Network and Parallel Computing

Name: Network and Parallel Computing | 8th IFIP International Conference, NPC 2011, Changsha, China, October 21-23, 2011, Proceedings
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

8th IFIP International Conference, NPC 2011, Changsha, China, October 21-23, 2011, Proceedings

Erik Altman Weisong Shi(Herausgeber*in)

Springer (Verlag)

Erschienen am 18. Oktober 2011

XIII, 368 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-642-24403-2 (ISBN)

53,49 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Title page
Program Chairs' Welcome
Organization
Table of Contents
Session 1: Filesystems and Data
Elastic Phoenix: Malleable MapReduce for Shared-Memory Systems
Introduction
Background
Related Work
Design Goals
Implementation
Sharing Data
Assigning Tasks
Coordinating Workers
Porting Phoenix Applications
Evaluation
Overhead
Elasticity Advantages
Limitations and Future Work
Concluding Remarks
References
A Load-Aware Data Placement Policy on Cluster File System
Introduction
Related Work
Background
Load-Aware Data Placement and Load Balance Mechanism
Storage Server Load
Loads-Aware Data Placement
Load Balance Mechanism
Data Migration
Performance Evaluation
Load Balance Interval Constant
Throughput
mpiBLAST Application
Conclusion
References
H-Fuzzing: A New Heuristic Method for Fuzzing Data Generation
Introduction
Related Work
Program Model
Heuristic Program Fuzzing Data Generation Method
The Program Static Analysis Procedure
The Fuzzing Data Generation Algorithm
Experimental Evaluation
Conclusion and Discussion
References
Improved Compact Routing Schemes for Power-Law Networks
Introduction
Related Work
Definitions and Notations
Our Routing Schemes
HDLR
HDLR+
Analytical Performance Analysis
Theory of Random Power-Law Graphs
Analysis
Simulations
Diameter of the Subgraph Induced by Landmarks
Average Routing Table Size
Average Stretch
Conclusions
References
Session 2: Network and Parallel Algorithms
Accelerating the Requirement Space Exploration through Coarse-Grained Parallel Execution
Introduction
Monotonic Indices Space (MIS) Method
MIS Method
Time and Memory Costs of the MIS Method
Coarse-Grained Approach
Parallelized Resolving in C/S Mode
Task Pool Work Balancing
Speedup Analysis
Experimental Results
Conclusions and Future Work
References
A Global Benefit Maximization Task-Bundle Allocation
Introduction
Related Work
Problem Formulation
System Model
Pricing
Design and Analysis of ISGBMAT Algorithm
Calculation Method for ISCT
Solution of Task Assignment Amount Vector
ISGBMAT Algorithm
Performance Evaluation
Experimental Setup
Experimental Results and Analysis
Conclusion
References
Service Quality Assurance Mechanisms for P2P SIP VoIP
Introduction
Related Work
Skype Relay Calls
Application-Layer Relay Algorithm
Design and Architecture
Design Principle
System Architecture Overview
RN-S Register
Redundancy Detecting
NAT Traversal
Dynamic Relay Algorithm
Performance Evaluations
Evaluation Environment
Re-Hit Ratio
Dialog Completion Ratio
Latency of Switching Relay Nodes
Conclusions
References
Session 3: Cluster and Grid
VSCM: A Virtual Server Consolidation Manager for Cluster
Introduction
Problem Statements
The Virtual Machine Bin Packing Problem
The Virtual Machine Migration Problem
Design and Implementation of VSCM
The Globally Optimal Virtual Machine Bin Packing Algorithm
Feedback Based Virtual Machine Migration Algorithm
Evaluation
Design of Migration Overhead
Experimental Setup
Results and Analysis
Related Work
Conclusions
References
Informed Live Migration Strategies of Virtual Machines for Cluster Load Balancing
Introduction
Background and Motivation
Experimental Evaluation and Analysis
The Migrating VMs Selection Problem
The Destination Host Choose Problem
The Migration Sequence
