Euro-Par 2011: Parallel Processing Workshops

Name: 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 II
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

CCPI, CGWS, HeteroPar, HiBB, HPCVirt, HPPC, HPSS, MDGS, ProPer, Resilience, UCHPC, VHPC, Bordeaux, France, August 29 -- September 2, 2011, Revised Selected Papers, Part II

Michael Alexander Pasqua D'Ambra Adam Belloum George Bosilca Mario Cannataro Marco Danelutto Beniamino Di Martino Michael Gerndt Emmanuel Jeannot Raymond Namyst Jean Roman Stephen L. Scott Jesper Larsson Traff Geoffroy Vallee Josef Weidendorfer(Editor)

Springer (Publisher)

Published on 14. April 2012

XXX, 480 pages

E-Book

PDF with digital watermarking

System requirements

978-3-642-29740-3 (ISBN)

€53.49incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Title
Preface
Organization
Table of Contents - Part II
HiBB 2011: 2nd Workshop on High-Performance Bioinformatics and Biomedicine
HiBB 2011: 2nd Workshop on High Performance Bioinformatics and Biomedicine
On Parallelizing On-Line Statistics for Stochastic Biological Simulations
Introduction
The CWC Formalism and Its Parallel Simulator
On Line Statistical Tools
Typical Patterns for Biological Trajectories
Examples
Related Work
Conclusions
References
Scalable Sequence Similarity Search and Join in Main Memory on Multi-cores
Introduction
Preliminaries
Data Structure and Algorithms
Algorithms
Parallelization with In-memory MapReduce
Evaluation
Performance of Similarity-Based Operators
Scalability of PeARL
Related Work
Conclusions and Future Work
References
Enabling Data and Compute Intensive Workflows in Bioinformatics
Introduction
Motivation
Workflow Technology
Workflows in Bioinformatics
Proteomics: MassMatrix
RNA Sequencing: Transcriptional Atlas of the Developing Human Brain
RNA Sequencing: Cancer Genome Atlas Using SeqWare
Quality Assurance and Quality Control: Population Architecture Using Genomics and Epidemiology (PAGE)
Workflows in a Virtual Machine
Related Works
Future Works and Conclusion
References
Homogenizing Access to Highly Time-Consuming Biomedical Applications through a Web-Based Interface
Introduction
BioBench: A Framework to Integrate and to Transform Text-Based User Interfaces into Homogeneous Web-Based Ones
Description
Design
Software Development
BiosH: A Case Study
Conclusions
References
Distributed Management and Analysis of Omics Data
Introduction
Management of Omics Data
Genomics Databases
Proteomics Databases
Interactomics Databases
Omics Data Analysis
Microarray Data Analysis
Mass Spectrometry Data Analysis
Protein-to-Protein Interaction Data Analysis
Tools for Distributed Management of Omics Data
Micro-CS
MS-Analyzer
IMPRECO
OntoPIN
Conclusion and Future Work
References
Managing and Delivering Grid Services (MDGS)
Managing and Delivering Grid Services (MDGS)
Resource Allocation for the French National Grid Initiative
Context and Definitions
Context of Current Work
Definition of "User Communities"
Definition of "Resource Providers" and "Service Providers"
Definition of "Resource Allocation"
Identified Needs and Goals
Improve Service Delivery to the French Community
Measure What Is Done
Inspiration from Existing Resource Allocation Mechanisms
High Performance Computing (HPC) World
Worldwide LHC Computing Grid (WLCG)
Other National Grids
Definition of the Strategy
Key Principles
Allocating Resources to New Users and French Communities
Allocating Resources to Established International Communities
Next Steps
References
On Importance of Service Level Management in Grids
Introduction
Background: On the Relevance of SLM
Model of SLM for Grids
Elements of SLM in Grid Theory
Actors Perspective
Customer Perspective
Sites Perspective
Grid Infrastructure Provider Perspective
Application Scenarios That Require SLM
Large Collaboration Case
Data Challenges
Urgent Computing
Related Works
Examples of Implementation of SLM Elements in Grids
Summary
References
On-Line Monitoring of Service-Level Agreements in the Grid
Introduction and Motivation
Related Work
On-Line SLA Monitoring Framework
Architecture
