You are previewing The Handbook of Research on Scalable Computing Technologies.
O'Reilly logo
The Handbook of Research on Scalable Computing Technologies

Book Description

The Handbook of Research on Scalable Computing Technologies presents ideas and experiences in significant advancements and future challenges of enabling technologies. This Handbook of Research contains valuable findings for those involved with developing programming tools and environments in computing as well as those in related upper-level undergraduate and graduate courses.

Table of Contents

  1. Copyright
  2. Editorial Advisory Board
  3. List of Contributors
  4. Foreword
  5. Preface
  6. Acknowledgment
  7. 1. Grid Architectures and Applications
    1. 1. Pervasive Grid and its Applications
      1. ABSTRACT
      2. INTRODUCTION
      3. CURRENT AND FUTURE RESEARCH TRENDS
      4. APPLICATION OF PERVASIVE GRID
        1. System Overview
        2. Pervasive Grid Platform
        3. Grid Service Provider
        4. Applications
        5. Implementation
        6. Performance Evaluation and Analyses
      5. CONCLUSION
      6. RFFERENCES
      7. KEY TERMS AND DEFINITIONS
    2. 2. Pervasive Grids: Challenges and Opportunities
      1. ABSTRACT
      2. INTRODUCTION
      3. PERVASIVE GRID APPLICATIONS AND THEIR REQUIREMENTS
      4. CROSSCUTTING CHALLENGES
      5. RESEARCH OPPORTUNITIES IN PERVASIVE GRID COMPUTING
      6. RELATED WORK
        1. Research Landscape in Grid and Autonomic Computing
        2. Pervasive Grid Efforts
      7. CONCLUSION
      8. REFERENCES
      9. KEY TERMS AND DEFINITIONS
    3. 3. Desktop Grids: From Volunteer Distributed Computing to High Throughput Computing Production Platforms
