Parallel and Distributed Processing and Applications: 4th International Symposium, ISPA 2006, Sorrento, Italy, December 4-6, 2006, Proceedings (Lecture Notes in Computer Science, 4330)

✍ Scribed by Minyi Guo (editor), Laurence T. Yang (editor), Beniamino Di Martino (editor), Hans Zima (editor), Jack Dongarra (editor), Feilong Tang (editor)

• Publisher: Springer
• Year: 2006
• Tongue: English
• Leaves: 971
• Category: Library

✦ Synopsis

Welcome to the proceedings of the 4th International Symposium on Parallel and Distributed Processing and Applications (ISPA 2006), which was held in Sorrento, Italy, December, 4–6 2006. Parallel computing has become a mainstream research area in computer s- ence and the ISPA conference has become one of the premier forums for the p- sentation of new and exciting research on all aspects of parallel and distributed computing. We are pleased to present the proceedings for ISPA 2006, which comprises a collection of excellent technical papers and keynote speeches. The accepted papers cover a wide range of exciting topics including architectures, languages, algorithms, software, networking and applications. The conference continues to grow and this year a record total of 277 ma- scriptsweresubmitted forconsiderationbytheProgramCommittee.Fromthese submissions the Program Committee selected only 79 regular papers in the p- gram, which re?ects the acceptance rate as 28%. An additional 10 workshops complemented the outstanding paper sessions. The submission and review process worked as follows. Each submission was assigned to at least three Program Committee members for review. Each P- gram Committee member prepared a single review for each assigned paper or assigned a paper to an outside reviewer for review. In addition, the Program Chairs and Program Vice-Chairs read the papers when a con?icting review - sult occurred. Finally, after much discussion among the Program Chairs and ProgramVice-Chairs, based on the review scores, the ProgramChairs made the ?nal decision. Given the large number of submissions, each ProgramCommittee member was assigned roughly 7–12 papers.

