Default Cover Image

Parallel and Distributed Processing Symposium, International

Apr. 26 2004 to Apr. 30 2004

Santa Fe, New Mexico

ISBN: 0-7695-2132-0

Volume:

Table of Contents

Message from the general vice-chairFreely available from IEEE.pp. xliv-xliv
Message from the tutorials chairFreely available from IEEE.pp. xlv-xlv
Message from the industrial track chairFreely available from IEEE.pp. xlvi-xlvi
IPDPS 2004 organizationFreely available from IEEE.pp. xlviii-xlix
IPDPS 2004 technical programFreely available from IEEE.pp. li-liii
Session 1: Scheduling and Mapping
Average-Case Performance Analysis and Validation of Online Scheduling of Independent Parallel TasksFull-text access may be available. Sign in or learn about subscription options.pp. 2a
Session 1: Scheduling and Mapping
Scheduling of Query Execution Plans in Symmetric Multiprocessor Database SystemsFull-text access may be available. Sign in or learn about subscription options.pp. 2b
Session 1: Scheduling and Mapping
A Novel Static Task Scheduling Algorithm in Distributed Computing EnvironmentsFull-text access may be available. Sign in or learn about subscription options.pp. 3a
Session 1: Scheduling and Mapping
Two Fast and Efficient Message Scheduling Algorithms for Data Redistribution through a BackboneFull-text access may be available. Sign in or learn about subscription options.pp. 3b
Session 1: Scheduling and Mapping
Assignment of Shortest Paths Spanning Trees in MeshesFull-text access may be available. Sign in or learn about subscription options.pp. 4a
Session 1: Scheduling and Mapping
Parallel Maximum Weight Bipartite Matching Algorithms for Scheduling in Input-Queued SwitchesFull-text access may be available. Sign in or learn about subscription options.pp. 4b
Session 2: Scientific Applications I
Employing Nested OpenMP for the Parallelization of Multi-Zone Computational Fluid Dynamics ApplicationsFull-text access may be available. Sign in or learn about subscription options.pp. 6a
Session 2: Scientific Applications I
Performance Characteristics of the Multi-Zone NAS Parallel BenchmarksFull-text access may be available. Sign in or learn about subscription options.pp. 6b
Session 2: Scientific Applications I
Distributed Correction of Proximity Effect in Electron Beam Lithography on a Heterogeneous ClusterFull-text access may be available. Sign in or learn about subscription options.pp. 7a
Session 2: Scientific Applications I
A Parallel Object-Oriented Application for 3D ElectromagnetismFull-text access may be available. Sign in or learn about subscription options.pp. 7b
Session 2: Scientific Applications I
Solving Large Sparse Linear Systems in End-to-end Accelerator Structure SimulationsFull-text access may be available. Sign in or learn about subscription options.pp. 8a
Session 2: Scientific Applications I
Optimization of the POLCOMS Hydrodynamic Code for Terascale High-Performance ComputersFull-text access may be available. Sign in or learn about subscription options.pp. 8b
Session 3: Interconnection Networks
BLACK-BUS: A New Data-Transfer Technique Using Local Address on Networks-on-ChipsFull-text access may be available. Sign in or learn about subscription options.pp. 10a
Session 3: Interconnection Networks
Fast and Scalable MPI-Level Broadcast Using InfiniBand's Hardware Multicast SupportFull-text access may be available. Sign in or learn about subscription options.pp. 10b
Session 3: Interconnection Networks
A Multiple LID Routing Scheme for Fat-Tree-Based InfiniBand NetworksFull-text access may be available. Sign in or learn about subscription options.pp. 11a
Session 3: Interconnection Networks
A Scalable Architecture for Distributed Shared Memory Multiprocessors Using Optical InterconnectsFull-text access may be available. Sign in or learn about subscription options.pp. 11b
Session 3: Interconnection Networks
On Constructing the Minimum Orthogonal Convex Polygon in 2-D Faulty MeshesFull-text access may be available. Sign in or learn about subscription options.pp. 12a
Session 3: Interconnection Networks
LORE — Local Reconfiguration for Fault Management in Irregular InterconnectsFull-text access may be available. Sign in or learn about subscription options.pp. 12b
Session 4: Parallel Programming Models/Implementations
High Performance Implementation of MPI Derived Datatype Communication over InfiniBandFull-text access may be available. Sign in or learn about subscription options.pp. 14a
Session 4: Parallel Programming Models/Implementations
Integrating Java and Matlab Components into the Same Parallel and Distributed Application Using JavaPortsFull-text access may be available. Sign in or learn about subscription options.pp. 14b
Session 4: Parallel Programming Models/Implementations