      1. ABSTRACT
      2. ORIGINS AND PRINCIPLES
        1. Origins of Desktop Grids
        2. Main Principles
      3. CLASSIFICATION OF DESKTOP GRIDS
        1. Local Desktop Grids
        2. Collaborative Desktop Grids
        3. Internet Volunteer Desktop Grids
          1. Single-Application Internet Volunteer Desktop Grids.
          2. XtremWeb.
          3. BOINC
      4. EVOLUTION OF MIDDLEWARE ARCHITECTURE
        1. Job Management
        2. Resource Scheduling
        3. Volatility Tolerance
        4. Data Management
        5. Security Model
        6. Bridging Service Grids and Desktop Grids
        7. The Superworker Approach
        8. The Gliding-In Approach
        9. Result Certification
      5. RESEARCH AND EXPLORATION TOOLS
        1. Platforms Observations
        2. DG Simulation: SimBOINC
          1. Simulator Overview.
          2. Execution.
          3. Client States File.
          4. Simulation File
          5. Using Availability Traces
          6. Logging
          7. Simulator Output and Performance Metrics.
          8. Use of SimGrid.
      6. APPLICATIONS
        1. Bag of Task Applications
        2. Data Intensive
        3. Long Running Applications
        4. Real-Time Applications
        5. Network-Intensive Applications
      7. CONCLUSION
      8. REFERENCES
      9. KEY TERMS AND DEFINITIONS
        1. Glossary
      10. ENDNOTES
    4. 4. Porting Applications to Grids1
      1. ABSTRACT
      2. INTRODUCTION
      3. APPLICATIONS AND THE GRID INFRASTRUCTURE
        1. Applications and Security
        2. Applications and Resource Management
        3. Applications and Resource Information Services
        4. Applications and Data Management
        5. The Simple API for Grid Applications (SAGA)
      4. GRID APPLICATIONS AND DATA
        1. Shared Data Access
        2. Data Topology
        3. Data Volume
      5. PORTING AND PROGRAMMING GRID APPLICATIONS
        1. Grid Programming Models and Environments
        2. Grid-Enabling Application Programs and Numerical Algorithms
        3. Applications and Grid Portals
          1. Example: The EnginFrame Portal Environment
      6. CASE STUDY: APPLICATIONS ON THE DEISA INFRASTRUCTURE
        1. The DEISA Project
        2. The DEISA Infrastructure Services
        3. DECI - DEISA Extreme Computing Initiative for Supercomputing Applications
        4. Applications Adapted to the DEISA Grid Infrastructure
      7. APPLICATIONS IN THE CLOUD
      8. CONCLUSION: 10 RULES FOR BUILDING A SUSTAINABLE GRID FOR SUSTAINABLE APPLICATIONS
      9. REFERENCES
      10. KEY TERMS AND DEFINITIONS
        1. Glossary
      11. ENDNOTE
    5. 5. Benchmarking Grid Applications for Performance and Scalability Predictions
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
        1. Related Work
        2. Grid Resource Model
        3. Grid Workflow Model
        4. Grid Execution Model
      4. PREDICTION REQUIREMENTS
      5. ARCHITECTURE DESIGN
      6. EXPERIMENTAL DESIGN
        1. Experiment Specification
        2. Training Experiments
        3. Performance Sharing and Translation
        4. Inter-Platform PST
          1. Intra-Platform PST
        5. Reduced Experiment Set
        6. Experiment Execution
        7. Background Load
        8. Performance and Scalability Prédictions
        9. Grid Site Comparisons
      7. EXPERIMENTS
        1. Workflow Applications
          1. WIEN2k
          2. MeteoAG
          3. Invmod
        2. Experiment Set Reduction
        3. Normalized Benchmarks
        4. Grid Site Comparison
        5. Prediction Accuracy
      8. FUTURE TRENDS
      9. CONCLUSION
      10. REFERENCES
      11. KEY TERMS AND DEFINITIONS
  8. 2. P2P Computing
    1. 6. Scalable Index and Data Management for Unstructured Peer-to-Peer Networks
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND AND RELATED WORK
        1. BitTorrent
        2. Blind Search Methods
        3. Index Caching Mechanisms
        4. Interest Group and Data Clustering
        5. Popularity of Content:
      4. AN INDEX DIFFUSION MECHANISM
        1. Request Popularity and Disposition
        2. Network Architecture
          1. Control of Indices
          2. Flooding of Query
        3. Modified Search Methods
      5. SIMULATION RESULTS
      6. DISCUSSIONS AND FUTURE TRENDS
      7. CONCLUSION
      8. REFERENCES
      9. KEY TERMS AND DEFINITIONS
    2. 7. Hierarchical Structured Peer-to-Peer Networks
      1. ABSTRACT
      2. INTRODUCTION
      3. DISTRIBUTED HASH TABLES
      4. HIERARCHICAL DHT
