Showing papers on "Distributed algorithm published in 2000"

Topology control of multihop wireless networks using transmit power adjustment

Abstract: We consider the problem of adjusting the transmit powers of nodes in a multihop wireless network (also called an ad hoc network) to create a desired topology. We formulate it as a constrained optimization problem with two constraints-connectivity and biconnectivity, and one optimization objective-maximum power used. We present two centralized algorithms for use in static networks, and prove their optimality. For mobile networks, we present two distributed heuristics that adaptively adjust node transmit powers in response to topological changes and attempt to maintain a connected topology using minimum power. We analyze the throughput, delay, and power consumption of our algorithms using a prototype software implementation, an emulation of a power-controllable radio, and a detailed channel model. Our results show that the performance of multihop wireless networks in practice can be substantially increased with topology control.

The state of the art in distributed query processing

Abstract: Distributed data processing is becoming a reality. Businesses want to do it for many reasons, and they often must do it in order to stay competitive. While much of the infrastructure for distributed data processing is already there (e.g., modern network technology), a number of issues make distributed data processing still a complex undertaking: (1) distributed systems can become very large, involving thousands of heterogeneous sites including PCs and mainframe server machines; (2) the state of a distributed system changes rapidly because the load of sites varies over time and new sites are added to the system; (3) legacy systems need to be integrated—such legacy systems usually have not been designed for distributed data processing and now need to interact with other (modern) systems in a distributed environment. This paper presents the state of the art of query processing for distributed database and information systems. The paper presents the “textbook” architecture for distributed query processing and a series of techniques that are particularly useful for distributed database systems. These techniques include special join techniques, techniques to exploit intraquery paralleli sm, techniques to reduce communication costs, and techniques to exploit caching and replication of data. Furthermore, the paper discusses different kinds of distributed systems such as client-server, middleware (multitier), and heterogeneous database systems, and shows how query processing works in these systems.

A game theoretic framework for bandwidth allocation and pricing in broadband networks

Abstract: In this paper, we present a game theoretic framework for bandwidth allocation for elastic services in high-speed networks. The framework is based on the idea of the Nash bargaining solution from cooperative game theory, which not only provides the rate settings of users that are Pareto optimal from the point of view of the whole system, but are also consistent with the fairness axioms of game theory. We first consider the centralized problem and then show that this procedure can be decentralized so that greedy optimization by users yields the system optimal bandwidth allocations. We propose a distributed algorithm for implementing the optimal and fair bandwidth allocation and provide conditions for its convergence. The paper concludes with the pricing of elastic connections based on users' bandwidth requirements and users' budget. We show that the above bargaining framework can be used to characterize a rate allocation and a pricing policy which takes into account users' budget in a fair way and such that the total network revenue is maximized.

Power control for wireless data

Abstract: With cellular phones mass-market consumer items, the next frontier is mobile multimedia communications. This situation raises the question of how to perform power control for information sources other than voice. To explore this issue, we use the concepts and mathematics of microeconomics and game theory. In this context, the quality of service of a telephone call is referred to as the "utility" and the distributed power control problem for a CDMA telephone is a "noncooperative game." The power control algorithm corresponds to a strategy that has a locally optimum operating point referred to as a "Nash equilibrium." The telephone power control algorithm is also "Pareto efficient," in the terminology of game theory. When we apply the same approach to power control in wireless data transmissions, we find that the corresponding strategy, while locally optimum, is not Pareto efficient. Relative to the telephone algorithm, there are other algorithms that produce higher utility for at least one terminal, without decreasing the utility for any other terminal. This article presents one such algorithm. The algorithm includes a price function proportional to transmitter power. When terminals adjust their power levels to maximize the net utility (utility-price), they arrive at lower power levels and higher utility than they achieve when they individually strive to maximize utility.

