Euro-Par 2011: Parallel Processing Workshops
CCPI, CGWS, HeteroPar, HiBB, HPCVirt, HPPC, HPSS, MDGS, ProPer, Resilience, UCHPC, VHPC, Bordeaux, France, August 29 -- September 2, 2011, Revised Selected Papers, Part I
Published on 14. April 2012
XXIX, 524 pages
978-3-642-29737-3 (ISBN)
Description
This book constitutes thoroughly refereed post-conference proceedings of the workshops of the 17th International Conference on Parallel Computing, Euro-Par 2011, held in Bordeaux, France, in August 2011. The papers of these 12 workshops CCPI, CGWS, HeteroPar, HiBB, HPCVirt, HPPC, HPSS HPCF, PROPER, CCPI, and VHPC focus on promotion and advancement of all aspects of parallel and distributed computing.
More details
Other editions
Additional editions
Michael Alexander | Pasqua D'Ambra | Adam Belloum
Euro-Par 2011: Parallel Processing Workshops
CCPI, CGWS, HeteroPar, HiBB, HPCVirt, HPPC, HPSS, MDGS, ProPer, Resilience, UCHPC, VHPC, Bordeaux, France, August 29 -- September 2, 2011, Revised Selected Papers, Part I
Book
05/2012
1st Edition
Springer
€53.49
Shipment within 7-9 days
Content
- Title
- Preface
- Organization
- Table of Contents - Part I
- CCPI 2011: Workshop on Cloud Computing Projects and Initiatives
- CCPI 2011: Workshop on Cloud Computing Projects and Initiatives
- Towards Cross-Platform Cloud Computing
- Introduction
- Background and Context
- Proposed Abstractions
- ADAPAS Overview
- Application Model
- Resource Model
- Application Execution Model
- ADAPAS Design Approach
- Virtually Unified Capabilities
- Dependency-Capability Objects
- Adapters
- Contexts
- Summary
- References
- QoS Monitoring in a Cloud Services Environment: The SRT-15 Approach
- Introduction
- QoSMONaaS: Quality of Service MONitoring as a Service
- QoSMONaaS Internals
- Case Study
- Conclusions and Directions of Future Work
- References
- Enabling e-Science Applications on the Cloud with COMPSs
- Introduction
- COMPSs and the VENUS-C Platform
- Evaluation of an e-Science Application
- Porting of the Application
- The COMPSs Application Interface
- Performance Analysis
- Conclusions
- References
- OPTIMIS and VISION Cloud: How to Manage Data in Clouds
- Introduction
- Related Work
- Infrastructure Model
- The OPTIMIS Approach
- The VISION Cloud Approach
- Data and Account Model
- Data Management
- Comparison
- Conclusion
- References
- Integrated Monitoring of Infrastructures and Applications in Cloud Environments
- Introduction
- Technological Background
- The Lattice Framework
- Portability Issues
- Architectural Organization
- Implementation of Infrastructure Oriented Probes
- Implementation of Data Platform Oriented Probes
- Startup and Base Message Tagging Rules
- Implementation of the Optimized-Transmission Service
- Implementation of the Log Service
- Summary
- References
- Towards Collaborative Data Management in the VPH-Share Project
- Introduction
- VPH-Share Infostructure
- Patient Avatar
- VPH-Share Workflows
- VPH-Share Cloud Infrastructure
- VPH-Share Data Infrastructure
- VPH-Share Data Sources
- Data Services
- VPHSemantics
- Related Work
- Conclusion
- References
- SLM and SDM Challenges in Federated Infrastructures
- Setting the Scene
- The Nature of Federated Infrastructures
- The ICT Management Challenge
- Why Federation Is Necessary?
