Task Assignment in a Distributed System: Improving Performance by Unbalancing Load

We consider the problem of task assignment in a distributed system (such as a distributed Web server) in which task sizes are drawn from a heavy-tailed distribution. Many task assignment algorithms are based on the heuristic that balancing the load at the server hosts will result in optimal performance. We show this conventional wisdom is less true when the task size distribution is heavy-tailed (as is the case for Web file sizes). We introduce a new task assignment policy, called Size Interval Task Assignment with Variable Load (SITA-V). SITA-V purposely operates the server hosts at different loads, and directs smaller tasks to the lighter-loaded hosts. The result is that SITA-V provably decreases the mean task slowdown by significant factors (up to 1000 or more) where the more heavy-tailed the workload, the greater the improvement factor. We evaluate the tradeoff between improvement in slowdown and increase in waiting time in a system using SITA-V, and show conditions under which SITA-V represents a particularly appealing policy. We conclude with a discussion of the use of SITA-V in a distributed Web server, and show that it is attractive because it has a simple implementation which requires no communication from the server hosts back to the task router.

A Framework for Local Anonymity in the Internet

We describe and evaluate options for providing anonymous IP service, argue for the further investigation of local anonymity, and sketch a framework for the implementation of locally anonymous networks.

Robust Identification of Shared Losses Using End-to-End Unicast Probes

ERRATA: We present corrections to Fact 3 and (as a consequence) to Lemma 1 of BUCS Technical Report BUCS-TR-2000-013 (also published in IEEE INCP'2000)[1]. These corrections result in slight changes to the formulae used for the identifications of shared losses, which we quantify.

Program Representation Size in an Intermediate Language with Intersection and Union Types

The CIL compiler for core Standard ML compiles whole programs using a novel typed intermediate language (TIL) with intersection and union types and flow labels on both terms and types. The CIL term representation duplicates portions of the program where intersection types are introduced and union types are eliminated. This duplication makes it easier to represent type information and to introduce customized data representations. However, duplication incurs compile-time space costs that are potentially much greater than are incurred in TILs employing type-level abstraction or quantification. In this paper, we present empirical data on the compile-time space costs of using CIL as an intermediate language. The data shows that these costs can be made tractable by using sufficiently fine-grained flow analyses together with standard hash-consing techniques. The data also suggests that non-duplicating formulations of intersection (and union) types would not achieve significantly better space complexity.

BRITE: Universal Topology Generation from a User's Perspective

Effective engineering of the Internet is predicated upon a detailed understanding of issues such as the large-scale structure of its underlying physical topology, the manner in which it evolves over time, and the way in which its constituent components contribute to its overall function. Unfortunately, developing a deep understanding of these issues has proven to be a challenging task, since it in turn involves solving difficult problems such as mapping the actual topology, characterizing it, and developing models that capture its emergent behavior. Consequently, even though there are a number of topology models, it is an open question as to how representative the topologies they generate are of the actual Internet. Our goal is to produce a topology generation framework which improves the state of the art and is based on design principles which include representativeness, inclusiveness, and interoperability. Representativeness leads to synthetic topologies that accurately reflect many aspects of the actual Internet topology (e.g. hierarchical structure, degree distribution, etc.). Inclusiveness combines the strengths of as many generation models as possible in a single generation tool. Interoperability provides interfaces to widely-used simulation and visualization applications such as ns and SSF. We call such a tool a universal topology generator. In this paper we discuss the design, implementation and usage of the BRITE universal topology generation tool that we have built. We also describe the BRITE Analysis Engine, BRIANA, which is an independent piece of software designed and built upon BRITE design goals of flexibility and extensibility. The purpose of BRIANA is to act as a repository of analysis routines along with a user–friendly interface that allows its use on different topology formats.

Automatic 3D Registration of Lung Surfaces in Computed Tomography Scans

We developed an automated system that registers chest CT scans temporally. Our registration method matches corresponding anatomical landmarks to obtain initial registration parameters. The initial point-to-point registration is then generalized to an iterative surface-to-surface registration method. Our "goodness-of-fit" measure is evaluated at each step in the iterative scheme until the registration performance is sufficient. We applied our method to register the 3D lung surfaces of 11 pairs of chest CT scans and report promising registration performance.