Algorithms for Distributed Constraint Satisfaction: A Review

Abstract: When multiple agents are in a shared environment, there usually exist constraints among the possible actions of these agents. A distributed constraint satisfaction problem (distributed CSP) is a problem to find a consistent combination of actions that satisfies these inter-agent constraints. Various application problems in multi-agent systems can be formalized as distributed CSPs. This paper gives an overview of the existing research on distributed CSPs. First, we briefly describe the problem formalization and algorithms of normal, centralized CSPs. Then, we show the problem formalization and several MAS application problems of distributed CSPs. Furthermore, we describe a series of algorithms for solving distributed CSPs, i.e., the asynchronous backtracking, the asynchronous weak-commitment search, the distributed breakout, and distributed consistency algorithms. Finally, we show two extensions of the basic problem formalization of distributed CSPs, i.e., handling multiple local variables, and dealing with over-constrained problems.

Generic support for distributed applications

Abstract: In the late 1980s, software designers introduced middleware platforms to support distributed computing systems. Since then, the rapid evolution of technology has caused an explosion of distributed-processing requirements. Application developers now routinely expect to support multimedia systems and mobile users and computers. Timely response to asynchronous events is crucial to such applications, but current platforms do not adequately meet this need. Another need of existing and emerging applications is the secure interoperability of independent services in large-scale, widely distributed systems. Information systems serving organizations such as universities, hospitals, and government agencies require cross-domain interaction. To meet the needs of these applications, Cambridge University researchers developed middleware extensions that provide a flexible, scalable approach to distributed-application development. This article details the extensions they developed, explaining their distributed software approach and the support it has provided for emerging applications.

A distributed algorithm for delay-constrained unicast routing

Abstract: We study the NP-hard delay-constrained least cost (DCLC) path problem. A solution to this problem is needed to provide real-time communication service to connection-oriented applications, such as video and voice. We propose a simple, distributed heuristic solution, called the delay-constrained unicast routing (DCUR) algorithm, DCUR requires limited network state information to be kept at each node: a cost vector and a delay vector. We prove DCUR's correctness by showing that it is always capable of constructing a loop-free delay-constrained path within finite time, if such a path exists. The worst case message complexity of DCUR is O(|V|/sup 2/) messages, where |V| is the number of nodes. However, simulation results show that, on the average, DCUR requires much fewer messages. Therefore, DCUR scales well to large networks. We also use simulation to compare DCUR to the optimal algorithm, and to the least delay path algorithm. Our results show that DCUR's path costs are within 10% of those of the optimal solution.

Inside MASSIVE-3: flexible support for data consistency and world structuring

Abstract: MASSIVE-3 is our third generation of Collaborative Virtual Environment (CVE) system. This paper describes the goals, design and implementation of key aspects of the MASSIVE-3 system, and in particular its support for data consistency, and world structuring and interest management. MASSIVE-3 adopts a distributed database model, in which all changes to items in the database are represented by explicit events that are themselves visible to the system. Networking is logically multicast, but physically client-server (the reasons for this are explained). MASSIVE-3 makes application behaviours explicitly visible within the database in the form of “Behaviour” data items. MASSIVE-3 implements and extends work on consistency by the University of Reading. In particular, it adds an explicit “Update Request” data item, which allows the system to support a number of different consistency mechanisms within a single virtual world. World structuring in MASSIVE-3 extends the notion of “Locales” from the SPLINE system to include distinctions based on functional class, organisational scope and fidelity. It also allows flexible and general replication and rendering policies to be specified and used for interest management.

Distributed data clustering can be efficient and exact

Abstract: Data clustering is one of the fundamental techniques in scientific data analysis and data mining. It partitions a data set into groups of similar items, as measured by some distance metric. Over the years, data set sizes have grown rapidly with the exponential growth of computer storage and increasingly automated business and manufacturing processes. Many of these datasets are geographically distributed across multiple sites, e.g. different sales or warehouse locations. To cluster such large and distributed data sets, efficient distributed algorithms are called for to reduce the communication overhead, central storage requirements, and computation time, as well as to bring the resources of multiple machines to bear on a given problem as the data set sizes scale-up. We describe a technique for parallelizing a family of center-based data clustering algorithms. The central idea is to communicate only sufficient statistics, yielding linear speed-up with excellent efficiency. The technique does not involve approximation and may be used orthogonally in conjunction with sampling or aggregation-based methods, such as BIRCH, to lessen the quality degradation of their approximation or to handle larger data sets. We demonstrate in this paper that even for relatively small problem sizes, it can be more cost effective to cluster the data in-place using an exact distributed algorithm than to collect the data in one central location for clustering.

Optimal implementation of the weakest failure detector for solving consensus

Abstract: The concept of unreliable failure detector was introduced by T.D. Chandra and S. Toueg (1996) as a mechanism that provides information about process failures. Depending on the properties which the failure detectors guarantee, they proposed a taxonomy of failure detectors. It has been shown that one of the classes of this taxonomy, namely Eventually Strong (/spl nabla/S), is the weakest class allowing a solution of the Consensus problem. The authors present a new algorithm implementing /spl nabla/S. Our algorithm guarantees that eventually all the correct processes agree on a common correct process. This property trivially allows us to provide the accuracy and completeness properties required by /spl nabla/S. We show then that our algorithm is better than any other proposed implementation of /spl nabla/S in terms of the number of messages and the total amount of information periodically sent. In particular, previous algorithms require periodic exchange of at least a quadratic amount of information, while ours only requires O(n log n) (where n is the number of processes). However, we also propose a new measure to evaluate the efficiency of this kind of algorithm, the eventual monitoring degree, which does not rely on a periodic behavior and expresses the degree of processing required by the algorithms better. We show that the runs of our algorithm have optimal eventual monitoring degree.

FIRM: a class of distributed scheduling algorithms for high-speed ATM switches with multiple input queues

Abstract: Advanced input queuing is an attractive, promising architecture for high-speed ATM switches, because it combines the low cost of input queuing with the high performance of output queuing. The need for scalable schedulers for advanced input queuing switch architectures has led to the development of efficient distributed scheduling algorithms. We introduce a new distributed scheduling algorithm, FIRM, which provides improved performance characteristics over alternative distributed algorithms. FIRM achieves saturation throughput 1 with lower delay than the most efficient alternative (up to 50% at high load). Furthermore, it provides improved fairness (it approximates FCFS) and tighter service guarantee than others. FIRM provides a basis for a class of distributed scheduling algorithms, many of which provide even more improved performance characteristics.

Deterministic Radio Broadcasting

Abstract: We consider broadcasting in radio networks: one node of the network knows a message that needs to be learned by all the remaining nodes. We seek distributed deterministic algorithms to perform this task. Radio networks are modeled as directed graphs. They are unknown, in the sense that nodes are not assumed to know their neighbors, nor the size of the network, they are aware only of their individual identifying numbers. If more than one message is delivered to a node in a step then the node cannot hear any of them. Nodes cannot distinguish between such collisions and the case when no messages have been delivered in a step. The fastest previously known deterministic algorithm for deterministic distributed broadcasting in unknown radio networks was presented in [6], it worked in time O(n11/6). We develop three new deterministic distributed algorithms. Algorithm A develops further the ideas of [6] and operates in time O(n1:77291) = O(n9/5), for general networks, and in time O(n1+a+H(a)+o(1)) for sparse networks with in-degrees O(na) fora < 1=2; here H is the entropy function. Algorithm B uses a new approach and works in time O(n3/2 log1/2 n) for general networks or O(n1+a+o(1)) for sparse networks. Algorithm C further improves the performance for general networks running in time O(n3/2).

Clustering algorithms for wireless ad hoc networks

Abstract: Efficient clustering algorithms play a very important role in the fast connection establishment of ad hoc networks. In this paper, we describe a communication model that is derived directly from that of Bluetooth, an emerging technology for pervasive computing; this technology is expected to play a major role in future personal area network applications. We further propose two new distributed algorithms for clustering in wireless ad hoc networks. The existing algorithms often become infeasible because they use models where the discovering devices broadcast their Ids and exchange substantial information in the initial stages of the algorithm.We propose a 2-stage distributed O(N) randomized algorithm for an N node complete network, that always finds the minimum number of star-shaped clusters, which have maximum size. We then present a completely deterministic O(N) distributed algorithm for the same model, which achieves the same purpose. We describe in detail how these algorithms can be applied to Bluetooth for efficient scatternet formation. Finally, we evaluate both algorithms using simulation experiments based on the Bluetooth communication model, and compare their performance.

Asynchronous group mutual exclusion

Abstract: Mutual exclusion and concurrency are two fundamental and essentially opposite features in distributed systems. However, in some applications such as Computer Supported Cooperative Work (CSCW) we have found it necessary to impose mutual exclusion on different groups of processes in accessing a resource, while allowing processes of the same group to share the resource. To our knowledge, no such design issue has been previously raised in the literature.In this paper we address this issue by presenting a new problem, called Congenial Talking Philosophers, to model group mutual exclusion. We also propose several criteria to evaluate solutions of the problem and to measure their performance. Finally, we provide an efficient and highly concurrent distributed algorithm for the problem in a shared-memory model where processes communicate by reading from and writing to shared variables. The distributed algorithm meets the proposed criteria, and has performance similar to some naive but centralized solutions to the problem.

Power controlled multiple access (PCMA) in wireless communication networks

Abstract: We address the issue of power-controlled shared channel access in future wireless networks supporting packetized data traffic, beyond the voice-oriented continuous traffic primarily supported by current-generation networks. First, some novel formulations of the power control problem are introduced, which become progressively more general by incorporating various relevant costs. The analysis of the models under simple, yet natural, assumptions yields certain ubiquitous structural properties of 'optimal' power control algorithms. Based on such structural properties, we design a new family of distributed and asynchronous PCMA algorithms and evaluate them experimentally by simulation. They are found to perform substantially better than a standard benchmark algorithm for power control. This is a first step towards the design of full PCMA protocols for autonomous channel access in high-performance wireless networks.

Self-Stabilizing Local Mutual Exclusion and Daemon Refinement

Abstract: Refining self-stabilizing algorithms which use tighter scheduling constraints (weaker daemon) into corresponding algorithms for weaker or no scheduling constraints (stronger daemon), while preserving the stabilization property, is useful and challenging. Designing transformation techniques for these refinements has been the subject of serious investigations in recent years. This paper proposes a transformation technique to achieve the above task. The heart of the transformer is a self-stabilizing local mutual exclusion algorithm. The local mutual exclusion problem is to grant a process the privilege to enter the critical section if and only if none of the neighbors of the process has the privilege. The contribution of this paper is twolold. First, we present a bounded-memory self-stabilizing local mutual exclusion algorithm for arbitrary network, assuming any arbitrary daemon. After stabilization, this algorithm maintains a bound on the service time (the delay between two successive executions of the critical section by a particular process). This bound is n×(n-1)/2 where n is the network size. Second, we use the local mutual exclusion algorithm to design two scheduler transformers which convert the algorithms working under a weaker daemon to ones which work under the distributed, arbitrary (or unfair) daemon, both transformers preserving the self-stabilizing property. The first transformer refines algorithms written under the central daemon, while the second transformer refines algorithms designed for the k-fair (k ? (n - 1)) daemon.

Agent-based resource discovery

Abstract: Presents a distributed discovery method allowing individual nodes to gather information about resources in a wide-area distributed system made up of autonomous systems linked together by a network technology substrate. We introduce an algorithm and a model for distributed awareness and a framework for the dynamic assembly of agents monitoring network resources. Whenever an agent needs detailed information about the individual components of another system, it uses the information gathered by the distributed awareness mechanism to identify the target system, then creates a description of a monitoring agent that is capable of providing the information about remote resources, and sends this description to the remote site. There, an agent factory dynamically assembles the monitoring agent. This solution is scalable and is suitable for heterogeneous environments where the architecture and the hardware resources of individual nodes differ, where the services provided by the system are diverse, and where the bandwidth and latency of the communication links cover a broad range.

