background transfers transfers of data that humans are not waiting for to improve availability, reliability, latency or consistency. However, given the rapid fluctuations of available network bandwidth and changing resource costs due to technology trends, hand tuning the aggressiveness of background transfers risks (1) complicating applications, (2) being too aggressive and interfering with other applications, and (3) being too timid and not gaining the benefits of background transfers. Our goal is for the operating system to manage network resources in order to provide a simple abstraction of near zero-cost background transfers. Our system, TCP Nice, can provably bound the interference inflicted by background flows on foreground flows in a restricted network model. And our microbenchmarks and case study applications suggest that in practice it interferes little with foreground flows, reaps a large fraction of spare network bandwidth, and simplifies application construction and deployment. For example, in our prefetching case study application, aggressive prefetching improves demand performance by a factor of three when Nice manages resources; but the same prefetching hurts demand performance by a factor of six under standard network congestion control.

The high tech bubble was inflated by myths of astronomical Internet traffic growth rates. Yet although these myths were false, Internet traffic was increasing very rapidly, close to doubling each year since 1997. Moreover, it continues growing close to this rate. This rapid growth reflects a poorly understood combination of many feedback loops operating on different time scales. Evidence about past and current growth rates and their sources is presented, together with speculations about the future. The expected rapid but not astronomical growth of Internet traffic is likely to have important implications for networking technologies that are deployed and for industry structure. Backbone transport is likely to remain a commodity and be provided as a single high quality service. It is probable that backbone revenues will stay low, as the complexity, cost, and revenue and profit opportunities continue to migrate towards the edges of the network.

We present NPS, a novel non-intrusive web prefetching system that (1) utilizes only spare resources to avoid interference between prefetch and demand requests at the server as well as in the network , and (2) is deployable without any modifications to servers, browsers, network or the HTTP protocol. NPS's self-tuning architecture eliminates the need for traditional "thresholds" or magic numbers typically used to limit interference caused by prefetching, thereby allowing applications to improve bene ts and reduce the risk of aggressive prefetching.

This paper introduces a metric for measuring backbone traffic variability that is grounded on simple but powerful traffic theory. What sets this metric apart, however, is that we present a method for practical implementation of the metric using widely available SNMP traffic measurements. In addition to simulations, we use a large set of SNMP data from an operational IP network on the order of 1000 nodes to test our methods. We also delve into the degree and sources of variability in real backbone traffic, providing insight into the true nature of traffic variability

The Cactus software is representative for a whole class of scientific applications; typically those that are tightly coupled, have regular space decomposition, and huge memory and processor time requirements.

Traditional journals, even those available electronically, are changing slowly. However, there is rapid evolution in scholarly communication. Usage is moving to electronic formats. In some areas, it appears that electronic versions of papers are being read about as often as the printed journal versions. Although there are serious difficulties in comparing figures from different media, the growth rates in usage of electronic scholarly information are sufficiently high that if they continue for a few years, there will be no doubt that print versions will be eclipsed. Further, much of the electronic information that is accessed is outside the formal scholarly publication process. There is also vigorous growth in forms of electronic communication that take advantage of the unique capabilities of the Web, and which simply do not fit into the traditional journal publishing format. This paper presents some statistics on usage of print and electronic information. It also discusses some preliminary evidence about the changing patterns of usage. It appears that much of the online usage comes from new readers (esoteric research papers assigned in undergraduate classes, for exam-ple) and often from places that do not have access to print journals. Also, the reactions to even slight barriers to usage suggest that even high quality scholarly papers are not irreplaceable. Readers are faced with a “river of knowledge ” that allows them to select among a multitude of sources, and to find near substitutes when necessary. To stay relevant, scholars, publishers, and librarians will have to make even larger efforts to make their material easily accessible. 1.

The Cactus software package is suitable for a class of scientific applications that are tightly coupled, have regular space decompositions, and involve huge memory and processor time requirements. Cactus has proved to be a valuable tool for astrophysicists, who first initiated its development. However, today's fastest supercomputers are not powerful enough to perform realistic large-scale astrophysics simulations with Cactus. Instead, we must turn to innovative resource environments -- in particular, computational Grids -- to satisfy this need for computational power. Our paper addresses issues related to the execution of applications such as Cactus in Grid environments. We focus on two types of Grids: a set of geographically distributed supercomputers and a collection of one million Internet-connected workstations.

Understanding the culture of operational networking can help to illuminate the question of why QoS has floundered. Network administrators have a well-founded aversion to complexity, in part because they experience failures attributable to design complexity on a regular basis. I argue that IP multicast defines a functional limit-case for deployable complexity in today’s Internet. That limit is relevant to the deployment of QoS, since many flavors of QoS entail equal or greater complexity. The notion of a functional constraint on complexity draws attention to the economic, historical, and institutional forces which influence the fate of networking technologies. QoS will not be compelling for most network administrators until its design takes account of these forces. Categories and Subject Descriptors C.2.1 [Computer-Communication Networks]: Network

The cooperative anarchy of the global Internet defies easy characterization or measurement of its behavior. Fortunately, lack of global understanding has not stalled the advancement of network engineering technologies that enable and support Internet growth - for the moment. Both Internet users and providers can benefit from measurements that detect and isolate Internet problems, and identify traffic bottlenecks. Yet it is neither practical nor particularly effective to monitor and measure every single link. Common sense supports the establishment of a measurement infrastructure strategically designed to yield maximal Internet coverage at reasonable cost. However, while individual ISPs monitor their own infrastructure and quality of service, business and other practical concerns often prevent sharing of such information. We survey existing public and mission-specific Internet measurement infrastructures, comparing them using a variety of criteria. Community awareness of similar measurement activities will hopefully facilitate opportunities for collaboration, leveraging experiences and investment across groups. Cataloguing these sources of Internet measurements also provides operations researchers with places to seek topology, workload, performance, and routing data that can help them refine metrics and methodologies for effective management of the global Internet.

In this dissertation we posit a coherent area of research called Internet Algorithmics. We study a number of different algorithmic problems which are motivated by or are relevant to the various processes which activate the Internet or are activated by it.