      5. VARYING FREQUENCY OF STABILIZATION
      6. VARYING SIZE OF ROUTING TABLES
        1. Hierarchical Chord
      7. COLLISIONS OF GROUP IDENTIFIERS
      8. COLLISION DETECTION AND RESOLUTION SCHEME
        1. Supernode Initiated
        2. Node Initiated
      9. COLLISION AVOIDANCE SCHEME
        1. Join Protocol
        2. Leave Protocol
        3. Failures
      10. SUMMARY AND OPEN ISSUES
      11. REFERENCES
      12. KEY TERMS AND DEFINITIONS
        1. Glossary
      13. ENDNOTES
    3. 8. Load Balancing in Peer-to-Peer Systems
      1. ABSTRACT
      2. 1. INTRODUCTION
      3. 2. BACKGROUND
      4. 3. LOAD BALANCING METHODS
        1. 3.1 Examples of Load Balancing Methods
          1. 3.1.1 Virtual Server
          2. 3.1.2 ID Assignment or Reassignment
          3. 3.1.3 Load Transfer
        2. 3.2 Case Study: Locality-Aware Randomized Load Balancing Algorithms
          1. 3.2.1 Cycloid: A Constant-Degree DHT
          2. 3.2.2 Load Balancing Framework
          3. 3.2.3 Locality-Aware Randomized Load Balancing Algorithms
          4. 3.2.4 Performance Evaluation
          5. 3.2.6. Summary
      5. 4. FUTURE TRENDS
      6. 5. CONCLUSION
      7. REFERENCES
      8. KEY TERMS AND DEFINITIONS
    4. 9. Decentralized Overlay for Federation of Enterprise Clouds
      1. ABSTRACT
      2. INTRODUCTION
      3. DESIGNING DECENTRALIZED ENTERPRISE CLOUD OVERLAY
      4. ANEKA ENTERPRISE CLOUD: AN OVERVIEW
      5. ANEKA-FEDERATION
      6. DESIGN AND IMPLEMENTATION
      7. CONTENT-BASED DECENTRALIZED CLOUD SERVICES
      8. EXPERIMENTAL EVALUATION AND DISCUSSION
        1. Results and Discussion
      9. RELATED WORK
      10. CONCLUSION AND FUTURE DIRECTIONS
      11. ACKNOWLEDGMENT
      12. REFERENCES
      13. KEY TERMS AND DEFINITIONS
        1. Glossary
      14. ENDNOTE
  9. 3. Programming Models and Tools
    1. 10. Reliability and Performance Models for Grid Computing
      1. ABSTRACT
      2. INTRODUCTION
      3. GRID SERVICE RELIABILITY AND PERFORMANCE
        1. Description of the Grid Computing
        2. Failure Analysis of Grid Service
        3. Grid Service Reliability and Performance
        4. Grid Service Time Distribution and Reliability/Performance Measures
      4. STAR TOPOLOGY GRID ARCHITECTURE
        1. Universal Generating Function
        2. Illustrative Example
      5. TREE TOPOLOGY GRID ARCHITECTURE
        1. Algorithms for Determining the pmf of the Task Execution Time
          1. Minimal Task Spanning Tree (MTST)
          2. pmf of The Task Execution Time
        2. Illustrative Example
          1. The Service MTST
          2. Parameters of MTSTs' Paths
          3. List of MTST Elements
          4. pmf of Task Completion Time
          5. 4.2.5. Calculating the Reliability Indices
        3. Parameterization and Monitoring
      6. CONCLUSION
      7. ACKNOWLEDGMENT
      8. REFERENCES
      9. KEY TERMS AND DEFINITIONS
    2. 11. Mixed Programming Models Using Parallel Tasks
      1. ABSTRACT
      2. INTRODUCTION
      3. TASK-BASED PROGRAMMING APPROACHES
        1. Language Extensions
        2. Skeleton-Based Approaches
        3. Library-Based Approaches
        4. Coordination-Based Approaches
      4. HIERARCHICAL M-TASK PROGRAMMING
        1. TwoL Model
        2. Programming Interface Tlib
      5. EXTENDED PROGRAMMING MODEL
        1. CM-Task Programming Model
        2. Development of CM-Task Programs
      6. SCHEDULING AND MAPPING
      7. EXAMPLE AND EXPERIMENTS
      8. CONCLUSION
      9. REFERENCES
      10. KEY TERMS AND DEFINITIONS
    3. 12. Programmability and Scalability on Multi-Core Architectures
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND STUDY: HARDWARE DESIGNS OF MULTI-CORE ARCHITECTURE
        1. Multi-Core Processor Architecture: Homogeneous or Heterogeneous
        2. Memory Hierarchy Design
      4. EXPLOITING SOFTWARE PARALLELISM ON A MULTI-CORE SYSTEM
        1. Network Coding on Multi-Core Systems
        2. Implementing Intrusion Detection System on Multi-Core Architecture
          1. Overview of IDS (Intrusion Detection System)
          2. Parallelization of Intrusion Detection System
      5. GPU DEVELOPMENT ON MULTI-CORE SYSTEM
        1. Background of 3D Graphics Processing and GPU
        2. Parallelization of 3D Graphic Processing
      6. CONCLUSION
      7. REFERENCES
      8. KEY TERMS AND DEFINITIONS
    4. 13. Assembling of Parallel Programs for Large Scale Numerical Modeling
      1. ABSTRACT
      2. INTRODUCTION
      3. THE PIC METHOD AND THE PROBLEMS OF ITS PARALLEL IMPLEMENTATION