An overview of reliable multicast transport protocol II

Abstract: This document provides an overview of the reliable multicast transport protocol II, RMTP-II. RMTP-II is a reliable multicast protocol, designed to reliably and efficiently send data from a few senders to large groups of simultaneous recipients. It works over both symmetric networks and asymmetrical network topologies such as those provided by satellite, cable modem, or ADSL carriers. Before sending, each sender must connect with a trusted top node to receive permission and control parameters for its data stream. The top node provides network managers with a single point of control for the senders, allowing them to monitor and control the traffic being sent. RMTP-II builds on a rich field of existing work, and adds to it the following novel contributions. It differentiates the roles of the nodes in the protocol, provides algorithms for smoothing and control of the return (TRACK) traffic, and provides explicit support for highly asymmetrical networks. It provides explicit network management controls through a centralized point of control, a fully distributed membership protocol that enables positive confirmation of data delivery, and fault recovery algorithms which are integrated to the reliability semantics of the protocol. It includes a novel reliability level called time bounded reliability, and offers a unique combination of TRACKs, NACKs, and FEC for increased scalability and real-time performance. Finally, it integrates distributed algorithms for RTT calculation to each receiver, and provides automatic configuration of receiver nodes.

Energy efficient routing for connection-oriented traffic in ad-hoc wireless networks

Abstract: We address the problem of routing connection-oriented traffic in wireless ad-hoc networks under minimum energy expenditures. We outline the tradeoffs that arise by the flexibility to transmit at different power levels and propose distributed algorithms on how to select connection paths relying only on local information. We propose a set of link metrics that capture network parameters and use them for the computation of shortest paths in static topologies. The performance is measured by the call blocking probability and average consumed energy. The algorithms are evaluated by a detailed simulation model and their key characteristics are discussed.

A Distributed Algorithm to Evaluate Quantified Boolean Formulae

Abstract: In this paper, we present PQSOLVE, a distributed theorem-prover for Quantified Boolean Formulae First, we introduce our sequential algorithm QSOLVE, which uses new heuristics and improves the use of known heuristics to prune the search tree As a result, QSOLVE is more efficient than the QSAT-solvers previously known We have parallelized QSOLVE The resulting distributed QSAT-solver PQSOLVE uses parallel search techniques, which we have developed for distributed game tree search PQSOLVE runs efficiently on distributed systems, i e parallel systems without any shared memory We briefly present experiments that show a speedup of about 114 on 128 processors To the best of our knowledge we are the first to introduce an efficient parallel QSAT-solver

Indulgent algorithms (preliminary version)

Abstract: Informally, an indulgent algorithm is a distributed algorithm that tolerates unreliable failure detection: the algorithm is indulgent towards its failure detector. This paper formally characterises such algorithms and states some of their interesting features. We show that indulgent algorithms are inherently safe and uniform. We also state impossibility results for indulgent solutions to divergent problems like consensus, and failure-sensitive problems like non-blocking atomic commit and terminating reliable broadcast.

