Algorithms and Architectures for Parallel Processing, Part I
11th International Conference, ICA3PP 2011, Melbourne, Australia,October 24-26, 2011, Proceedings, Part I
Erschienen am 23. Oktober 2011
XVIII, 497 Seiten
978-3-642-24650-0 (ISBN)
Beschreibung
This two volume set LNCS 7016 and LNCS 7017 constitutes the refereed proceedings of the 11th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2011, held in Melbourne, Australia, in October 2011. The first volume presents 24 revised regular papers and 17 revised short papers together with the abstract of the keynote lecture - all carefully reviewed and selected from 85 initial submissions. The papers cover the many dimensions of parallel algorithms and architectures, encompassing fundamental theoretical approaches, practical experimental results, and commercial components and systems and focus on two broad areas of parallel and distributed computing, i.e., architectures, algorithms and networks, and systems and applications.
Weitere Details
Weitere Ausgaben
Andere Ausgaben
Yang Xiang | Alfredo Cuzzocrea | Michael Hobbs
Algorithms and Architectures for Parallel Processing, Part I
11th International Conference, ICA3PP 2011, Melbourne, Australia,October 24-26, 2011, Proceedings, Part I
Buch
10/2011
Springer
53,49 €
Versand in 7-9 Tagen
Inhalt
- Title
- Table of Contents
- ICA3PP 2011 Keynote
- Keynote: Assertion Based Parallel Debugging
- ICA3PP 2011 Regular Papers
- Secure and Energy-Efficient Data Aggregation with Malicious Aggregator Identification in Wireless Sensor Networks
- Introduction
- Related Work
- System Model
- Secure and Energy-Efficient Data Aggregation with Malicious Aggregator Identification
- Aggregation Commitment
- Aggregation Verification
- Theoretical Analysis on Communication Overhead
- Discussion
- Simulation Evaluation
- Conclusions
- References
- Dynamic Data Race Detection for Correlated Variables
- Introduction
- Problem Description
- Related Work
- Race Detection for Correlated Variables
- Inferring Correlated Sets and Computational Units
- Adapting the Lockset Algorithm
- Calculating a Computational Unit's Lockset
- Happens-Before Analysis - Hybrid Race Detector
- Evaluation
- Detecting Extended Data Races
- Conclusion and Future Work
- References
- Improving the Parallel Schnorr-Euchner LLL Algorithm
- Introduction
- Preliminaries
- Parallel LLL Reduction
- Motivation and Approach
- Improved Parallel LLL Reduction
- Scalar Product Part
- Orthogonalization Part
- µ-Update and Size-Reduction Part
- Using Sequences of Reduction Parameters
- Experiments
- Setup
- Results
- Future Work
- References
- Distributed Mining of Constrained Frequent Sets from Uncertain Data
- Introduction
- Background
- Mining Frequent Sets from Uncertain Data
- Mining Frequent Sets That Satisfy User Constraints
- Our Proposed Distributed Mining System
- Finding Constrained Locally Frequent Sets
- Finding Constrained Globally Frequent Sets
- Experimental Results
- Conclusions
- References
- Set-to-Set Disjoint-Paths Routing in Recursive Dual-Net
- Introduction
- Recursive Dual-Net
- Set-to-Set Disjoint-Path Routing in RDN
- Concluding Remarks
- References
- Redflag: A Framework for Analysis of Kernel-Level Concurrency
- Introduction
- Design
- Instrumentation and Logging
- Lockset Algorithm
- Block-Based Algorithms
- Algorithm Enhancements
- Filtering False Positives and Benign Warnings
- Evaluation
- Related Work
- Conclusions
- References
- Exploiting Parallelism in the H.264 Deblocking Filter by Operation Reordering
- Introduction
- Background
- H.264 Deblocking Filter
- Related Work
- Algorithm
- Analysis of Data Dependencies
- Proposed Deblocking Order
- Analysis
- Architectural Requirements
- Conclusion and Future Work
- References
- Compiler Support for Concurrency Synchronization
- Introduction
- Related Work
- Background
- The Proposed Priority Scheduling Algorithm
- Algorithm
- Priority Assignment
- PS on TM Pathologies
- Experimental Results
- Setup
- Results
- Conclusion
- References
- Fault-Tolerant Routing Based on Approximate Directed Routable Probabilities for Hypercubes
- Introduction
- Related Works
- Preliminaries
- Directed Routable Probabilities
- Fault-Tolerant Routing Algorithms
- Naive Algorithm ADRP0
- Improved Algorithm ARDP1
- Performance Evaluation
- Time Complexity
- Computer Experiment
- Discussion
- Conclusion
- References
- Finding a Hamiltonian Cycle in a Hierarchical Dual-Net with Base Network of p -Ary q -Cube
- Introduction
- The Hierarchical Dual-Net
- Topological Properties of HDN
- Hamiltonian Cycle Embedding
- Concluding Remarks
- References
- Adaptive Resource Remapping through Live Migration of Virtual Machines
- Introduction
- Related Work
- Heterogeneity-Aware Schedulers
- Heterogeneity-Aware Applications
- Comparison with ARRIVE-F
- Framework