Performance Comparison of Pure MPI vs Hybrid MPI-OpenMP Parallelization Models on SMP ClustersFull-text access may be available. Sign in or learn about subscription options.pp. 15a
Session 4: Parallel Programming Models/Implementations
Architecture of LA-MPI, A Network-Fault-Tolerant MPIFull-text access may be available. Sign in or learn about subscription options.pp. 15b
Session 4: Parallel Programming Models/Implementations
The UPC Memory Model: Problems and ProspectsFull-text access may be available. Sign in or learn about subscription options.pp. 16a
Session 4: Parallel Programming Models/Implementations
Design and Implementation of MPICH2 over InfiniBand with RDMA SupportFull-text access may be available. Sign in or learn about subscription options.pp. 16b
Session 5: Network Algorithms
Random Broadcasting in Product Networks and Symmetric NetworksFull-text access may be available. Sign in or learn about subscription options.pp. 18a
Session 5: Network Algorithms
Optimal Multi-Channel Data Allocation with Flat Broadcast Per ChannelFull-text access may be available. Sign in or learn about subscription options.pp. 18b
Session 5: Network Algorithms
On the IP Routing Tables Minimization with Addresses ReassignmentsFull-text access may be available. Sign in or learn about subscription options.pp. 19a
Session 5: Network Algorithms
Pipelining Broadcasts on Heterogeneous PlatformsFull-text access may be available. Sign in or learn about subscription options.pp. 19b
Session 5: Network Algorithms
Towards Efficient Load Balancing in Structured P2P SystemsFull-text access may be available. Sign in or learn about subscription options.pp. 20a
Session 5: Network Algorithms
Load Balancing: Dimension Exchange on Product GraphsFull-text access may be available. Sign in or learn about subscription options.pp. 20b
Session 6: Grid Applications and Sensor Networks
A Trust Brokering System and Its Application to Resource Management in Public-Resource GridsFull-text access may be available. Sign in or learn about subscription options.pp. 22a
Session 6: Grid Applications and Sensor Networks
Single Sign-On in In-VIGO: Role-Based Access via Delegation Mechanisms Using Short-Lived User IdentitiesFull-text access may be available. Sign in or learn about subscription options.pp. 22b
Session 6: Grid Applications and Sensor Networks
A Cluster Oriented Model for Dynamically Balanced DHTsFull-text access may be available. Sign in or learn about subscription options.pp. 23a
Session 6: Grid Applications and Sensor Networks
Policy Based Scheduling for Simple Quality of Service in Grid ComputingFull-text access may be available. Sign in or learn about subscription options.pp. 23b
Session 6: Grid Applications and Sensor Networks
A New Algorithm for Relative Localization in Wireless Sensor NetworksFull-text access may be available. Sign in or learn about subscription options.pp. 24a
Session 6: Grid Applications and Sensor Networks
Malicious Node Detection in Wireless Sensor NetworksFull-text access may be available. Sign in or learn about subscription options.pp. 24b
Session 7: Distributed System Architecture
Cycloid: A Constant-Degree and Lookup-Efficient P2P Overlay NetworkFull-text access may be available. Sign in or learn about subscription options.pp. 26a
Session 7: Distributed System Architecture
Characterizing and Evaluating Desktop Grids: An Empirical StudyFull-text access may be available. Sign in or learn about subscription options.pp. 26b
Session 7: Distributed System Architecture
Distributed Embedded Systems for Low Power: A Case StudyFull-text access may be available. Sign in or learn about subscription options.pp. 27a
Session 7: Distributed System Architecture
How to Run Experiments with Large Peer-to-Peer Data StructuresFull-text access may be available. Sign in or learn about subscription options.pp. 27b
Session 7: Distributed System Architecture
Mobility-Sensitive Topology Control in Mobile Ad Hoc NetworksFull-text access may be available. Sign in or learn about subscription options.pp. 28a
Session 7: Distributed System Architecture
Architecture and Protocol for Reliable Event Delivery to Clients of a High-Availability ClusterFull-text access may be available. Sign in or learn about subscription options.pp. 28b
Session 8: Shared Memory Operations/Optimizations/Models
Adaptive Memory Paging for Efficient Gang Scheduling of Parallel ApplicationsFull-text access may be available. Sign in or learn about subscription options.pp. 30a
Session 8: Shared Memory Operations/Optimizations/Models
Integrating Remote Invocation and Distributed Shared StateFull-text access may be available. Sign in or learn about subscription options.pp. 30b
Session 8: Shared Memory Operations/Optimizations/Models
Host-Assisted Zero-Copy Remote Memory Access Communication on InfiniBandFull-text access may be available. Sign in or learn about subscription options.pp. 31a
