and differentiate bandwidth and throughput, and discuss which part of each is achievable and which is available. Also, we introduce and discuss a new concept, maximum burst size, that is crucial to obtaining good network performance. A new tool, netest, is introduced which is designed to help users to determine the available bandwidth. It provides information to achieve better throughput while fairly sharing the available bandwidth, thus reducing misuse of the network.

[1]  Robert L. Grossman,et al.  Simple Available Bandwidth Utilization Library for High-Speed Wide Area Networks , 2005, The Journal of Supercomputing.

[2]  Robert L. Grossman,et al.  PSockets: The Case for Application-level Network Striping for Data Intensive Applications using High Speed Wide Area Networks , 2000, ACM/IEEE SC 2000 Conference (SC'00).

[3]  George Yang,et al.  Network Characterization Service (NCS) , 2001, Proceedings 10th IEEE International Symposium on High Performance Distributed Computing.

[4]  Guojun Jin,et al.  Algorithms and Requirements for Measuring Network Bandwidth , 2002 .

[5]  Brian Tierney,et al.  An infrastructure for passive network monitoring of application data streams , 2003 .

[6]  Ian T. Foster,et al.  Secure, Efficient Data Transport and Replica Management for High-Performance Data-Intensive Computing , 2001, 2001 Eighteenth IEEE Symposium on Mass Storage Systems and Technologies.

[7]  Sally Floyd,et al.  HighSpeed TCP for Large Congestion Windows , 2003, RFC.

[8]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[9]  Ravi P. Agarwal,et al.  On nonoscillatory solutions of differential inclusions , 2002 .

[10]  Karsten P. Ulland,et al.  Vii. References , 2022 .

[11]  Brian D. Noble,et al.  The end-to-end performance effects of parallel TCP sockets on a lossy wide-area network , 2002, Proceedings 16th International Parallel and Distributed Processing Symposium.

[12]  Brian Tierney,et al.  TCP Tuning Guide for Distributed Application on Wide Area Networks , 2001, login Usenix Mag..