Informed Live Migration Strategies
Related Work
Conclusions and Future Work
References
Dynamic Advance Reservation for Grid System Using Resource Pools
Introduction
Related Work
SLAs for QoS Negotiation
Grid Architecture Using Resource Pool for Advance Reservation
Grid System Architecture
Theoretical Analysis
kAPAC Algorithm
QoS Distance Computation
Algorithm Design
Experimental Results
Impact of k-set Length and Recent Queue Length
kAPAC vs. SNAP Strict Negotiation
Conclusion
References
Session 4: Trust and Authentication
A Way of Key Management in Cloud Storage Based on Trusted Computing
Introduction
Background and Motivation
Key Management and Cloud
RNG
Key Structure
Migratable Keys versus Non-migratable Keys
Motivation
Architecture of a Private Data Protecting Service
Create Key Structure of Cloud
Data Processing Service
The Backup Service
The Key Sharing Service
Constructing the Prototype and the Performance Evaluation
Conclusions and Future Work
References
A Trust-Based Defensive System Model for Cloud Computing
Introduction
Architecture of Defensive Model
Physical Structure
Logical Structure
Relevant Description
Behavioral Evidences
Obtaining Evidences
Standardizing Evidences
Evaluation Model Based on FAHP
A Hierarchical Structure of Evidences
The Weight Determined by FAHP
The Assessment of User Behavior Character
The Overall Trust Evaluation of User Behavior
Evaluation Model for Suspicious Files
Multi-engine Detection Mechanism
Dynamic Behavior Analysis of Suspicious Files
Simulation and Analysis
Experimental Simulation and Description
Example of Trust Evaluation
Experimental Comparison
Performance Comparison
Conclusion
References
A New Method for Authentication Based on Covert Channel
Introduction
Theory
Covert Channel
Categories of Covert Channel
Comparison between Storage and Timing Covert Channel
Methodology
Result
Conclusion and Future Work
References
Session 5: Monitor, Diagnose, and Then Optimize
A Layered Detection Method for Malware Identification
Introduction
Machine Learning Based Approach
Feature Extraction and Definition
Performance Measures
Feature Selection
Classification Algorithm
Layered Framework and Method for Detecting Malwares
Experiment and Discussion
Feature Selection
Malware Classification
Conclusions
References
Pedestrian Detection and Tracking Using HOG and Oriented-LBP Features
Introduction
Previous Work
Feature Pool and Classifier
Feature Pool
Classifier
Pedestrian Tracking
Detection Validation
Pedestrian Tracking
Experimental Results
Performance of Different Features with SVM Classifier
Comparison of SVM and Adaboost Classifier
Detection Results with HOG-Oriented LBP Feature
Conclusions
References
An Efficient Checkpointing Scheme Using Price History of Spot Instances in Cloud Computing Environment
Introduction
Related Work
System Architecture
Layer Structure
Instances Types
The SLA Based Checkpointing Scheme
SLA Based on Price History Using Spot Instances
Fault Tolerance Mechanisms Using Checkpoints
Performance Evaluation
Simulation Environments
The Analysis of Computing-Type Instances
The Analysis of Memory-Type Instances
Conclusion
References
Session 6: Best Paper - 1
Towards a Universal Sketch for Origin-Destination Network Measurements
Introduction
Problem Statement
Conditional Random Sampling
The CRS Algorithm
Implementation Details
Unbiasedness of CRS
Experiments
Related Work
Conclusions
References
A Power Adjusting Algorithm on Mobility Control in Mobile Ad Hoc Networks
Introduction
Propagation Model, Mobility Model, and Network Assumptions
Propagation Model
Mobility Model
Network Assumptions
The Power Adjusting Algorithm Design
Description of PAA
Analysis of PAA
Parameters Setting of PAA
The Average Distance of Any Two MHs
Energy Comsumption on Route Dicovery
The Distance Variety between Adjacent MHs on the Routing Path