Session 8: Shared Memory Operations/Optimizations/Models
Nemos: A Framework for Axiomatic and Executable Specifications of Memory Consistency ModelsFull-text access may be available. Sign in or learn about subscription options.pp. 31b
Session 8: Shared Memory Operations/Optimizations/Models
Flecc: A Flexible Cache Coherence Protocol for Dynamic Component-Based SystemsFull-text access may be available. Sign in or learn about subscription options.pp. 32a
Session 8: Shared Memory Operations/Optimizations/Models
Dynamic versus Static Locking in Real-Time Parallel Database SystemsFull-text access may be available. Sign in or learn about subscription options.pp. 32b
Session 9: Plenary Session: Best Papers
Translating Submachine Locality into Locality of ReferenceFull-text access may be available. Sign in or learn about subscription options.pp. 34a
Session 9: Plenary Session: Best Papers
Efficient Synthesis of Out-of-Core Algorithms Using a Nonlinear Optimization SolverFull-text access may be available. Sign in or learn about subscription options.pp. 34b
Session 9: Plenary Session: Best Papers
Designing WDM Optical Interconnects with Full Connectivity by Using Limited Wavelength ConversionFull-text access may be available. Sign in or learn about subscription options.pp. 35a
Session 9: Plenary Session: Best Papers
Running OpenMP Applications Efficiently on an Everything-Shared SDSMFull-text access may be available. Sign in or learn about subscription options.pp. 35b
Session 10: Parallel Algorithms for Graphs and Multiprocessors
A Fast, Parallel Spanning Tree Algorithm for Symmetric MultiprocessorsFull-text access may be available. Sign in or learn about subscription options.pp. 38a
Session 10: Parallel Algorithms for Graphs and Multiprocessors
Fast and Scalable Parallel Algorithms for Euclidean Distance Transform on LARPBSFull-text access may be available. Sign in or learn about subscription options.pp. 38b
Session 10: Parallel Algorithms for Graphs and Multiprocessors
Fast Shared-Memory Algorithms for Computing the Minimum Spanning Forest of Sparse GraphsFull-text access may be available. Sign in or learn about subscription options.pp. 39a
Session 10: Parallel Algorithms for Graphs and Multiprocessors
Optimising Static Workload Allocation in MulticlustersFull-text access may be available. Sign in or learn about subscription options.pp. 39b
Session 10: Parallel Algorithms for Graphs and Multiprocessors
A Multiprocessor Implementation of the Total Bandwidth ServerFull-text access may be available. Sign in or learn about subscription options.pp. 40a
Session 10: Parallel Algorithms for Graphs and Multiprocessors
An Algorithm for Geometric Load Balancing with Two ConstraintsFull-text access may be available. Sign in or learn about subscription options.pp. 40b
Session 11: Scientific Applications II
A Large Scale Monte Carlo Simulator for Cellular MicrophysiologyFull-text access may be available. Sign in or learn about subscription options.pp. 42a
Session 11: Scientific Applications II
A Hierarchical Parallel Scheme for Global Parameter Estimation in Systems BiologyFull-text access may be available. Sign in or learn about subscription options.pp. 42b
Session 11: Scientific Applications II
Parallel Simulation of Fluid Slip in a MicrochannelFull-text access may be available. Sign in or learn about subscription options.pp. 43a
Session 11: Scientific Applications II
Isocoupling: Reusing Kernel Coupling Values to Predict the Performance of Parallel ApplicationsFull-text access may be available. Sign in or learn about subscription options.pp. 43b
Session 11: Scientific Applications II
A Self-Adapting Distributed Memory Package for Fast Signal TransformsFull-text access may be available. Sign in or learn about subscription options.pp. 44a
Session 11: Scientific Applications II
Parallel Brutus: The First Distributed, FPGA Accelerated Chess ProgramFull-text access may be available. Sign in or learn about subscription options.pp. 44b
Session 12: Distributed Memory and Networks
Taking Advantage of the Overlay Geometrical Structures for Mobile Agent CommunicationsFull-text access may be available. Sign in or learn about subscription options.pp. 46a
Session 12: Distributed Memory and Networks
Building a Scalable Bipartite P2P Overlay NetworkFull-text access may be available. Sign in or learn about subscription options.pp. 46b
Session 12: Distributed Memory and Networks
Specification and Architecture Supports for Component Adaptations on Distributed EnvironmentsFull-text access may be available. Sign in or learn about subscription options.pp. 47a
Session 12: Distributed Memory and Networks
Integrating Program Component Executables on Distributed Memory Architectures via MPHFull-text access may be available. Sign in or learn about subscription options.pp. 47b