✦ Table of Contents

Frontmatter
Keynote Speech
The Walls of Computer Design
The Impact of Multicore on Math Software and Exploiting Single Precision Computing to Obtain Double Precision Results
Implementing Tomorrow's Programming Languages
Trends in High Performance Computing for Industry and Research
Emerging Technologies and Large Scale Facilities in HPC
Track 1. Architectures and Networks
Session 1. Architectures
Architecture and Performance of Dynamic Offloader for Cluster Network
Session Key Forwarding Scheme Based on AAA Architecture in Wireless Networks
Dynamic Anchor Based Mobility Management Scheme for Mobile IP Networks
Forest Search: A Paradigm for Faster Exploration of Scale-Free Networks
Choosing a Load Balancing Scheme for Agent-Based Digital Libraries
A Novel Software Solution for Localized Thermal Problems
The Optimum Network on Chip Architectures for Video Object Plane Decoder Design
A Hardware NIC Scheduler to Guarantee QoS on High Performance Servers
Studying Several Proposals for the Adaptation of the DTable Scheduler to Advanced Switching
Asymmetrical SSL Tunnel Based VPN
Session 2. Networks
Multihomed Routing in Multicast Service Overlay Network
Region-Based Multicast Routing Protocol with Dynamic Address Allocation Scheme in Mobile Ad-Hoc Networks
A Novel Approach for Topology-Aware Overlay Multicasting
Limitations of Fixed Timers for Wireless Links
Performance Analysis of IEEE 802.11e WLANs with Hidden Stations and Heterogeneous Traffic
Study on High Performance IPv4/IPv6 Transition and Access Service
Limiting the Effects of Deafness and Hidden Terminal Problems in Directional Communications
Enforcing Dimension-Order Routing in On-Chip Torus Networks Without Virtual Channels
A Distributed Backbone Formation Algorithm for Mobile Ad Hoc Networks
A Service Differentiation Mechanism for Improving the Performance of IEEE 802.15.4 Sensor Networks
Randomized Leader Election Protocols in Noisy Radio Networks with a Single Transceiver
A Hybrid Intelligent Preventive Fault-Tolerant QoS Unicast Routing Scheme in IP over DWDM Optical Internet
Track 2. Languages and Algorithms
Session 3. Languages
From XML Specifications to Parallel Programs
Managing Grid Computations: An ORC-Based Approach
Adaptive Technique for Automatic Communication Access Pattern Discovery Applied to Data Prefetching in Distributed Applications Using Neural Networks and Stochastic Models
Process Scheduling Using Ant Colony Optimization Techniques
Interlocking Control by Distributed Signal Boxes: Design and Verification with the SPIN Model Checker
A Delay Time-Based Peak Load Control for Stable Performance
Interference Aware Dynamic Subchannel Allocation in a Multi-cellular OFDMA System Based on Traffic Situation
Sized-Based Replacement-k Replacement Policy in Data Grid Environments
Barrier Elimination Based on Access Dependency Analysis for OpenMP
Session 4. Algorithms
Parallelization Algorithms for Three-Body Interactions in Molecular Dynamics Simulation
An Efficient Distributed Algorithm for Mining Association Rules
Adaptive Algorithms Using Bounded Memory Are Inherently Non-uniform
On the Implementation of Parallel Shortest Path Algorithms on a Supercomputer
An Efficient K-Coverage Eligibility Algorithm on Sensor Networks
A Distributed Algorithm for a b-Coloring of a Graph
Topology-Sensitive Epidemic Algorithm for Information Spreading in Large-Scale Systems
Performance Modeling and Optimal Block Size Selection for the Small-Bulge Multishift QR Algorithm
A Parallel Mutual Information Based Image Registration Algorithm for Applications in Remote Sensing
A Novel Broadcasting Algorithm for Wireless Sensor Networks
Track 3. Middleware and Cooperative Computing
Session 5. Middleware
Middleware Services for Pervasive Grids
Allocating QoS-Constrained Workflow-Based Jobs in a Multi-cluster Grid Through Queueing Theory Approach
Managing Multiple Isolation Levels in Middleware Database Replication Protocols
Proof and Evaluation of a 1CS Middleware Data Replication Protocol Based on O2PL
Reconfiguration of Information Management Framework Based on Adaptive Grid Computing
Framework for Enabling Highly Available Distributed Applications for Utility Computing
A Bluetooth MPI Framework for Collaborative Computer Graphics
Study on Application Server Aging Prediction Based on Wavelet Network with Hybrid Genetic Algorithm
Autonomic Service Reconfiguration in a Ubiquitous Computing Environment
Remote Class Prefetching: Improving Performance of Java Applications on Grid Platforms
Session 6. Cooperative Computing
Using Data Consistency Protocol in Distributed Computing Environments
Design and Evaluation of a Parallel Data Redistribution Component for TGrid
MetaLoRaS: A Predictable MetaScheduler for Non-dedicated Multiclusters
The Cost of Transparency: Grid-Based File Access on the Avaki Data Grid
A Topology Self-adaptation Mechanism for Efficient Resource Location
A Replication-Based Fault Tolerance Protocol Using Group Communication for the Grid
Increasing Availability in a Replicated Partitionable Distributed Object System
Exploiting Multidomain Non Routable Networks
Recovery Strategies for Linear Replication
Mobile Agents Based Collective Communication: An Application to a Parallel Plasma Simulation
Track 4. Software and Applications
Session 7. Software
A Static Parallel Multifrontal Solver for Finite Element Meshes
A Software Architecture Framework for On-Line Option Pricing
An Open Source Web Service Based Platform for Heterogeneous Clusters
Cost Models and Performance Evaluation of Similarity Search in the HON P2P System
Matrix-Based Programming Optimization for Improving Memory Hierarchy Performance on Imagine
On Design and Implementation of Adaptive Data Classification Scheme for DSM Systems
TraceDo: An On-Chip Trace System for Real-Time Debug and Optimization in Multiprocessor SoC
Automatic Performance Optimization of the Discrete Fourier Transform on Distributed Memory Computers
Session 8. Applications
Debugging Distributed Shared Memory Applications
Implications of Memory Performance for Highly Efficient Supercomputing of Scientific Applications
Optimisation of the Parallel Performance of a 3D Device Simulator for HEMTs
Video Shot Extraction on Parallel Architectures
Performance of Real-Time Data Scheduling Heuristics Under Data Replacement Policies and Access Patterns in Data Grids
Multi-site Scheduling with Multiple Job Reservations and Forecasting Methods
Evaluating the Dynamic Behaviour of {\bfseries\itshape PROSA} P2P Network
Towards Fully Adaptive Pipeline Parallelism for Heterogeneous Distributed Environments
Compiler-Directed Energy-Time Tradeoff in MPI Programs on DVS-Enabled Parallel Systems
A GPGPU Approach for Accelerating 2-D/3-D Rigid Registration of Medical Images
Backmatter