Design and Implementation
Defining Composite SLA Metrics Using EPL
Registry
SLA Monitoring for Data-Intensive Computations
Conclusion
References
Challenges of Future e-Infrastructure Governance
Introduction
e-IRG Recommendations
e-Governance Management
Future of Research Networks
Authentication, Authorization and Accounting
Energy and Green IT
Exascale Computing and Related Software
e-Infrastructure Services
Data Infrastructures
Service-Orientation of the Future Open e-Infrastructures
Conclusions
References
Influences between Performance Based Scheduling and Service Level Agreements
Introduction
An SLA Aware Job Allocation Architecture
Benchmarks Characterizing Resource Performance
Evaluation of a Performance-Based Job Allocation
Characterizing the Test Bed
Simulating the Architecture
Conclusion
References
User Centric Service Level Management in mOSAIC Applications
Introduction
Related Work
mOSAIC API
Vision and Approach
General Overview of the SLA Architecture
An Example of mOSAIC Approach
Conclusions
References
Service Level Management for Executable Papers
Introduction
Executable Papers Lifecycle
Exploitation of Executable Papers
Challenges Facing the Implementation of Executable Papers
Discussions
Conclusion
References
Change Management in e-Infrastructures to Support Service Level Agreements
IT Service Management and e-Infrastructures
Case Study: e-Infrastructure for High Performance Computing
Related Work
Design of an Inter-organizational Change-Management
Summary
References
PROPER 2011: Fourth Workshop on Productivity and Performance: Tools for HPC Application Development
PROPER 2011: Fourth Workshop on Productivity and Performance Tools for HPC Application Development
Scout: A Source-to-Source Transformator for SIMD-Optimizations
Introduction
Related Tools
The Vectorizing Source-to-Source Transformator Scout
Unroll-and-Jam
Configuring Scout
Partial Vectorization
Practical Results
Summary and Conclusion
Future Work
References
Scalable Automatic Performance Analysis on IBM BlueGene/P Systems
Introduction
BlueGene/P Architecture
Periscope Design
Tool Architecture
Current Design Shortcomings
BlueGene/P Tailored Design Alternatives
MPMD Approach
Implementation Based on the HTC Mode
I/O Nodes Agents Placement
Comparison of Design Alternatives
Evaluation
Related Works
Conclusion and Outlook
References
An Approach to Creating Performance Visualizations in a Parallel Profile Analysis Tool
Introduction
Design Approach
Examples
Sweep3D
S3D Use Case
GCRM/ZGrd Use Case: 3D Correlation Plot
Related Work
Conclusion
References
INAM - A Scalable InfiniBand Network Analysis and Monitoring Tool
Introduction
Background
InfiniBand
OFED
Design and Implementation of INAM
Features of INAM
Experimental Results
Experimental Setup
Visualizing Port Counters
Point to Point Visualization: Synthetic Communication Pattern
Link Utilization of Collective Operations: Case Study with MPI_Bcast Operation
Application Visualization: SpecMPI - LU
Overhead of Running INAM
Related Tools
Conclusions and Future Work
References
Auto-tuning for Energy Usage in Scientific Applications
Introduction
Motivation
Experiments
Power/Energy Measurement
Auto-tuning Feedback Metric
Experimental Platform
Results for Poisson's Equation Solver (PES)
Future Work
Related Work
Conclusion
References
Automatic Source Code Transformation for GPUs Based on Program Comprehension
Introduction
Related Works
Source Code Transformation
Algorithm Recognition
Source to Source Transformation
Prototype Tool
Case Study
Conclusion
References
Enhancing Brainware Productivity through a Performance Tuning Workflow
Introduction
Brainware for Green HPC
A Performance Tuning Workflow
Initial Meeting
Base-Line
Pathology
Tuning Cycle
Results/Handover
Conclusion
References
Workshop on Resiliency in High-Performance Computing (Resilience) in Clusters, Clouds, and Grids
Workshop on Resiliency in High Performance Computing (Resilience) in Clusters, Clouds, and Grids
The Malthusian Catastrophe Is Upon Us! Are the Largest HPC Machines Ever Up?
Introduction
Defining a Resilient System