Matchmaking frameworks for distributed resource management

Abstract: Federated distributed systems present new challenges to resource management. Conventional resource managers are based on a relatively static resource model and a centralized allocator that assigns resources to customers. Distributed environments, particularly those built to support high-throughput computing (HTC), are often characterized by distributed management and distributed ownership. Distributed management introduces resource heterogeneity: Not only the set of available resources, but even the set of resource types is constantly changing. Distributed ownership introduces policy heterogeneity: Each resource may have its own idiosyncratic allocation policy. We propose a resource management framework based on a matchmaking paradigm to address these shortcomings. Matchmaking services enable discovery and exchange of goods and services in marketplaces. Agents that provide or require services advertise their presence by publishing constraints and preferences on the entities they would like to be matched with, as well as their own characteristics. A matchmaker uses a matching operation to discover pairings between compatible agents. Since the notion of “compatible” is completely determined by the content of agent classified advertisements (classads), a matchmaker can match classads from different kinds of entities in a general manner. Matched agents activate a separate claiming protocol to confirm the match and establish an allocation. The resulting framework is robust, scalable, flexible and evolvable, and has been demonstrated in Condor, a production-quality distributed high throughput computing system developed at the University of Wisconsin-Madison. The goal of this dissertation is to show the power, flexibility, desirability and feasibility of resource management through matchmaking. We detail the architecture and operation of matchmaking frameworks, and describe mechanisms to implement the components and interactions in such systems. We describe the architecture of a matchmaking framework that distinguishes itself by providing both bilateral and multilateral matchmaking (i.e., gangmatching) services. The classad language, a semi-structured agent specification language, is presented, and an indexing model for the classad data model is defined. The indexing solution tolerates the lax semantics of semi-structured data models, and indexes both classad attributes and constraints to efficiently identify compatible advertisements. Finally, algorithms that implement the proposed gangmatching model are described, and their performance characteristics analyzed.

Braess-like paradoxes in distributed computer systems

Abstract: We consider optimal distributed decisions in distributed computer systems. We identify a Braess like paradox in which adding capacity to the system may degrade the performance of all users. Unlike the original Braess paradox, we show that this behavior occurs only in the case of finitely many users and not in the case of infinite number of users.

Static and adaptive data replication algorithms for fast information access in large distributed systems

Abstract: Creating replicas of frequently accessed objects across a read-intensive network can result in large bandwidth savings which, in turn, can lead to reduction in user response time. On the contrary, data replication in the presence of writes incurs extra cost due to multiple updates. The set of sites at which an object is replicated constitutes its replication scheme. Finding an optimal replication scheme that minimizes the amount of network traffic given read and write frequencies for various objects, is NP-complete in general. We propose two heuristics to deal with this problem for static read and write patterns. The first is a simple and fast greedy heuristic that yields good solutions when the system is predominantly read-oriented. The second is a genetic algorithm that through an efficient exploration of the solution space provides better solutions for cases where the greedy heuristic does not perform well. We also propose an extended genetic algorithm that rapidly adapts to the dynamically changing characteristics such as the frequency of reads and writes for particular objects.

Distributed power control and spreading gain allocation in CDMA data networks

Abstract: We study the radio resource allocation problem of distributed joint transmission power control and spreading gain allocation in a DS-CDMA mobile data network. The network consists of K base stations and M wireless data users. The data streams generated by the users are treated as best-effort traffic, in the sense that there are no pre-specified constraints on the quality of the radio channels. We are interested in designing a distributed algorithm that achieves maximal (or near-maximal in some reasonable sense) aggregate throughput, subject to peak power constraints. We provide an algorithm where base stations coordinate in a distributed fashion to control the powers and spreading gains of the users, and show that it converges to a Nash equilibrium point. In general, there may be multiple equilibrium points; however, certain structural properties of the throughput expression can be exploited to significantly trim the search space and induce an ordering on the users in each cell. The numerical results indicate that with these modifications, the algorithm frequently converges in just a few iterations to the throughput maximizing (globally optimal) power and spreading gain allocation.

Concurrent exception handling and resolution in distributed object systems

Abstract: We address the problem of how to handle exceptions in distributed object systems. In a distributed computing environment, exceptions may be raised simultaneously in different processing nodes and thus need to be treated in a coordinated manner. Mishandling concurrent exceptions can lead to catastrophic consequences. We take two kinds of concurrency into account: 1) Several objects are designed collectively and invoked concurrently to achieve a global goal and 2) multiple objects (or object groups) that are designed independently compete for the same system resources. We propose a new distributed algorithm for resolving concurrent exceptions and show that the algorithm works correctly even in complex nested situations, and is an improvement over previous proposals in that it requires only O(n/sub max/N/sup 2/) messages, thereby permitting quicker response to exceptions.