      4. BASIC CONCEPTS OF THE TECHNOLOGY OF FRAGMENTED PROGRAMMING
        1. Algorithm and Program Fragmentation
        2. Assembling vs. Partitioning
        3. Separation of the Fine Grain and the Coarse Grain Computations
        4. Explicitly Two-Level Programming
        5. Automatic Providing of Dynamic Properties of a Program
        6. Computation and Communicating in Parallel
        7. Separation of Semantics and Scheme of Computation
      5. PARALLELIZATION OF NUMERICAL METHODS WITH AT
        1. PIC Parallelization on the line of PEs
        2. PIC parallelization on the 2D grid of PEs
        3. General PIC Method Fragmentation
      6. IMPLEMENTATION OF THE PIC METHOD ON MULTICOMPUTERS
      7. DYNAMIC LOAD BALANCING
        1. Initial Load Balancing
        2. Dynamic Load Balancing
        3. Centralized Dynamic Load Balancing Algorithm
        4. Decentralized Dynamic Load Balancing Algorithm for the Constant Number of Particles
        5. Specialized Decentralized Algorithm
        6. Diffusive Load Balancing Algorithms
      8. AUTOMATIC GENERATION OF PARALLEL CODE
      9. CONCLUSION
      10. REFERENCES
      11. KEY TERMS AND DEFINITIONS
    5. 14. Cell Processing for two Scientific Computing Kernels
      1. ABSTRACT
      2. INTRODUCTION
      3. CELL BROADBAND ENGINE PROCESSOR
      4. FINITE DIFFERENCE TIME DOMAIN ALGORITHM
        1. Finite Difference Time Domain Algorithm
        2. FDTD on Distributed-Memory Machines
        3. FDTD on Homogeneous Multicore Achitecture
        4. FDTD on Cell/B.E. Processor
        5. Experiment Results and Comparisons
        6. FFT
          1. Fast Fourier Transform
        7. Cooley-Tukey Butterfly Network and ISN
        8. Parallel FFT Based on ISN on Cell/B.E.
          1. Experimental Results
      5. CONCLUSION
      6. ACKNOWLEDGMENT
      7. REFERENCES
      8. KEY TERMS AND DEFINITIONS
  10. 4. Scheduling and Communication Techniques
    1. 15. On Application Behavior Extraction and Prediction to Support and Improve Process Scheduling Decisions
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND: BEHAVIOR EXTRACTION APPROACHES
        1. MPI Application Behavior Extraction
        2. Distributed Application Behavior Extraction
      4. APPROACHES FOR APPLICATION BEHAVIOR CLASSIFICATION AND PREDICTION
      5. APPLICATIONS
      6. CONCLUSION
      7. REFERENCES
      8. KEY TERMS AND DEFINITIONS
    2. 16. A Structured Tabu Search Approach for Scheduling in Parallel Computing Systems
      1. ABSRACT
      2. INTRODUCTION
      3. TASK SCHEDULING AND PROBLEM FORMULATION
      4. TABU SEARCH
        1. Short Term Memory
        2. Longer Term Memory
          1. Intensification
          2. Diversification
        3. Implementation
        4. Design Overview
        5. Initialization
        6. Initial Solution
        7. Neighborhood
        8. Candidate List
        9. Tabu List
        10. Evaluation Functions
        11. Execution Trace
        12. RESULTS: COMPUTATION-INTENSIVE CASES
        13. Test Parameters
        14. Performance Results
          1. Comparisons with Differing Evaluation Criteria
        15. Comparisons with Number of Processors
          1. Minimizing Communication
          2. Comparisons with Various Graph Features
        16. Results: Communication-Intensive Cases
        17. Test Parameters
        18. Performance Results
        19. Comparisons with Differing Evaluation Criteria
        20. Comparisons with Number of Processors
        21. Minimizing Communication
      5. CONCLUSION
      6. REFERENCES
      7. KEY TERMS AND DEFINTIONS
    3. 17. Communication Issues in Scalable Parallel Computing1
      1. ABSTRACT
      2. INTRODUCTION
      3. PARALLEL COMPUTATION MODEL
      4. THE STRING SIMILARITY PROBLEM
      5. THE BASIC SIMILARITY ALGORITHM
        1. Experimental Results of the Basic Algorithm
      6. THE IMPROVED SIMILARITY ALGORITHM
        1. Experimental Results of the Improved Similarity Algorithm
        2. Quadratic vs. Linear Space Implementation
      7. CONCLUSION
      8. REFERENCES
      9. KEY TERMS AND DEFINITIONS
        1. Glossary
      10. ENDNOTE
    4. 18. Scientific Workflow Scheduling with Time–Related QoS Evaluation
      1. ABSTRACT
      2. INTRODUCTION
      3. PRELIMINARY KNOWLEDGE OF QOS
        1. A Temporal Model of Service-Driven Scientific Workflow Execution
      4. APPLICATION CONTEXT ANALYSES OF SCIENTIFIC WORKFLOW EXECUTION
        1. A Temporal Reasoning Rule for Collaborative Scheduling Application of Scientific Workflow