Session 12: Distributed Memory and Networks
Hierarchical Routing with Soft-State Replicas in TerraDirFull-text access may be available. Sign in or learn about subscription options.pp. 48a
Session 12: Distributed Memory and Networks
Application-Perceived Multicast Push PerformanceFull-text access may be available. Sign in or learn about subscription options.pp. 48b
Session 13: Distributed Algorithms and Data Structures
Almost Wait-Free Resizable HashtablesFull-text access may be available. Sign in or learn about subscription options.pp. 50a
Session 13: Distributed Algorithms and Data Structures
Star-Coloring of Graphs for Conflict-Free Access to Parallel Memory SystemsFull-text access may be available. Sign in or learn about subscription options.pp. 50b
Session 13: Distributed Algorithms and Data Structures
A Distributed Hash Table for Computational GridsFull-text access may be available. Sign in or learn about subscription options.pp. 51a
Session 13: Distributed Algorithms and Data Structures
An Efficient Distributed Mutual Exclusion Algorithm Based on Relative Consensus VotingFull-text access may be available. Sign in or learn about subscription options.pp. 51b
Session 13: Distributed Algorithms and Data Structures
Bounded Service Time and Memory Space Optimal Self-Stabilizing Token Circulation Protocol on Unidirectional RingsFull-text access may be available. Sign in or learn about subscription options.pp. 52a
Session 13: Distributed Algorithms and Data Structures
Distributed Adaptive Task Allocation in Heterogeneous Computing Environments to Maximize ThroughputFull-text access may be available. Sign in or learn about subscription options.pp. 52b
Session 14: P2P and Networking Applications
A Neural Network Based Approach for Overlay Multicast in Media Streaming SystemsFull-text access may be available. Sign in or learn about subscription options.pp. 54a
Session 14: P2P and Networking Applications
Secure and Reliable Decentralized Peer-to-Peer Web CacheFull-text access may be available. Sign in or learn about subscription options.pp. 54b
Session 14: P2P and Networking Applications
Exploiting Client Cache: A Scalable and Efficient Approach to Build Large Web CacheFull-text access may be available. Sign in or learn about subscription options.pp. 55a
Session 14: P2P and Networking Applications
Diagnostics for Causes of Packet Loss in a High Performance Data Transfer SystemFull-text access may be available. Sign in or learn about subscription options.pp. 55b
Session 14: P2P and Networking Applications
Prediction-Based Routing through Least Cost Delay ConstraintFull-text access may be available. Sign in or learn about subscription options.pp. 56a
Session 14: P2P and Networking Applications
A SNAP-Based Community Resource Broker Using a Three-Phase Commit ProtocolFull-text access may be available. Sign in or learn about subscription options.pp. 56b
Session 15: Parallel System Architecture
Highly Efficient Synchronization Based on Active Memory OperationsFull-text access may be available. Sign in or learn about subscription options.pp. 58a
Session 15: Parallel System Architecture
On the Feasibility of Incremental Checkpointing for Scientific ComputingFull-text access may be available. Sign in or learn about subscription options.pp. 58b
Session 15: Parallel System Architecture
Utilizing Home Node Prediction to Improve the Performance of Software Distributed Shared MemoryFull-text access may be available. Sign in or learn about subscription options.pp. 59a
Session 15: Parallel System Architecture
Multithreaded Home-Based Lazy Release Consistency over VIAFull-text access may be available. Sign in or learn about subscription options.pp. 59b
Session 15: Parallel System Architecture
A Novel Method for Adding Multiprocessor Support to a Large and Complex Uniprocessor KernelFull-text access may be available. Sign in or learn about subscription options.pp. 60a
Session 15: Parallel System Architecture
Assignment and Scheduling of Real-time DSP Applications for Heterogeneous Functional UnitsFull-text access may be available. Sign in or learn about subscription options.pp. 60b
Session 16: Thread/Job Scheduling, Load Balancing and Management
Unobtrusiveness and Efficiency in Idle Cycle Stealing for PC GridsFull-text access may be available. Sign in or learn about subscription options.pp. 62a
Session 16: Thread/Job Scheduling, Load Balancing and Management
Packet Probing as Network Load Detection for Scientific Applications at Run-TimeFull-text access may be available. Sign in or learn about subscription options.pp. 62b
Session 16: Thread/Job Scheduling, Load Balancing and Management
Queue Scheduling and Advance Reservations with COSYFull-text access may be available. Sign in or learn about subscription options.pp. 63a
Showing 100 out of 171