Difficulties in Obtaining and Comparing Data
Reliability of Kraken at the National Institute for Computational Sciences
Cray XT Architecture Description
NICS Kraken System Description
Data Collection
Kraken Node Failure Data
Comparison of Large Cray Systems Resiliency
Data Collection
The Machines
Future Work
Conclusions
References
Simulating Application Resilience at Exascale
Introduction
Requirements
Future Systems
Simulation and Modeling
Design of Our Simulation Infrastructure
The Structural Simulation Toolkit (SST)
Router Model
Storage Component
Communication Pattern Generators and Applications
Implementing Pattern Generators
Tying It All Together
Work in Progress
References
Framework for Enabling System Understanding
Introduction
Approaches
Numeric Information Collection and Storage
Text Based Information Collection and Storage
Numeric Analysis
Text Based Analysis
Applications
Related Work
Conclusions
References
Cooperative Application/OS DRAM Fault Recovery
Introduction
DRAM Failures
Fault and Recovery Model
Fault Model
Recovery Model
Application / OS Interface
Design
Implementation
Testing Support
Fault-Tolerant GMRES
Initial Results
Related Work
DRAM Error Handling
Fault Tolerant Algorithms
Conclusions and Future Work
References
A Tunable, Software-Based DRAM Error Detection and Correction Library for HPC
Introduction
Design
Extensions for Error Correction
Assumptions and Limitations
Implementation
Handling User Pointers with System Calls
Results
Related Work
Conclusions and Future Work
References
Reducing the Impact of Soft Errors on Fabric-Based Collective Communications
Introduction
Fabric-Based Collectives
Handling Soft Errors on InfiniBand
Collective Replication Approaches
Spatial Replication
Temporal Replication
Experiments
Results
Related Work
Conclusions
References
Evaluating Application Vulnerability to Soft Errors in Multi-level Cache Hierarchy
Introduction
Graphite Simulator and Bit Error Injection
Fast Processor Simulation
Cache Error Injection
Multiple Cell Upset and Bit Error Patterns
Simulation Setup
Applications
Simulation Parameters
Simulation Results
Application Response to Fault Injections
Conclusions
References
Experimental Framework for Injecting Logic Errors in a Virtual Machine to Profile Applications for Soft Error Resilience
Introduction
Overview of Methodology
Startup
Probe
Fault Injection
Experiments
Floating Point Multiply Fault Injection
Floating Point Addition Fault Injection
Intended Uses
Related Work
Future Work
Conclusion
References
High Availability on Cloud with HA-OSCAR
Introduction
Related Work
HA-OSCAR 2.0
HA-OSCAR Software Architecture
System Architecture
OSCAR-V Cluster Architecture
HA-OSCAR-V Cluster Architecture
Case Study: HA-OSCAR-Enabled Nimbus
System Model
OSCAR-V Cluster System Model
HA-OSCAR-V Cluster System Model
Availability Analysis
OSCAR-V Cluster System Analysis
HA-OSCAR-V Cluster System Analysis
HA Comparison
Conclusions and Future Work
References
On the Viability of Checkpoint Compression for Extreme Scale Fault Tolerance
Introduction
Checkpoint Compression
A Checkpoint Compression Viability Model
Previous Work
Evaluating Checkpoint Compression
The Applications
The Checkpoint Library
The Compression Algorithms
The Tests
Compression Results
Discussion
Current Limitations
References
Can Checkpoint/Restart Mechanisms Benefit from Hierarchical Data Staging?
Introduction
Background
Detailed Design
Experimental Evaluation
Experimental Testbed
Profiling of a Stand-Alone Staging Server
Scalability Analysis
Evaluation with Applications
Related Work
Conclusion and Future Work
References
Impact of Over-Decomposition on Coordinated Checkpoint/Rollback Protocol
Introduction
Background
Over-Decomposition for Rollback
Over-Decomposition for Global Restart
Over-Decomposition for Partial Restart
Experimental Results
Global Restart
Partial Restart
Related Works
Conclusion and Future Work
References
UCHPC 2011: Fourth Workshop on UnConventional High-Performance Computing
UCHPC 2011: Fourth Workshop on UnConventional High Performance Computing
PACUE: Processor Allocator Considering User Experience
Introduction
Designing PACUE
Dynamic Compute Device Redirection
System Resource Management
Evaluation
Evaluation Policy
Results
Analysis
Conclusions and Future Work
References
Workload Balancing on Heterogeneous Systems: A Case Study of Sparse Grid Interpolation
Introduction
Related Work
Optimizing Programs for Heterogeneous Computing
Sparse Grid Interpolation
Interpolation and Heterogeneous Computing
Dynamic Task Based Load Balancing
Static Load Balancing
Evaluation
Conclusion
References
Performance Evaluation of a Multi-GPU Enabled Finite Element Method for Computational Electromagnetics
Introduction
The Physical Problem and Its Numerical Treatment
Implementation Aspects
DGTD CUDA Kernels
Multi-GPU Strategy
Performance Results
Weak Scalability
Strong Scalability
Conclusion
References
Study of Hierarchical N-Body Methods for Network-on-Chip Architectures
Introduction
Modeling of the Network-on-Chip
The Hierarchical N-Body Methods
Experimental Evaluation
Experiment Setup
Result Analysis
Conclusion
References
Extending a Highly Parallel Data Mining Algorithm to the Intel R Many Integrated Core Architecture
Introduction
Intel MIC Architecture in Comparison to NVIDIA Fermi Architecture
Data Mining Using Sparse Grids
Implementation and Performance Measurements
Conclusion
References
VHPC 2011: 6th Workshop on Virtualization in High-Performance Cloud Computing
VHPC 2011: 6th Workshop on Virtualization in High-Performance Cloud Computing
Group-Based Memory Deduplication for Virtualized Clouds
Introduction
Background and Motivation
Memory Deduplication
Performance Isolation in Clouds
Limitations of Memory Deduplication in Clouds
Group-Based Memory Deduplication
Design
Implementation
Evaluation
Experimental Environments
Effects of Group-Based Memory Deduplication
Discussion
VM Colocation
Grouping Policies
Per-group Deduplication Customization
Conclusions and Future Work
References
A Smart HPC Interconnect for Clusters of Virtual Machines
Introduction
Background and Related Work
Design and Implementation
Performance Evaluation
Results
Conclusions
References
Coexisting Scheduling Policies Boosting I/O Virtual Machines
Introduction
Background
Motivation
Related Work
Default Setup
Co-existing Scheduling Polices
Monitoring Tool
The Driver Domain Pool
Decoupling vCPUs Based on Workload Characteristics
Discussion
Credit Vulnerabilities to I/O Service
Exotic Scenarios
Dynamic Instead of Static
Conclusions
References
PIGA-Virt: An Advanced Distributed MAC Protection of Virtual Systems
Introduction
Motivation
Architecture of PIGA-Virt
In-Depth End-to-End Mandatory Protection
Experimentations
Related Works
Conclusion
References
An Economic Approach for Application QoS Management in Clouds
Introduction
Architecture
Use Cases
Adapting the Agent's Spending Rate
Adapting the Virtual Cluster Size
Related Work
Conclusions
References
Evaluation of the HPC Challenge Benchmarks in Virtualized Environments
Introduction
Related Work
Setup
Self-monitoring Results
Statistical Methods
Hardware and Software Used in Tests
Disparity between CPU and Wall Clock Timers
Accuracy-Drift of Timers with VM State Accumulation
Results
Conclusions and Future Work
References
DISCOVERY, Beyond the Clouds
Introduction
Related Work
The DISCOVERY Proposal
Architecture Overview
The DISCOVERY Agent
DISCOVERY in a Nutshell
Conclusion
References
Cooperative Dynamic Scheduling of Virtual Machines in Distributed Systems
Introduction
Related Work
Proposal Overview
Main Characteristics
The Iterative Scheduling Procedure
Implementation
Current State
Node Agent
Experiments
Future Work
Conclusion
References
Large-Scale DNA Sequence Analysis in the Cloud: A Stream-Based Approach
Introduction
Large-Scale DNA Sequence Analysis
A Stream-Based Approach
Incremental Data Processing with an SPE
Use Case Implementation
Initial Evaluation
Conclusions and Future Work
References
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Euro-Par 2011: Parallel Processing Workshops

Description

More details

Other editions

Additional editions

Content

System requirements