      5. EVALUATION: A SCIENTIFIC WORKFLOW ENGINEERING DEVELOPMENT WITH TIME-RELATED QOS EVALUATION
      6. RELATED WORKS AND COMPARISON ANALYSIS
      7. CONCLUSION
      8. ACKNOWLEDGMENT
      9. REFERENCES
      10. KEY TERMS AND DEFINITIONS
  11. 5. Service Computing
    1. 19. Grid Transaction Management and Highly Reliable Grid Platform
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
      4. MAIN FOCUS OF THE CHAPTER
        1. Grid Transaction Service and Highly Reliable Grid Platform
        2. Layered Architecture
      5. HOW TO USE GRIDTS
        1. Grid Transaction Processing Framework and Flow
      6. TRANSACTION COORDINATION
        1. Transaction Coordination for Atomic Grid Transaction
        2. Long-Lived Grid Transaction
      7. AUTOMATIC GENERATION OF COMPENSATING TRANSACTIONS
        1. Key Technologies for Automatic Generation of Compensating Transaction
        2. Set Compensating Rules
        3. Generate Compensating Operations
        4. Generate and Call Compensating Transactions
        5. Handle Noncompensable Transaction
      8. FUTURE TRENDS
      9. CONCLUSION
      10. ACKNOWLEDGMENT
      11. REFERENCES
      12. KEY TERMS AND DEFINITIONS
    2. 20. Error Recovery for SLA – Based Workflows Within the Business Grid
      1. ABSTRACT
      2. INTRODUCTION
        1. Grid-Based Workflow Model
        2. Grid Service Model
        3. Business Model
        4. Problem Statement
          1. The Large-Scale Error
        5. The Small-Scale Error
      3. RELATED WORKS
      4. ERROR RECOVERY FRAMEWORK
        1. Recovery the Large-Scale Error
          1. Checking Workflows having only Independent Sub-Jobs Affected
          2. Determining the Re-Mapping Priority
          3. Determining Sub-Jobs which need to be Re-Planned
        2. Recovery the Small-Scale Error
          1. Phase 1: Re-Mapping the Directly Affected Sub-Jobs
          2. Phase 2: Re-Mapping the Workflow to Meet the Pre-Determined Deadline
          3. Phase 3: Re-Mapping the Workflow to have Minimal Runtime
        3. Recovery Procedure
      5. RE-MAPPING ALGORITHMS
        1. Formal Mapping Problem Statement
        2. w-Tabu Algorithm
          1. Generating Referent Solution Set
          2. Solution Improvement Algorithm
            1. The Neighborhood Set Structure
            2. The Assigning Sequence of the Workflow
            3. Computing the Timetable Procedure
            4. The Modified Tabu Search Procedure
          3. Performance of w-Tabu Algorithm
        3. H-Map Algorithm
          1. Constructing the Set of Initial Configurations
          2. Improving Solution Quality Algorithm
          3. Performance of H-Map Algorithm
        4. G-Map Algorithm
          1. Refining the Solution Space
          2. Constructing the Set of Initial Configurations
          3. Determining the Assigning Order
          4. Checking the Feasibility of a Solution
          5. Improving Solution Quality Algorithm
          6. Performance of the G-Map Algorithm
      6. PERFORMANCE EXPERIMENT
        1. Large-Scale Error Recovery Experiment
        2. Small-Scale Error Recovery Performance
          1. The Effectiveness of the Error Recovery Mechanism
          2. The Effect of the Late Period to the Recovery Process
      7. FUTURE RESEARCH DIRECTION
      8. CONCLUSION
      9. REFERENCES
      10. KEY TERMS AND DEFINITIONS
        1. Glossary
      11. ENDNOTE
    3. 21. A Fuzzy Real Option Model to Price Grid Compute Resources
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
      4. RESEARCH METHODOLOGY
        1. Model Assumptions and Formulation
        2. Price Variant Factor
        3. Real Option Discretization of Trinomial Process
      5. GRID COMPUTE RESOURCES PRICING
        1. Fuzzy Logic Framework
      6. CONCLUSION AND FUTURE WORK
      7. REFERENCES
      8. KEY TERMS AND DEFINITIONS
        1. Glossary
      9. ENDNOTES
    4. 22. The State of the Art and Open Problems in Data Replication in Grid Environments
      1. ABSTRACT
      2. INTRODUCTION
      3. OVERVIEW OF REPLICATION IN DATA GRID SYSTEMS
        1. The Replication Process and Components
        2. Issues and Challenges in Data Replication
        3. Replica Placement
        4. Replica Selection
        5. Replica Consistency
        6. The Impact of Data Replication on Job Scheduling
      4. DATA REPLICATION: STATE OF THE ART
        1. Replica Placement Strategies
          1. Algorithme Focusing on Access Latency and Bandwidth Consumption