The War Between Mice and Elephants

Recent measurement based studies reveal that most of the Internet connections are short in terms of the amount of traffic they carry (mice), while a small fraction of the connections are carrying a large portion of the traffic (elephants). A careful study of the TCP protocol shows that without help from an Active Queue Management (AQM) policy, short connections tend to lose to long connections in their competition for bandwidth. This is because short connections do not gain detailed knowledge of the network state, and therefore they are doomed to be less competitive due to the conservative nature of the TCP congestion control algorithm. Inspired by the Differentiated Services (Diffserv) architecture, we propose to give preferential treatment to short connections inside the bottleneck queue, so that short connections experience less packet drop rate than long connections. This is done by employing the RIO (RED with In and Out) queue management policy which uses different drop functions for different classes of traffic. Our simulation results show that: (1) in a highly loaded network, preferential treatment is necessary to provide short TCP connections with better response time and fairness without hurting the performance of long TCP connections; (2) the proposed scheme still delivers packets in FIFO manner at each link, thus it maintains statistical multiplexing gain and does not misorder packets; (3) choosing a smaller default initial timeout value for TCP can help enhance the performance of short TCP flows, however not as effectively as our scheme and at the risk of congestion collapse; (4) in the worst case, our proposal works as well as a regular RED scheme, in terms of response time and goodput.

Retrieval by Shape Population: An Index Tree Approach

Based on our previous work in deformable shape model-based object detection, a new method is proposed that uses index trees for organizing shape features to support content-based retrieval applications. In the proposed strategy, different shape feature sets can be used in index trees constructed for object detection and shape similarity comparison respectively. There is a direct correspondence between the two shape feature sets. As a result, application-specific features can be obtained efficiently for shape-based retrieval after object detection. A novel approach is proposed that allows retrieval of images based on the population distribution of deformed shapes in each image. Experiments testing these new approaches have been conducted using an image database that contains blood cell micrographs. The precision vs. recall performance measure shows that our method is superior to previous methods.

Surface Reconstruction from Multiple Views using Rational B-Splines

A method for reconstructing 3D rational B-spline surfaces from multiple views is proposed. The method takes advantage of the projective invariance properties of rational B-splines. Given feature correspondences in multiple views, the 3D surface is reconstructed via a four step framework. First, corresponding features in each view are given an initial surface parameter value (s; t), and a 2D B-spline is fitted in each view. After this initialization, an iterative minimization procedure alternates between updating the 2D B-spline control points and re-estimating each feature's (s; t). Next, a non-linear minimization method is used to upgrade the 2D B-splines to 2D rational B-splines, and obtain a better fit. Finally, a factorization method is used to reconstruct the 3D B-spline surface given 2D B-splines in each view. This surface recovery method can be applied in both the perspective and orthographic case. The orthographic case allows the use of additional constraints in the recovery. Experiments with real and synthetic imagery demonstrate the efficacy of the approach for the orthographic case.

Inference and Labeling of Metric-Induced Network Topologies

The development and deployment of distributed network-aware applications and services over the Internet require the ability to compile and maintain a model of the underlying network resources with respect to (one or more) characteristic properties of interest. To be manageable, such models must be compact, and must enable a representation of properties along temporal, spatial, and measurement resolution dimensions. In this paper, we propose a general framework for the construction of such metric-induced models using end-to-end measurements. We instantiate our approach using one such property, packet loss rates, and present an analytical framework for the characterization of Internet loss topologies. From the perspective of a server the loss topology is a logical tree rooted at the server with clients at its leaves, in which edges represent lossy paths between a pair of internal network nodes. We show how end-to-end unicast packet probing techniques could b e used to (1) infer a loss topology and (2) identify the loss rates of links in an existing loss topology. Correct, efficient inference of loss topology information enables new techniques for aggregate congestion control, QoS admission control, connection scheduling and mirror site selection. We report on simulation, implementation, and Internet deployment results that show the effectiveness of our approach and its robustness in terms of its accuracy and convergence over a wide range of network conditions.

On Class-based Isolation of UDP, Short-lived and Long-lived TCP Flows

The congestion control mechanisms of TCP make it vulnerable in an environment where flows with different congestion-sensitivity compete for scarce resources. With the increasing amount of unresponsive UDP traffic in today's Internet, new mechanisms are needed to enforce fairness in the core of the network. We propose a scalable Diffserv-like architecture, where flows with different characteristics are classified into separate service queues at the routers. Such class-based isolation provides protection so that flows with different characteristics do not negatively impact one another. In this study, we examine different aspects of UDP and TCP interaction and possible gains from segregating UDP and TCP into different classes. We also investigate the utility of further segregating TCP flows into two classes, which are class of short and class of long flows. Results are obtained analytically for both Tail-drop and Random Early Drop (RED) routers. Class-based isolation have the following salient features: (1) better fairness, (2) improved predictability for all kinds of flows, (3) lower transmission delay for delay-sensitive flows, and (4) better control over Quality of Service (QoS) of a particular traffic type.

DNS-based Internet Client Clustering and Characterization

This paper proposes a novel protocol which uses the Internet Domain Name System (DNS) to partition Web clients into disjoint sets, each of which is associated with a single DNS server. We define an L-DNS cluster to be a grouping of Web Clients that use the same Local DNS server to resolve Internet host names. We identify such clusters in real-time using data obtained from a Web Server in conjunction with that server's Authoritative DNS―both instrumented with an implementation of our clustering algorithm. Using these clusters, we perform measurements from four distinct Internet locations. Our results show that L-DNS clustering enables a better estimation of proximity of a Web Client to a Web Server than previously proposed techniques. Thus, in a Content Distribution Network, a DNS-based scheme that redirects a request from a web client to one of many servers based on the client's name server coordinates (e.g., hops/latency/loss-rates between the client and servers) would perform better with our algorithm.

Open Issues on TCP for Mobile Computing

We discuss the design principles of TCP within the context of heterogeneous wired/wireless networks and mobile networking. We identify three shortcomings in TCP's behavior: (i) the protocol's error detection mechanism, which does not distinguish different types of errors and thus does not suffice for heterogeneous wired/wireless environments, (ii) the error recovery, which is not responsive to the distinctive characteristics of wireless networks such as transient or burst errors due to handoffs and fading channels, and (iii) the protocol strategy, which does not control the tradeoff between performance measures such as goodput and energy consumption, and often entails a wasteful effort of retransmission and energy expenditure. We discuss a solution-framework based on selected research proposals and the associated evaluation criteria for the suggested modifications. We highlight an important angle that did not attract the required attention so far: the need for new performance metrics, appropriate for evaluating the impact of protocol strategies on battery-powered devices.

How does TCP generate Pseudo-self-similarity?

Long-range dependence has been observed in many recent Internet traffic measurements. In addition, some recent studies have shown that under certain network conditions, TCP itself can produce traffic that exhibits dependence over limited timescales, even in the absence of higher-level variability. In this paper, we use a simple Markovian model to argue that when the loss rate is relatively high, TCP's adaptive congestion control mechanism indeed generates traffic with OFF periods exhibiting power-law shape over several timescales and thus introduces pseudo-long-range dependence into the overall traffic. Moreover, we observe that more variable initial retransmission timeout values for different packets introduces more variable packet inter-arrival times, which increases the burstiness of the overall traffic. We can thus explain why a single TCP connection can produce a time-series that can be misidentified as self-similar using standard tests.

A Spectrum of TCP-Friendly Window-Based Congestion Control Algorithms

The increased diversity of Internet application requirements has spurred recent interests in transport protocols with flexible transmission controls. In window-based congestion control schemes, increase rules determine how to probe available bandwidth, whereas decrease rules determine how to back off when losses due to congestion are detected. The parameterization of these control rules is done so as to ensure that the resulting protocol is TCP-friendly in terms of the relationship between throughput and loss rate. In this paper, we define a new spectrum of window-based congestion control algorithms that are TCP-friendly as well as TCP-compatible under RED. Contrary to previous memory-less controls, our algorithms utilize history information in their control rules. Our proposed algorithms have two salient features: (1) They enable a wider region of TCP-friendliness, and thus more flexibility in trading off among smoothness, aggressiveness, and responsiveness; and (2) they ensure a faster convergence to fairness under a wide range of system conditions. We demonstrate analytically and through extensive ns simulations the steady-state and transient behaviors of several instances of this new spectrum of algorithms. In particular, SIMD is one instance in which the congestion window is increased super-linearly with time since the detection of the last loss. Compared to recently proposed TCP-friendly AIMD and binomial algorithms, we demonstrate the superiority of SIMD in: (1) adapting to sudden increases in available bandwidth, while maintaining competitive smoothness and responsiveness; and (2) rapidly converging to fairness and efficiency.