Determining WWW User's Next Access and Its Application to Pre-fetching

World-Wide Web (WWW) services have grown to levels where significant delays are expected to happen. Techniques like pre-fetching are likely to help users to personalize their needs, reducing their waiting times. However, pre-fetching is only effective if the right documents are identified and if user's move is correctly predicted. Otherwise, pre-fetching will only waste bandwidth. Therefore, it is productive to determine whether a revisit will occur or not, before starting pre-fetching. In this paper we develop two user models that help determining user's next move. One model uses Random Walk approximation and the other is based on Digital Signal Processing techniques. We also give hints on how to use such models with a simple pre-fetching technique that we are developing.

Scalability of Multicast Delivery for Non-sequential Streaming Access

To serve asynchronous requests using multicast, two categories of techniques, stream merging and periodic broadcasting have been proposed. For sequential streaming access where requests are uninterrupted from the beginning to the end of an object, these techniques are highly scalable: the required server bandwidth for stream merging grows logarithmically as request arrival rate, and the required server bandwidth for periodic broadcasting varies logarithmically as the inverse of start-up delay. However, sequential access is inappropriate to model partial requests and client interactivity observed in various streaming access workloads. This paper analytically and experimentally studies the scalability of multicast delivery under a non-sequential access model where requests start at random points in the object. We show that the required server bandwidth for any protocols providing immediate service grows at least as the square root of request arrival rate, and the required server bandwidth for any protocols providing delayed service grows linearly with the inverse of start-up delay. We also investigate the impact of limited client receiving bandwidth on scalability. We optimize practical protocols which provide immediate service to non-sequential requests. The protocols utilize limited client receiving bandwidth, and they are near-optimal in that the required server bandwidth is very close to its lower bound.

Scholarship, Libraries, and Open Access Archiving Initiative

The resolution passed by the BU University Council approving an initiative to establish an archive of the research and scholarship produced by the faculty of the University.

OPEN DOORS AND OPEN MINDS: WHAT FACULTY AUTHORS CAN DO TO ENSURE OPEN ACCESS TO THEIR WORK THROUGH THEIR INSTITUTION

The Internet has brought unparalleled opportunities for expanding availability of research by bringing down economic and physical barriers to sharing. The digitally networked environment promises to democratize access, carry knowledge beyond traditional research niches, accelerate discovery, encourage new and interdisciplinary approaches to ever more complex research challenges, and enable new computational research strategies. However, despite these opportunities for increasing access to knowledge, the prices of scholarly journals have risen sharply over the past two decades, often forcing libraries to cancel subscriptions. Today even the wealthiest institutions cannot afford to sustain all of the journals needed by their faculties and students. To take advantage of the opportunities created by the Internet and to further their mission of creating, preserving, and disseminating knowledge, many academic institutions are taking steps to capture the benefits of more open research sharing. Colleges and universities have built digital repositories to preserve and distribute faculty scholarly articles and other research outputs. Many individual authors have taken steps to retain the rights they need, under copyright law, to allow their work to be made freely available on the Internet and in their institutionâ s repository. And, faculties at some institutions have adopted resolutions endorsing more open access to scholarly articles. Most recently, on February 12, 2008, the Faculty of Arts and Sciences (FAS) at Harvard University took a landmark step. The faculty voted to adopt a policy requiring that faculty authors send an electronic copy of their scholarly articles to the universityâ s digital repository and that faculty authors automatically grant copyright permission to the university to archive and to distribute these articles unless a faculty member has waived the policy for a particular article. Essentially, the faculty voted to make open access to the results of their published journal articles the default policy for the Faculty of Arts and Sciences of Harvard University. As of March 2008, a proposal is also under consideration in the University of California system by which faculty authors would commit routinely to grant copyright permission to the university to make copies of the facultyâ s scholarly work openly accessible over the Internet. Inspired by the example set by the Harvard faculty, this White Paper is addressed to the faculty and administrators of academic institutions who support equitable access to scholarly research and knowledge, and who believe that the institution can play an important role as steward of the scholarly literature produced by its faculty. This paper discusses both the motivation and the process for establishing a binding institutional policy that automatically grants a copyright license from each faculty member to permit deposit of his or her peer-reviewed scholarly articles in institutional repositories, from which the works become available for others to read and cite.

Complying with the NIH Public Access Policy - Copyright Considerations and Options

On January 11, 2008, the National Institutes of Health ('NIH') adopted a revised Public Access Policy for peer-reviewed journal articles reporting research supported in whole or in part by NIH funds. Under the revised policy, the grantee shall ensure that a copy of the author's final manuscript, including any revisions made during the peer review process, be electronically submitted to the National Library of Medicine's PubMed Central ('PMC') archive and that the person submitting the manuscript will designate a time not later than 12 months after publication at which NIH may make the full text of the manuscript publicly accessible in PMC. NIH adopted this policy to implement a new statutory requirement under which: The Director of the National Institutes of Health shall require that all investigators funded by the NIH submit or have submitted for them to the National Library of Medicine's PubMed Central an electronic version of their final, peer-reviewed manuscripts upon acceptance for publication to be made publicly available no later than 12 months after the official date of publication: Provided, That the NIH shall implement the public access policy in a manner consistent with copyright law. This White Paper is written primarily for policymaking staff in universities and other institutional recipients of NIH support responsible for ensuring compliance with the Public Access Policy. The January 11, 2008, Public Access Policy imposes two new compliance mandates. First, the grantee must ensure proper manuscript submission. The version of the article to be submitted is the final version over which the author has control, which must include all revisions made after peer review. The statutory command directs that the manuscript be submitted to PMC 'upon acceptance for publication.' That is, the author's final manuscript should be submitted to PMC at the same time that it is sent to the publisher for final formatting and copy editing. Proper submission is a two-stage process. The electronic manuscript must first be submitted through a process that requires input of additional information concerning the article, the author(s), and the nature of NIH support for the research reported. NIH then formats the manuscript into a uniform, XML-based format used for PMC versions of articles. In the second stage of the submission process, NIH sends a notice to the Principal Investigator requesting that the PMC-formatted version be reviewed and approved. Only after such approval has grantee's manuscript submission obligation been satisfied. Second, the grantee also has a distinct obligation to grant NIH copyright permission to make the manuscript publicly accessible through PMC not later than 12 months after the date of publication. This obligation is connected to manuscript submission because the author, or the person submitting the manuscript on the author's behalf, must have the necessary rights under copyright at the time of submission to give NIH the copyright permission it requires. This White Paper explains and analyzes only the scope of the grantee's copyright-related obligations under the revised Public Access Policy and suggests six options for compliance with that aspect of the grantee's obligation. Time is of the essence for NIH grantees. As a practical matter, the grantee should have a compliance process in place no later than April 7, 2008. More specifically, the new Public Access Policy applies to any article accepted for publication on or after April 7, 2008 if the article arose under (1) an NIH Grant or Cooperative Agreement active in Fiscal Year 2008, (2) direct funding from an NIH Contract signed after April 7, 2008, (3) direct funding from the NIH Intramural Program, or (4) from an NIH employee. In addition, effective May 25, 2008, anyone submitting an application, proposal or progress report to the NIH must include the PMC reference number when citing articles arising from their NIH funded research. (This includes applications submitted to the NIH for the May 25, 2008 and subsequent due dates.) Conceptually, the compliance challenge that the Public Access Policy poses for grantees is easily described. The grantee must depend to some extent upon the author(s) to take the necessary actions to ensure that the grantee is in compliance with the Public Access Policy because the electronic manuscripts and the copyrights in those manuscripts are initially under the control of the author(s). As a result, any compliance option will require an explicit understanding between the author(s) and the grantee about how the manuscript and the copyright in the manuscript are managed. It is useful to conceptually keep separate the grantee's manuscript submission obligation from its copyright permission obligation because the compliance personnel concerned with manuscript management may differ from those responsible for overseeing the author's copyright management. With respect to copyright management, the grantee has the following six options: (1) rely on authors to manage copyright but also to request or to require that these authors take responsibility for amending publication agreements that call for transfer of too many rights to enable the author to grant NIH permission to make the manuscript publicly accessible ('the Public Access License'); (2) take a more active role in assisting authors in negotiating the scope of any copyright transfer to a publisher by (a) providing advice to authors concerning their negotiations or (b) by acting as the author's agent in such negotiations; (3) enter into a side agreement with NIH-funded authors that grants a non-exclusive copyright license to the grantee sufficient to grant NIH the Public Access License; (4) enter into a side agreement with NIH-funded authors that grants a non-exclusive copyright license to the grantee sufficient to grant NIH the Public Access License and also grants a license to the grantee to make certain uses of the article, including posting a copy in the grantee's publicly accessible digital archive or repository and authorizing the article to be used in connection with teaching by university faculty; (5) negotiate a more systematic and comprehensive agreement with the biomedical publishers to ensure either that the publisher has a binding obligation to submit the manuscript and to grant NIH permission to make the manuscript publicly accessible or that the author retains sufficient rights to do so; or (6) instruct NIH-funded authors to submit manuscripts only to journals with binding deposit agreements with NIH or to journals whose copyright agreements permit authors to retain sufficient rights to authorize NIH to make manuscripts publicly accessible.

Services to support BU Authors with Open Access Issues

A working paper written for Boston University Libraries to foster discussion about how to provide better support for BU faculty authors.