A fair distributed mutual exclusion algorithm

Abstract: This paper presents a fair decentralized mutual exclusion algorithm for distributed systems in which processes communicate by asynchronous message passing. The algorithm requires between N-1 and 2(N-1) messages per critical section access, where N is the number of processes in the system. The exact message complexity can be expressed as a deterministic function of concurrency in the computation. The algorithm does not introduce any other overheads over Lamport's and Ricart-Agrawala's algorithms, which require 3(N-1) and 2(N-1) messages, respectively, per critical section access and are the only other decentralized algorithms that allow mutual exclusion access in the order of the timestamps of requests.

Distributed fault-tolerant channel allocation for cellular networks

Abstract: A channel allocation algorithm includes channel acquisition and channel selection algorithms. Most of the previous work concentrates on the channel selection algorithm since early channel acquisition algorithms are centralized and rely on a mobile switching center (MSC) to accomplish channel acquisition. Distributed channel acquisition algorithms have received considerable attention due to their high reliability and scalability. However, in these algorithms, a borrower needs to consult with its interference neighbors in order to borrow a channel. Thus, the borrower fails to borrow channels when it cannot communicate with any interference neighbor. In real-life networks, under heavy traffic load, a cell has a large probability to experience an intermittent network congestion or even a communication link failure. In existing distributed algorithms, since a cell has to consult with a large number of interference neighbors to borrow a channel, the failure rate will be much higher under heavy traffic load. Therefore, previous distributed channel allocation algorithms are not suitable for real-life networks. We first propose a fault-tolerant channel acquisition algorithm which tolerates communication link failures and node (MH or MSS) failures. Then, we present a channel selection algorithm and integrate it into the distributed acquisition algorithm. Detailed simulation experiments are carried out in order to evaluate our proposed methodology. Simulation results show that our algorithm significantly reduces the failure rate under network congestion, communication link failures, and node failures compared to nonfault-tolerant channel allocation algorithms. Moreover, our algorithm has low message overhead compared to known distributed channel allocation algorithms, and outperforms them in terms of failure rate under uniform as well as nonuniform traffic distribution.

Experiences with parallel N-body simulation

Abstract: This paper describes our experiences developing high-performance code for astrophysical N-body simulations. Recent N-body methods are based on an adaptive tree structure. The tree must be built and maintained across physically distributed memory; moreover, the communication requirements are irregular and adaptive. Together with the need to balance the computational work-load among processors, these issues pose interesting challenges and tradeoffs for high-performance implementation. Our implementation was guided by the need to keep solutions simple and general. We use a technique for implicitly representing a dynamic global tree across multiple processors which substantially reduces the programming complexity as well as the performance overheads of distributed memory architectures. The contributions include methods to vectorize the computation and minimize communication time which are theoretically and experimentally justified. The code has been tested by varying the number and distribution of bodies on different configurations of the Connection Machine CM-5. The overall performance on instances with 10 million bodies is typically over 48 percent of the peak machine rate, which compares favorably with other approaches.

A multi-version approach to conflict resolution in distributed groupware systems

Abstract: Groupware systems are a special class of distributed computing systems which support human-computer-human interaction. Real-time collaborative graphics editors allow a group of users to view and edit the same graphics document at the same time from geographically dispersed sites connected by communication networks. Resolving conflict access to shared objects is one of the core issues in the design of this type of system. This paper proposes a novel distributed multi-version approach to conflict resolution. This approach aims to preserve the work concurrently produced by multiple users in the face of conflicts, and to minimize the number of object versions for accommodating combined effects of conflicting and compatible operations. Major technical contributions of this work include a formal specification of a unique combined effect for any group of conflicting and compatible operations, a distributed algorithm for incremental creation of multiple object versions, and a consistent object identification scheme for multi-version and multireplica graphics editing systems. All algorithms and schemes preserved in this paper have been used in the GRACE (Graphics Collaborative Editing) system implemented in Java.