Finding the Optimal Length of the Period
Simulation Results
Conclusions
References
Power-Aware Run-Time Incremental Mapping for 3-D Networks-on-Chip
Introduction
Related Work
Problem Formulation
Communication Power Model
Application and Architecture Model
Problem Description
Motivating Example and Problem Decomposition
Motivating Example
A Mapping Flow by Problem Decomposition
Algorithm Description
NoC Region Selection (P1)
Set Matching (P2)
CTG to NoC Region Mapping (P3)
Performance Evaluation
Mapping Single Application to 3-D NoC
Mapping Multiple Applications Incrementally to 3-D NoC
Conclusion
References
Session 7: Best Paper - 2
WeLe-RAID: A SSD-Based RAID for System Endurance and Performance
Introduction
Problem Description
Why Need Wear-Leveling
Why Not RAID5
Why Not other Schemes
WeLe-RAID
The Architecture of WeLe-RAID
Data Layout
Addressing Method
Evaluation
Performance of WeLe-RAID
Endurance of WeLe-RAID
Reliability of WeLe-RAID
Related Work
Conclusion and Future Work
References
Reorder the Write Sequence by Virtual Write Buffer to Extend SSD's Lifespan
Introduction
Backgrounds and Related Works
SSD Overview
Write Amplification Analysis
Related Works
The Virtual Write Buffer Architecture
The DRAM Component
Reorder Area
Performance Evaluation
Experimental Setup
Comparison with Other Buffers
Comparison with FAST
Conclusion
References
Memory-Side Acceleration for XML Parsing
Introduction
Background
XML Parsing Techniques
Related Research Work
Methodology
XML Parsers and Benchmarks
Prefetchers
Performance and Memory Modeling
Native vs. Managed Execution
Performance Analysis of XML Parsing
Network Data Exchange
Disk Data Loading
Data Loading from Memory Side
Summary
Memory-Side Acceleration
Implementation Feasibility
Bandwidth Utilization
Hardware Cost and Energy Consumption
Conclusions
References
Session 8: Microarchitecture
Hardware Performance Monitoring for the Rest of Us A Position and Survey
Introduction
Background
Event-Based Sampling (EBS)
Instruction Based Sampling (IBS)
Event Address Registers (EAR)
Last Branch Record (LBR)
Lightweight Profiling (LWP)
Issues Facing HPM
Opportunities and Alternatives
A Pragmatic Proposition
Feature Requests
An Adventurous Approach
Conclusion
References
An Efficient Architectural Design of Hardware Interface for Heterogeneous Multi-core System
Introduction
Related Work
Construct of MPU
Application Model
MPU Structure
Object Scheduling and Synchronization Flow
Experimental Results
Evaluation Methodology
The Fast Fourier Transform (FFT)
Eigenvalue of Matrix
Physical Parameters of MPU
Conclusion
References
Session 9: Network and Mobile Computing
Evaluations on Effectiveness of Bump-Slot over DAP-NAD-Based Tactical Wideband Wireless Networks
Introduction
MIL-STD-188-220D with Change 1
Network Access Control (NAC)
DAP-NAD
Motivations
Simulations and Evaluations
Simulation Environments
Performance Evaluations
Conclusions
References
Access Path Based Source Address Validation in Mobile IPv6
Introduction
Problem Statement
Access Path Based Trust Communication Architecture in Mobile IPv6
Overview
Information Collecting and Recording
Access Path Construction
Access Path Based Authentic Source Address Validation
Performance and Security Analysis
Storage overhead
Security Analysis
Feasibility Analysis
Experimental Study
Effectiveness Evaluation Algorithm
Choice of Access Hops
Related Work
Conclusion
References
Improve Google Android User Experience with Regional Garbage Collection
Introduction
Related Work
An Overview of the Regional GC
Regional GC Design Details
The Heap Layout
The Minor Collection
The Major Collection
Experimental Results
Overall Performance Impact
Collection Pause Time Reduction
Write Barrier Overhead
Conclusion
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken