We analyze 2.5 million TCP connections that occurred during 14 wide-area traffic traces. The traces were gathered at five "stub" networks and two internetwork gateways, providing a diverse look at wide-area traffic. We derive analytic models describing the random variables associated with telnet, nntp, smtp, and ftp connections, and present a methodology for comparing the effectiveness of the analytic models with empirical models such as tcplib [DJ91]. Overall we find that the analytic models provide good descriptions, generally modeling the various distributions as well as empirical models and in some cases better.

The behavior of the TCP protocol in simple situations is well-understood, but when multiple connections share a set of network resources the protocol can exhibit surprising phenomena. Earlier studies have identified several such phenomena, and have analyzed them using simulation or observation of contrived situations. This paper shows how, by analyzing traces of a busy segment of the Internet, it is possible to observe these phenomena in "real life" and measure both their frequency and their effects on performance. A TCP implementation might use similar techniques to support rate-based congestion control.

Dynamics of Internet load are investigated using statistics of round-trip delays, packet losses and out-of-order sequence of acknowledgments. Several segments of the Internet are studied. They include a regional network (the Jon yon Neumann Center Network), a segment of the NSFNet backbone and a cross-country network consisting of regional and backbone segments.

This paper describes tcplib, a workload or source library for network simulation. This paper motivates the need for tools like tcplib and discusses how to incorporate it into a network simulator. Tcplib is available by anonymous ftp 1. 1.

Over the last decade Internet has grown by orders of magnitude in size. Many of the protocols that were designed several years ago are still in use. It is not clear if the assumptions made in the design of control schemes still hold, particularly when we consider end-to-end behavior of paths in the network, today. This paper describes a simple experiment designed to capture end-to-end behavior of the Internet. The measurements indicate that the IP level service provided in the network yields high losses, duplicates and reorderings of packets. In addition, the round-trip transit delay varies significantly. These measurements indicate that the network may have several problems which still need to be analysed in order to improve the efficiency of protocols and control mechanisms that it uses.

For performance and economic reasons, ATM networks must efficiently support the Internet family of protocols. We compare the transmission efficiency achieved by a range of ATM-related protocols when transporting TCP and UDP wide-area traffic. We also calculate the efficiency effects of several non-standard compression techniques, and discuss how to implement these techniques cheaply. To assure an accurate workload characterization, we drive these calculations with millions of wide-area packet lengths measured on the current Internet. We find that networks using standard ATM procedures are inefficient in carrying wide-area data traffic - depending on the protocols used, efficiency as seen by an application program ranges between 40 and 53%. Moreover, due to interaction between TCP-IP datagram lengths and ATM cell padding, efficiency responds abruptly to changes in certain protocol parameters - for example, a 4byte increase in ATM cell payload size can yield a 10% increase in efficiency....

Many application-level traffic streams, or conversations, are multiplexed at the points where local-area networks meet the wide-area portion of an internetwork. Multiplexing policies and mechanisms acting at these points should provide good performance to each conversation, allocate network resources fairly among conversations, and make efficient use of network resources. In order to characterize wide-area network traffic, we have analyzed traces from four Internet sites. We identify characteristics common to all conversations of each major type of traffic, and find that these characteristics are stable across time and geographic site. Our results contradict many prevalent beliefs. For example, previous simulation models of wide-area traffic have assumed bulk transfers ranging from 80 Kilobytes to 2 Megabytes of data. In contrast, we find that up to 90% of all bulk transfers involve 10 Kilobytes or less. This and other findings may affect results of previous studies and should be taken into account in future models of wide-area traffic. We derive from our traces a new workload model for driving simulations of wide-area internetworks. It generates traffic for individual conversations of each major type of traffic. The model accurately and efficiently reproduces behavior specific to each traffic type by sampling measured probability distributions through the inverse transform method. Our model is valid for network conditions other than those prevalent during the measurements because it samples only network-independent traffic characteristics. We also describe a new wide-area internetwork simulator that includes both our workload model and realistic models of network components. We then present a simulation study of policies for multiplexing datagrams over virtual circu...

The increasing transmission speed and the growing size of computer networks are creating new challenges for network routing. Increased transmission speed enables new, integrated, applications that require service guarantees from the network. In order to provide these guarantees, routes with sufficient resources must be located. The ability of a particular route to support performancesensitive applications will vary over time as the utilization of the resources along that route changes. Therefore, routing performance can improve with the distribution of dynamic information about the state of the network. At the same time, the increasing size of networks argues against the global distribution of dynamic routing information and against requiring consistent dynamic routing information at all nodes in the network. These conflicting constraints motivate us to consider schemes that allow partial distribution of dynamic routing information. In this paper we present a protocol for the distribu...

Abstract not found

Physically, the Internet is a collection of packet switching networks interconnected by gateways along with the TCP/IP (Transmission Control Protocol/Internet Protocol) protocol suite that allows them to function logically as a single large virtual network. Logically, the Internet is a vast library of information and services being exchanged among a worldwide user community. It provides electronic mail, file transfer capabilities, remote login, network news, and other services on top of both reliable, full duplex stream delivery and unreliable, connectionless packet delivery services. Internet traffic measurement allows a statistical analysis of the international information flow. Such analysis is the first step to provide a snapshot of real workloads on wide area networks. This thesis presents results from measurements of wide area network TCP conversations between the Campus Ethernet at the University of Saskatchewan and the Internet in April 1994. A total of 103,016 TCP conversati...