        2. Algorithms Focusing on System Reliability and Availability
        3. Algorithms Focusing on Overall Grid Performance
        4. Replica Selection Algorithms
        5. Algorithms Based on Selecting the Best Replica
        6. Algorithms Using Co-Allocation Mechanism
        7. Replica Consistency Algorithms
        8. Impact of Data Replication on Job Scheduling
      5. COMPARISON OF REPLICA PLACEMENT STRATEGIES
        1. Comparison of Replica Placement Algorithms
      6. SUMMARY AND OPEN PROBLEMS
        1. Fragmented Replication
        2. Algorithms that are Adaptive to Performance Variation
        3. Enhanced Algorithms Combining Computation and Data Scheduling
        4. New Models of Grid Architecture
        5. Increased Collaboration Using VO-based Data Grids
      7. CONCLUSION
      8. REFERENCES
      9. KEY TERMS AND THEIR DEFINITIONS
        1. Glossary
      10. ENDNOTE
    5. 23. Architectural Elements of Resource Sharing Networks
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
      4. CLASSIFICATIONS FOR GRID RESOURCE MANAGEMENT SYSTEMS
        1. Architecture and Operational Models of GRMSs
        2. Arrangements Between Brokers in Resource Sharing Networks
        3. Resource Control Techniques
        4. Taxonomy on Virtual Organisations
      5. A SURVEY OF EXISTING WORK
        1. Distributed Architecture Based Systems
        2. Hierarchical Systems, Brokers and Meta-Scheduling
        3. Inter-Operation of Resource Sharing Networks
        4. Virtual Organisations
          1. Enabling Technology
          2. Facilitators or Breeding Environments
          3. Other Important Work
      6. MAPPING OF SURVEYED WORK AGAINST THE TAXONOMIES
      7. FUTURE TRENDS
      8. CONCLUSION
      9. ACKNOWLEDGMENT
      10. REFERENCES
      11. ENDNOTES
  12. 6. Optimization Techniques
    1. 24. Simultaneous MultiThreading Microarchitecture
      1. ABSTRACT
      2. INTRODUCTION
        1. Moore's Law and Memory Wall
        2. Overcoming the Limits of Instruction-Level Parallelism
      3. BACKGROUND
        1. Multithreading Microarchitectures
        2. Hardware-Supported Multithreading
        3. Chip Multi-Processing
        4. Simultaneous MultiThreading
        5. Commercial Implementation of SMT
      4. SMT DESIGN ASPECTS
        1. Thread Selection Policy
        2. Handing Long-Latency Instructions
        3. Resource Partitioning among Multiple Threads
        4. SMT Synchronization and Communication
      5. POTENTIAL BENEFITS OF SMT
        1. SMT for Fault-Tolerance
        2. Lowering the Performance Overhead
        3. Deadlock Analysis and Prevention
        4. SMT for Secure Communication
      6. TRENDS AND CONLUSIONS
      7. REFERENCES
      8. KEY TERMS AND DEFINITIONS
        1. Glossary
      9. ENDNOTES
    2. 25. Runtime Adaption Techniques for HPC Applications
      1. ABSTRACT
      2. INTRODUCTION
        1. Motivating Example
        2. Problem Description
        3. Specific Goals
      3. BACKGROUND
        1. Static Tuning of Applications
        2. Dynamic Tuning of Applications
          1. FFTW
          2. STAR-MPI
          3. SALSA
      4. THE ABSTRACT DATA AND COMMUNICATION LIBRARY
        1. Description of the ADCL API
        2. Technical Concept
          1. Distributed Decision Logic
          2. Attribute Based Tuning
          3. Ongoing Research
      5. PERFORMANCE EVALUATION
        1. Optimizing 3-D Nearest Neighbor Communication
          1. ADCL Performance Results
        2. Tuning a Parallel Matrix-Matrix Multiply Kernel
        3. Using ADCL to Tune System Parameters of Open MPI
      6. CONCLUSION
      7. REFERENCES
      8. KEY TERMS AND DEFINITIONS
    3. 26. A Scalable Approach to Real-Time System Timing Analysis
      1. ABSTRACT
      2. INTRODUCTION
        1. Partitionable Analysis
        2. Evolvable Analysis
        3. Timing Analysis without System Integration and Implementation Details
        4. Timing Analysis for an Evolving System Definition
        5. Abstract Timing Analysis Model
        6. Real-Time Transaction Topology
        7. Real-Time Transaction Properties
        8. End-to-End Timing Analysis
        9. Timing Model Initialisation and Finalisation
        10. Example Transaction Definition and Decomposition
      3. RATE-BASED EXECUTION MODEL
        1. Basic Schedule Implementation Scheme
        2. Example Application of Basic Cyclic Scheme
        3. Bandwidth Server-Based Implementation Scheme
        4. Example Application of Server-Based Scheme
        5. Introducing Priorities to Improve Resource Bandwidth Allocation and System Responsiveness
        6. Introducing High Priority Activities for Improved Responsiveness
        7. Introducing Low Priority Activities for Improved Bandwidth Allocation
        8. Example of Reclaiming Over-Allocated Bandwidth
      4. RELATED WORK
      5. SUMMARY
      6. REFERENCES
      7. ENDNOTE
    4. 27. Scalable Algorithms for Server Allocation in Infostations
      1. ABSTRACT
      2. INTRODUCTION
      3. COMPUTATIONAL INTRACTABILITY
      4. ALGORITHM FOR INTERVAL COLORING WITH BOUNDED OVERLAPPING
      5. ALGORITHM FOR WEIGHTED INTERVAL COLORING
      6. FURTHER GENERALIZATIONS
      7. CONCLUSION
      8. REFERENCES
      9. KEY TERMS
  13. 7. Web Computing
    1. 28. Web Application Server Clustering with Distributed Java Virtual Machine1
      1. ABSTRACT
      2. INTRODUCTION
      3. EXISTING WEB APPLICATION CLUSTERING TECHNOLOGIES
      4. JESSICA2 DISTRIBUTED JVM
      5. APACHE TOMCAT ON DISTRIBUTED JVM
      6. GLOBAL OBJECT SPACE
        1. 5.1 Overview of the Extended JVM Heap
        2. 5.2 Home-based Lazy Release Consistency Protocol
        3. 5.3 Volatile Consistency Protocol
      7. PERFORMANCE ANALYSIS
        1. 6.1 Experimental Setup
        2. 6.2 Scalability Study
        3. 6.3 Comparison with Existing Tomcat Clustering
        4. 6.4 Effect of Implicit Cooperative Caching
        5. 6.5 GOS Overhead Breakdowns
      8. RELATED WORK
      9. CONCLUSION AND FUTURE WORK
      10. REFERENCES
      11. KEY TERMS AND DEFINITIONS
        1. Glossary
      12. ENDNOTE
    2. 29. Middleware for Community Coordinated Multimedia
      1. ABSTRACT
      2. INTRODUCTION
      3. DEFINITIONS
      4. USAGE SCENARIO
      5. CCM MIDDLEWARE REQUIREMENTS
      6. MIDDLEWARE ARCHITECTURE FOR CCM
      7. SURVEY IN MIDDLEWARE FOR CCM
      8. MIDDLEWARE FOR MOBILITY-AWARE CCM
        1. Limitations and Requirements
      9. STATE OF THE ART IN MIDDLEWARE FOR MOBILITY-AWARE CCM
      10. MIDDLEWARE FOR MULTIMEDIA-DRIVEN CCM
      11. STANDARDS FOR MULTIMEDIA REPRESENTATION AND COMPRESSION
      12. MIDDLEWARE FOR MULTIMEDIA COMMUNICATION
      13. MIDDLEWARE FOR SERVICE-ORIENTED CCM
      14. MIDDLEWARE FOR COMMUNITY-COORDINATED CCM
      15. CLASSIFICATION OF COMMUNITIES
      16. REQUIREMENTS FOR COMMUNITY MIDDLEWARE
      17. SURVEY ON MIDDLEWARE TECHNOLOGY FOR COMMUNITY-COORDINATED CCM
      18. FUTURE TRENDS
      19. CONCLUSION
      20. ACKNOWLEDGMENT
      21. REFERENCES
      22. KEY TERMS AND DEFINITIONS
        1. Glossary
      23. ENDNOTE
  14. 8. Mobile Computing and Ad Hoc Networks
    1. 30. Scalability of Mobile Ad Hoc Networks
      1. ABSTRACT
      2. INTRODUCTION
      3. MANAGEMENT OF MANET
        1. The Set of Management Operations for MANET
        2. Solutions for the Management of IP-based MANET
        3. MANETconf
        4. The Management of Non-IP MANET
        5. The MANET Management Operations
      4. CLUSTER COMPUTING ON MANET
        1. Global High Performance Mobile Computing
        2. Localised Mobile Parallel Computing
      5. CONCLUSION
      6. REFERENCES
      7. KEY TERMS AND DEFINITIONS
    2. 31. Network Selection Strategies and Resource Management Schemes in Integrated Heterogeneous Wireless and Mobile Networks
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
      4. NETWORK SELECTION STRATEGIES AND RESOURCE MANAGEMENT SCHEMES
        1. Cost-Function-Based Network Selection Strategies
          1. System Model
        2. Cost-Function-Based Network Selection Strategy
        3. Numerical Results
      5. RESOURCE MANAGEMENT SCHEMES FOR MULTIPLE TRAFFIC
        1. System Model
        2. Preemption-Based Resource Management Scheme
        3. Fairness Between Voice and Data Traffic
        4. Numerical Results
      6. FUTURE TRENDS
      7. CONCLUSION
      8. REFERENCES
      9. KEY TERMS AND DEFINITIONS
  15. 9. Fault Tolerance and QoS
    1. 32. Scalable Internet Architecture Supporting Quality of Service (QoS)
      1. ABSTRACT
      2. INTRODUCTION
      3. DESIGN OBJECTIVES
      4. AUTONOMOUS SYSTEM (AS) RELATIONSHIPS AND NETWORK POLICIES
        1. Internet Connectivity
        2. Border Gateway Protocol (BGP) and AS Relationships
      5. THREE-LAYER POLICY ARCHITECTURE FOR THE INTERNET
        1. Layer-1: Device Layer QoS
        2. Layer-2: Network Layer QoS
        3. Layer-3: End-to-End QOS
      6. CONCLUSION
      7. REFERENCES
      8. KEY TERMS AND DEFINITIONS
    2. 33. Scalable Fault Tolerance for Large-Scale Parallel and Distributed Computing