- Examples of Federated Infrastructures
- The State of Federated SLM
- The SLM Problem
- Modelling Federated SLM Relationships
- Enforcement, Inducement and Penalisation
- Conclusions and Future Work
- References
- Rapid Prototyping of Architectures on the Cloud Using Semantic Resource Description
- Introduction
- Description and Knowledge Levels
- The Need for SOA
- The Need for Semantics
- Semantic Resource Description and Resource Management
- Lack of Information Effects
- Semantically Augmented Resources
- Managing Resources
- Representing Resources as Services
- Managed Resource Framework (MRF)
- Implementation
- SEROM
- Under the Hood
- Related Works
- Conclusion
- References
- Cloud Patterns for mOSAIC-Enabled Scientific Applications
- Introduction
- Cloud Use Cases and Cloud Patterns
- Cloud Computing Use Cases
- Usage Scenario: A mOSAIC Perspective
- Execution Scenarios for mOSAIC-Enabled Applications
- Intelligent Maintenance System Execution Scenarios
- Model Exploration and Service Level Agreement Execution Scenarios
- Cloud Patterns for mOSAIC-Enabled Applications
- Conclusions
- References
- Enhancing an Autonomic Cloud Architecture with Mobile Agents
- Introduction
- An Autonomic Engine for Managing Clouds
- Gathering Resource Metadata
- Requirements for Monitoring
- Introducing Mobile Agents
- The mAGDA Toolset
- Supporting the Knowledge Function with mAGDA
- Supporting the Monitoring Function with mAGDA
- An Example Use Case: The Cloud@Home System
- Related Work
- Conclusions
- References
- Mapping Application Requirements to Cloud Resources
- Introduction
- Paper Organisation
- Cloud Resource Management
- Financial Use Case Scenarios
- Application Requirement's Challenges
- Using Constraints to Discover Cloud Resources
- Demonstrator
- Related Works
- Conclusion and Future Works
- References
- CoreGRID/ERCIM Workshop on Grids, Clouds and P2P Computing - CGWS2011
- CoreGRID/ERCIM Workshop on Grids, Clouds and P2P Computing - CGWS2011
- A Perspective on the CoreGRID Grid Component Model
- Towards Scheduling Evolving Applications
- Introduction
- Related Work
- Problem Statement
- Definitions and Notations
- An RMS for Fully-Predictably Evolving Applications
- Formal Problem Statement
- Scheduling Fully-Predictably Evolving Applications
- An Algorithm for Offline Scheduling of Evolving Applications
- The fit Function
- Discussions
- Evaluation
- Description of Experiments
- Analysis
- Conclusions
- References
- Model Checking Support for Conflict Resolution in Multiple Non-functional Concern Management
- Introduction
- Rule Based Management
- Multiple Non-functional Concern Management
- Fully Coordinated Co-design
- Coordinated Decision Commitment
- Ex Post Independent Manager Coordination
- Model Checking for Conflict Resolution
- Preliminary Results
- Future Work and Conclusions
- References
- Consistent Rollback Protocols for Autonomic ASSISTANT Applications
- Introduction
- The ASSISTANT Programming Model
- Consistent Reconfiguration Protocols
- Formalization
- Implementation
- Description of Reconfiguration Protocols
- Experiments
- Conclusions
- References
- A Dynamic Resource Management System for Real-Time Online Applications on Clouds
- Introduction and Related Work
- Scalable ROIA Development with RTF
- Resource Management for ROIA on Clouds
- The RTF-RMS Architecture
- A Load-Balancing Strategy for ROIA on Clouds
- Cloud Resource Allocation
- Experiments
- Conclusion
- References
- Cloud Federations in Contrail
- Introduction
- Contrail Federations: Motivations and Commitments
- Federation Architecture
- Core Layer
- Adapters Layer
- Related Work
- Conclusions
- References
- Semi-automatic Composition of Ontologies for ASKALON Grid Workflows
- Introduction
- Prerequisite
