Measuring network throughput

Abstract: 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.

Abstract: Picture archiving and communication Systems (PACS) for medical imaging have always suffered from band-width limitations, throughput, and proprietary protocols. Commercially available local area networks have been hard pressed to meet the requirements of image transfer in a time consistent with patient-care needs. Recent technologic advances provide potential solutions to these constraints. Asynchronous transfer mode (ATM) provides the aggregate bandwidth and throughput that may be sufficient to satisfy the medical imaging community. Networks using prototype ATM technology have been available and commercial hardware is now becoming available. This report presents initial performance results of an ATM network and its suitability for use in a digital imaging network. Throughput of 10 Mbytes/sec was attained with Transmission Control Protocol/Internet Protocol using commercially available hardware.

Abstract: We propose an analytical method for evaluating the TCP throughput performance in a wireless cellular network environment, where the slotted ALOHA protocol is adopted as a data link layer. By using our method, we show that improving throughput at the data link layer level does not necessarily lead to TCP throughput improvement. Furthermore, we evaluate TCP throughput by considering transmission errors on the radio link. It is shown that when we introduce FEC as an error correcting method, we show that TCP throughput can be improved by selecting an appropriate error correction code with careful consideration of the overhead and the error correction capability according to the quality of the wireless channel.

Abstract: The article proposes the application of genetic algorithms (GA) for optimizing an asynchronous transfer mode (ATM) network throughput based on the virtual path concept. The M/M/1/K queueing model is used to formulate the throughput of ATM networks which we want to optimize. We show the simple coding method in which each chromosome in the genetic model is represented as the part of each ATM* node pair. Also, the concept of past solution is used for periodical changes of traffic, in order to reduce calculation time. The numerical results show that, by our proposed method, we can obtain the optimum configuration of the ATM network which maximizes the network throughput.