Approximately Uniform Random Sampling in Sensor Networks

Recent work in sensor databases has focused extensively on distributed query problems, notably distributed computation of aggregates. Existing methods for computing aggregates broadcast queries to all sensors and use in-network aggregation of responses to minimize messaging costs. In this work, we focus on uniform random sampling across nodes, which can serve both as an alternative building block for aggregation and as an integral component of many other useful randomized algorithms. Prior to our work, the best existing proposals for uniform random sampling of sensors involve contacting all nodes in the network. We propose a practical method which is only approximately uniform, but contacts a number of sensors proportional to the diameter of the network instead of its size. The approximation achieved is tunably close to exact uniform sampling, and only relies on well-known existing primitives, namely geographic routing, distributed computation of Voronoi regions and von Neumann's rejection method. Ultimately, our sampling algorithm has the same worst-case asymptotic cost as routing a point-to-point message, and thus it is asymptotically optimal among request/reply-based sampling methods. We provide experimental results demonstrating the effectiveness of our algorithm on both synthetic and real sensor topologies.

On the Fractal Nature of WWW and Its Application to Cache Modeling

The World Wide Web (WWW or Web) is growing rapidly on the Internet. Web users want fast response time and easy access to a enormous variety of information across the world. Thus, performance is becoming a main issue in the Web. Fractals have been used to study fluctuating phenomena in many different disciplines, from the distribution of galaxies in astronomy to complex physiological control systems. The Web is also a complex, irregular, and random system. In this paper, we look at the document reference pattern at Internet Web servers and use fractal-based models to understand aspects (e.g. caching schemes) that affect the Web performance.

Changes in Web Client Access Patterns: Characteristics and Caching Implications

Understanding the nature of the workloads and system demands created by users of the World Wide Web is crucial to properly designing and provisioning Web services. Previous measurements of Web client workloads have been shown to exhibit a number of characteristic features; however, it is not clear how those features may be changing with time. In this study we compare two measurements of Web client workloads separated in time by three years, both captured from the same computing facility at Boston University. The older dataset, obtained in 1995, is well-known in the research literature and has been the basis for a wide variety of studies. The newer dataset was captured in 1998 and is comparable in size to the older dataset. The new dataset has the drawback that the collection of users measured may no longer be representative of general Web users; however using it has the advantage that many comparisons can be drawn more clearly than would be possible using a new, different source of measurement. Our results fall into two categories. First we compare the statistical and distributional properties of Web requests across the two datasets. This serves to reinforce and deepen our understanding of the characteristic statistical properties of Web client requests. We find that the kinds of distributions that best describe document sizes have not changed between 1995 and 1998, although specific values of the distributional parameters are different. Second, we explore the question of how the observed differences in the properties of Web client requests, particularly the popularity and temporal locality properties, affect the potential for Web file caching in the network. We find that for the computing facility represented by our traces between 1995 and 1998, (1) the benefits of using size-based caching policies have diminished; and (2) the potential for caching requested files in the network has declined.

Differentiated Predictive Fair Service for TCP Flows

The majority of the traffic (bytes) flowing over the Internet today have been attributed to the Transmission Control Protocol (TCP). This strong presence of TCP has recently spurred further investigations into its congestion avoidance mechanism and its effect on the performance of short and long data transfers. At the same time, the rising interest in enhancing Internet services while keeping the implementation cost low has led to several service-differentiation proposals. In such service-differentiation architectures, much of the complexity is placed only in access routers, which classify and mark packets from different flows. Core routers can then allocate enough resources to each class of packets so as to satisfy delivery requirements, such as predictable (consistent) and fair service. In this paper, we investigate the interaction among short and long TCP flows, and how TCP service can be improved by employing a low-cost service-differentiation scheme. Through control-theoretic arguments and extensive simulations, we show the utility of isolating TCP flows into two classes based on their lifetime/size, namely one class of short flows and another of long flows. With such class-based isolation, short and long TCP flows have separate service queues at routers. This protects each class of flows from the other as they possess different characteristics, such as burstiness of arrivals/departures and congestion/sending window dynamics. We show the benefits of isolation, in terms of better predictability and fairness, over traditional shared queueing systems with both tail-drop and Random-Early-Drop (RED) packet dropping policies. The proposed class-based isolation of TCP flows has several advantages: (1) the implementation cost is low since it only requires core routers to maintain per-class (rather than per-flow) state; (2) it promises to be an effective traffic engineering tool for improved predictability and fairness for both short and long TCP flows; and (3) stringent delay requirements of short interactive transfers can be met by increasing the amount of resources allocated to the class of short flows.

Notes on the Effect of Different Access Patterns on the Intensity of Mistreatment in Distributed Caching Groups

In this report, we extend our study of the intensity of mistreatment in distributed caching groups due to state interaction. In our earlier work (published as BUCS-TR-2006-003), we analytically showed how this type of mistreatment may appear under homogeneous demand distributions. We provided a simple setting where mistreatment due to state interaction may occur. According to this setting, one or more "overactive" nodes generate disproportionately more requests than the other nodes. In this report, we extend our experimental evaluation of the intensity of mistreatment to which non-overactive nodes are subjected, when the demand distributions are not homogeneous.