- Methodology
- Parsing and Canonicalization
- Creation of an Intermediate Ontology
- Semantic Association and Meaning Specification
- Experiments
- Ontology Evaluation
- Performance Measurements for Ontology Synthesis Process
- Related Work
- Conclusion
- References
- The Chemical Machine: An Interpreter for the Higher Order Chemical Language
- Introduction
- The Chemical Computational Model
- The Concept of the Chemical Interpreter
- Implementation
- Experiences Learned
- Conclusions
- References
- Design and Performance of the OP2 Library for Unstructured Mesh Applications
- Introduction
- Related Work
- OP2
- OP2 Code Generation and Compiler Framework
- Performance
- Conclusion
- References
- Mining Association Rules on Grid Platforms
- Introduction
- Related Work
- Load Balancing: Problem Definition
- The Grid Model
- The Dynamic Load Balancing Strategy
- The Dynamic Load Balancing Algorithms
- Performance Evaluation
- Conclusion
- References
- 5th Workshop on System-Level Virtualization for High-Performance Computing (HPCVirt 2011)
- 5th Workshop on System-Level Virtualization for High Performance Computing (HPCVirt 2011)
- Performance Evaluation of HPC Benchmarks on VMware's ESXi Server
- Introduction
- ESXi Server Architecture
- Virtual NUMA Support
- Experimental Evaluation
- SPEC MPI Results
- SPEC OMP Results
- Conclusion
- References
- Virtualizing Performance Counters
- Introduction
- x86 Performance Counter Hardware
- Profilers
- Time-Sharing Physical CPUs
- Trapping and Emulation
- Speculative and Non-speculative Events
- Combining Native and Emulated Performance Counters
- Hybrid Performance Counter Example
- Performance Counter Hardware Virtualization Methods
- Discussion
- Future Work
- Related Work
- Conclusion
- References
- A Case for Virtual Machine Based Fault Injection in a High-Performance Computing Environment
- Introduction
- Background
- Motivation
- HPC Environments
- Batch Allocation Systems
- Executable Linkage
- Performance Sensitive
- Exotic Hardware
- Design
- Front-End and Distributed Control
- Experiment Setup/Management
- Monitoring and Event Logging
- Fault Injection Mechanisms
- Conclusion
- References
- HPPC 2010: 5th Workshop on Highly Parallel Processing on a Chip
- HPPC 2010: 5th Workshop on Highly Parallel Processing on a Chip
- Thermal Management of a Many-Core Processor under Fine-Grained Parallelism
- Introduction
- Experimental Platform
- The XMT Architecture
- Simulation Environment
- Performance under Thermal Constraints
- Benchmarks
- Dynamic Thermal Management
- Analysis of DTM Results
- Conclusion
- References
- Mainstream Parallel Array Programming on Cell
- Introduction
- Related Work
- Implicit Parallelism Using the `F' Programming Language
- A Language Primer
- The E Compiler
- Targeting Offload C++
- A C++ Template Interface to Fortran Runtime Libraries
- Parallel Operational Semantics
- The Benchmark Programs
- Blackscholes
- Swaptions
- Mandelbrot
- The n-Body Problem
- Experimental Evaluation
- Conclusion
- References
- Generating GPU Code from a High-Level Representation for Image Processing Kernels
- Introduction
- Image Processing Framework
- Library
- Compiler
- Results
- Vertical Mean Filtering
- OpenCV Library
- Future Work
- Conclusion
- References
- A Greedy Heuristic Approximation Scheduling Algorithm for 3D Multicore Processors
- Introduction
- 3D NoC with through Silicon via Constraints
- Processors and Caches
- Constraints of the through Silicon via
- The Scheduling Algorithm
- Experimental Evaluation
- Experiment Setup and Application
- Result Analysis
- Conclusion and Future Work
- References
- Algorithms and Programming Tools for Next-Generation High-Performance Scientific Software HPSS 2011
- Algorithms and Programming Tools for Next-Generation
- European Exascale Software Initiative: Numerical Libraries, Solvers and Algorithms
- Introduction
- EESI: European Exascale Software Initiative
- Numerical Libraries, Solvers and Algorithms
- Main Areas Covered in the WG 4.3 Report
- Algorithmic Issues
- Software Issues
- Fault Tolerance
- Uncertainty Quantification
- Programmability
- Floating-Point Issues
- Training
- Conclusions
- On Reducing I/O Overheads in Large-Scale Invariant Subspace Projections
- Introduction
- The Strutures of H and Z Matrices
- Task Decomposition for Parallel Processing
- An Out-of-Core Approach
- Distributed In-Core Approach
- Future Challenges
- Conclusions
- References
- Enabling Next-Generation Parallel Circuit Simulation with Trilinos
- Motivation
- Background
- Xyce-Trilinos Interface
- Next-Generation Circuit Simulation
- Computational Setup
- Epetra
- ShyLU
- Zoltan
- Conclusion
- References
- DAG-Based Software Frameworks for PDEs
- Introduction
- A Brief Survey of Direct Acyclic Graph Approaches
- Overview of Uintah Software
- Uintah as Viewed through the Silver Model
- The Silver Model and Applications
- Using the DAG Approach at the Multi-core Level with Wasatch
- Summary
- References
- On Partitioning Problems with Complex Objectives
- Introduction
- Problem Definition
- Related Work and Contributions
- Background
- Hypergraph Partitioning
- Recursive Bisection Based Hypergraph Partitioning
- A Framework for Complex Partitioning Objectives
- The Singly Bordered Form
- The Block Form
- The Doubly Bordered Form
- Experiments
- The Singly Bordered Form
- The Block Form
- Conclusion
- References
- A Communication-Avoiding Thick-Restart Lanczos Method on a Distributed-Memory System
- Introduction
- Thick-Restart Lanczos Method
- Communication-Avoiding TRLan
- Krylov Subspace Generation
- Orthogonalization
- Tridiagonal Matrix Computation
- Numerical Experiments
- Conclusion
- References
- Spherical Harmonic Transform with GPUs
- Introduction
- Spherical Harmonic Transforms
- Algebraic Background
- Current Approach
- ALM2MAP with CUDA
- Optimizations for GPU
- Experimental Results
- Performance of ?m Computation
- Overall Performance
- Conclusions
- References
- Design Patterns for Scientific Computations on Sparse Matrices
- Introduction
- Design Patterns
- ``STATE'' Is Your Friend
- ``PROTOTYPE'' and ``BUILDER'' Are Good Ideas
- An Example: Adding Support for NVIDIA GPUs
- Sparse Matrix Computation on GPU
- Performance Results
- Conclusions
- References
- High-Performance Matrix-Vector Multiplication on the GPU
- Introduction
- Related Work
- Matrix-Vector Multiplication Kernels
- One Thread per Row
- Several Threads per Row
- Several Rows per Thread
- Several Threads, Several Columns
- Several Threads, Several Rows
- Results
- Auto-tuner Results
- Performance Results
- Performance Comparison
- Conclusion
- References
- Relaxed Synchronization with Ordered Read-Write Locks
- Introduction
- Ordered Read-Write Locks, an Adaptive Tool for Synchronization
- UserInterfaces
- Resource Protection
- Associating Data
- Recurrent Access to Resources
- Experiments
- Benchmark Application
- Experimental Results
- Conclusion and Future Work
- References
- The Parallel C++ Statistical Library 'QUESO': Quantification of Uncertainty for Estimation, Simulation and Optimization
- Introduction
- Stochastic Models and Algorithms
- The QUESO Design and Implementation
- Conclusions and Future Directions
- References
- Use of HPC-Techniques for Large-Scale Data Migration
- Introduction
- Popular Approaches toward Migrating Finance Data
- The Big Bang Approach
- The On-The-Fly Approach
- The Parallel Business Approach
- Introducing the QuickApply HPC-Software
- The Implementation of QuickApply: Combining Java, C & SQL
- Case Study: Time Needs For Migrating the Loyalty Program
- Conclusion
- References
- Algorithms, Models and Tools for Parallel Computing on Heterogeneous Platforms (HeteroPar 2011)
- Algorithms, Models and Tools for Parallel Computing on Heterogeneous Platforms (HeteroPar 2011)
- A Genetic Algorithm with Communication Costs to Schedule Workflows on a SOA-Grid
- Introduction
- Framework
- GA with Communication Costs
- Simulation Setup
- Results with General DAGs
- Results for a Set of Identical Intrees
- Conclusion and Future Work
- References
- An Extension of XcalableMP PGAS Lanaguage for Multi-node GPU Clusters
- Introduction
- Overview of XcalableMP
- Execution Model
- XcalableMP Directives
- Language Extension for Multi-node GPU
- Data Declaration
- Data Transfer
- Work Sharing
- Implementation of XcalableMP-ACC
- Performance Evaluation
- Related Work
- Conclusion
- References
- Performance Evaluation of List Based Scheduling on Heterogeneous Systems
- Introduction
- DAG Scheduling
- List Scheduling Algorithms
- HEFT Algorithm
- CPOP Algorithm
- Metaheuristic Algorithms
- Ant Colony System
- Simulated Annealing
- Tabu Search
- Results and Conclusions
- References
- Column-Based Matrix Partitioning for Parallel Matrix Multiplication on Heterogeneous Processors Based on Functional Performance Models
- Introduction
- Related Work
- Column-Based Partitioning (KL)
- Minimising Total Communication Volume (BR)
- Functional Performance Model-Based Partitioning (FPM-KL)
- Other Related Work
- New FPM-BR Two-Dimensional Matrix Partitioning Algorithm
- Experimental Results
- References
- A Framework for Distributing Agent-Based Simulations
- Introduction
- A Distributed Framework for Simulations
- MASON
- D-Mason: Distributed Mason
- Issues in D-Mason
- Architecture
- Testing
- Beyond Mason Limits
- Exploiting Heterogeneity
- Fixed Timing Analysis
- Conclusion and Future Work
- References
- Parallel Sparse Linear Solver GMRES for GPU Clusters with Compression of Exchanged Data
- Introduction
- GPU Architecture
- GMRES Implementation on GPU Clusters
- Parallel GMRES Algorithm for GPU Clusters
- Minimizing Communication Overheads
- Performance Evaluation
- Our GPU Cluster
- Sparse Matrices of Tests
- Experimental Results
- Conclusion
- References
- Two-Dimensional Discrete Wavelet Transform on Large Images for Hybrid Computing Architectures: GPU and CELL
- Introduction
- JPEG2000
- Cell B.E. Architecture
- GPGPU Architecture
- Related Work
- Discrete Wavelet Transform
- Sequential Algorithm
- Parallel Algorithm
- DWT on Cell B.E.
- Basic Optimizations
- Advanced Optimizations
- DWT on GPGPU
- Basic Optimizations
- Advanced Optimization
- Performance Analysis
- Cell B.E. - Scalability and Execution Time Analysis
- GPU - Scalability and Execution Time Analysis
- Conclusions
- References
- Scheduling Divisible Loads on Heterogeneous Desktop Systems with Limited Memory
- Introduction
- System Model and Problem Formulation
- Algorithm for Device-Level Multi-installment with Multi-distributions
- Iterative Procedure for Partial Performance Modeling
- Experimental Results
- Comparison with the State of the Art Approaches
- Conclusions
- References
- Peer Group and Fuzzy Metric to Remove Noise in Images Using Heterogeneous Computing
- Introduction
- Denoising Algorithm
- Multi-GPU and Multi-core CPU Implementations
- Experimental Study
- Concluding Remarks
- References
- Estimation of MPI Application Performance on Volunteer Environments
- Introduction
- Related Work
- Simulation Framework
- Experiments and Results
- Simulation Configuration
- Desktop over Internet Configuration Results
- Collection of Networked Computers Configuration Results
- Simulation on Real-Life Environments
- Conclusions
- References
- Author Index