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
        1. FT-MPI Overview
        2. FT-MPI Semantics
        3. FT-MPI Usage
      4. SCALABLE DISKLESS CHECKPOINTING FOR LARGE SCALE SCIENTIFIC COMPUTING
        1. Diskless Checkpointing: From an Application Point of View
          1. Checksum-Based Checkpointing
          2. Overhead and Scalability Analysis
        2. A Scalable Checkpoint Encoding Algorithm for Diskless Checkpointing
          1. Pipelining
          2. Chain-Pipelined Encoding for Diskless Checkpointing
        3. Overhead and Scalability Analysis
        4. Coding to Tolerate Multiple Simultaneous Process Failures
          1. The Basic Weighted Checksum Scheme
          2. One Dimensional Weighted Checksum Scheme
      5. CHECKPOINT-FREE FAULT TOLERANCE FOR MATRIX MULTIPLICATION
        1. Maintaining Checksum at the End of Computation
        2. Is the Checksum Maintained During Computation?
        3. An Algorithm That Maintains the Checksum during Computation
      6. PRACTICAL NUMERICAL ISSUES
        1. Numerical Stability of Real Number Codes
        2. Numerically Good Real Number Codes Based on Random Matrices
      7. EXPERIMENTAL EVALUATION
        1. Performance of the Chain-Pipelined Checkpoint Encoding
        2. Performance of the Algorithm-Based Checkpoint-Free Fault Tolerance
      8. CONCLUSION AND FUTURE WORK
      9. REFERENCES
      10. KEY TERMS AND DEFINITIONS
  16. 10. Applications
    1. 34. Efficient Update Control of Bloom Filter Replicas in Large Scale Distributed Systems
      1. ABSTRACT
      2. INTRODUCTION TO BLOOM FILTERS
      3. APPLICATIONS OF BLOOM FILTERS IN DISTRIBUTED SYSTEMS
      4. FALSE RATES IN THEORY
        1. False Rates of Bloom Filter Replicas
        2. Overall False Rates
      5. EXPERIMENTAL VALIDATION
      6. REPLICA UPDATE PROTOCOL
      7. RELATED WORK
      8. CONCLUSION
      9. ACKNOWLEDGMENT
      10. REFERENCES
      11. KEY TERMS AND DEFINITIONS
    2. 35. Image Partitioning on Spiral Architecture
      1. ABSTRACT
      2. INTRODUCTION
      3. RELATED WORK
        1. Spiral Architecture
        2. Spiral Addressing
        3. Mathematical Operations on Spiral Architecture
        4. Spiral Addition
        5. Spiral Multiplication
        6. Case 1: a is Not a Multiple of 10
        7. Case 2: a is a Multiple of 10
        8. Mimicking Spiral Architecture
      4. IMAGE PARTITIONING ON SPIRAL ARCHITECTURE
        1. General Image Partitioning on Spiral Architecture
        2. Uniform Image Partitioning on Spiral Architecture
        3. Computing the Number of Partitions
        4. Complete Image Partitioning in Spiral Architecture
      5. EXPERIMENTS
      6. CONCLUSION
      7. REFERENCES
      8. KEY TERMS AND DEFINITIONS
        1. Glossary
      9. ENDNOTE
    3. 36. Scheduling Large–Scale DNA Sequencing Applications
      1. ABSTRACT
      2. INTRODUCTION
      3. RELATED WORK
      4. PROBLEM FORMULATION
      5. LOAD SCHEDULING STRATEGIES AND ANALYSIS
      6. RESULTS AND DISCUSSION
        1. Variable Link Speed
        2. Constant Link Speed
      7. CONCLUSION
      8. REFERENCES
      9. KEY TERMS AND DEFINITIONS
        1. Glossary
      10. ENDNOTE
    4. 37. Multi-Core Supported Deep Packet Inspection
      1. ABSTRACT
      2. 1. INTRODUCTION
      3. 2. RELATED WORK
        1. 2.1 Development of Multi-Core Processors
        2. 2.2 Parallel Network Security Applications on Multi-Core Processors
        3. 2.3 Parallel Deep Packet Inspection
      4. 3. SYSTEM DESIGN
      5. 4. PACKET-LEVEL AND FLOW-LEVEL PARALLELIZATIONS
      6. 5. PARALLEL STRING-MATCHING
      7. 6. DEEPER THINKING OF USING MULTI-CORE
      8. 7. CONCLUSION
      9. REFERENCES
      10. KEY TERMS AND DEFINITIONS
    5. 38. State-Carrying Code for Computation Mobility
      1. ABSTRACT
      2. INTRODUCTION
        1. Computation Mobility
          1. Process Migration
          2. Thread Migration
          3. Checkpointing
          4. Virtual Machines
          5. Mobile Agents
        2. Deployment of SCC
          1. Granularity
          2. Positioning Computation States
          3. Issues in Application-level Approach
          4. Infrastructure
          5. State Construction
          6. Data Conversion Schemes
          7. State Restoration
          8. Safety Issues
          9. Adaptation Point Analysis
          10. Communication States
        3. Future Trends
          1. Resource Access Transparency
          2. Migration Policy
          3. New Languages
      3. CONCLUSION
      4. REFERENCES
      5. KEY TERMS AND DEFINITIONS
  17. Compilation of References
  18. About the Contributors