- Assumptions
- Performance Modeling
- Migration Prediction
- Migration Decisions
- Implementation
- Experimental Results
- Experimental Platform
- Applications
- Comparison of Predicted and Actual Execution Times
- Framework Overheads
- Compute Farm Throughput
- Experiment 1
- Further Experiments
- Conclusions and Future Work
- References
- LUTS: A Lightweight User-Level Transaction Scheduler
- Introduction
- Related Work
- Overview
- Scheduling Transactions
- LUTS-Based Heuristics
- CILUTS
- CTLUTS
- HASHLUTS
- Experimental Results
- Speedup
- Overhead
- Discussion
- Conclusion
- References
- Verification of Partitioning and Allocation Techniques on Teradata DBMS
- Introduction
- Related Work
- Background
- Validation on Teradata
- Teradata Description
- Experiments
- Implementation and Testing Joint and Sequential Approaches
- Conclusion
- References
- Memory Performance and SPEC OpenMP Scalability on Quad-Socket x86 64 Systems
- Introduction
- Related Work
- Test Systems
- Benchmarks
- Microbenchmarks
- SPEC OMPM2001
- Results
- Memory Latency
- Memory Bandwidth
- SPEC OMPM2001 Scaling with Multiple Cores
- SPEC OMPM2001 Scaling with Multiple Sockets or NUMA Nodes
- SPEC OMPM2001 Performance Comparison
- Conclusions
- References
- Anonymous Communication over Invisible Mix Rings
- Introduction
- Related Work
- Our Solution
- Structure
- Topology
- Anonymous Communications
- Attacks and Defense
- Anonymity Evaluation
- Performance Analysis
- Conclusion
- References
- Game-Based Distributed Resource Allocation in Horizontal Dynamic Cloud Federation Platform
- Introduction
- Mathematical Problem Formulation
- Resource Allocation Games in a HDCF Platform
- Non-cooperative Resource Allocation Game
- Cooperative Resource Allocation Game
- Simulation and Discussion
- Conclusions
- References
- Stream Management within the CloudMiner
- Introduction
- Challenges of Data Stream Management Applications
- Cloud-Enabled Stream Processing
- Contributions and Organization of the Paper
- StreamMiner Framework
- Architecture
- StreamMiner Services
- Technology and Application
- Cloud Services and Stream Transmission
- A Real-World Application: The StreamMiner Cloud
- Performance Experiments
- Experiment 1: Transmission Delays
- Experiment 2: Transmission and Processing
- Experiment 3: Limitation on Transmission Speed
- Related Work
- Conclusions and Future Work
- References
- Security Architecture for Virtual Machines
- Introduction
- Our Model
- Overview
- Operation at Source
- Operation at Destination
- Analysis
- Slammer Analysis
- Performance Analysis
- Related Work`
- Conclusion
- References
- Fast and Accurate Similarity Searching of Biopolymer Sequences with GPU and CUDA
- Introduction
- Theoretical Background
- Alignment Algorithm
- CUDA Programming Model and Architecture of Hardware Accelerator
- Related Works
- Implementation of Parallel Alignment on GPU and CUDA
- Calculation of Alignment Matrix
- Reducing the Number of Transactions
- Reducing Idle Time of Threads
- Data Arrangement
- Storing the Input Sequence and the Substitution Matrix
- Efficiency Tests
- Concluding Remarks
- References
- Read Invisibility, VirtualWorld Consistency and Probabilistic Permissiveness are Compatible
- Introduction
- Software Transactional Memory (STM) Systems
- Consistency Criteria for STM Systems
- Desirable Properties for STM Systems
- Content of the Paper
- STM Computation Model and Base Definitions
- Consistency Conditions: Opacity and Virtual World Consistency
- Invisible Reads, Opacity and Permissiveness are Incompatible
- Step 1: Ensuring Virtual World Consistency with Read Invisibility
- STM Interface, Incremental Reads and Deferred Updates
- The Underlying Data Structures
- The readT() and writeT() Operations
- The try_to_commitT() Operation
- Step 2: Adding Probabilistic Permissiveness to the Protocol
- Conclusion
- References
- Parallel Implementations of Gusfield's Cut Tree Algorithm
- Introduction
- Related Work
- Definitions
- Cut Tree Algorithms
- Gusfield's Algorithm - Sequential Version
- Parallelization of Cut Tree Algorithms
- MPI Version
- OpenMP Version
- Experimental Setup
- Experimental Results
- Conclusion
- References
- Efficient Parallel Implementations of Controlled Optimization of Traffic Phases
- Introduction
- Related Work
- Serial Intersection Algorithm
- Shared Memory Parallel Intersection Control Algorithm
- Message Passing Implementation
- Data Parallel Implementation
- Experimental Study
- Conclusion
- References
- Scheduling Concurrent Workflows in HPC Cloud through Exploiting Schedule Gaps
- Introduction
- Related Work
- PCH and Gap Search
- Distributed Gap Search
- Experiments and Discussions
- Conclusions
- References
- Efficient Decoding of QC-LDPC Codes Using GPUs
- Introduction
- Review of QC-LDPC Codes and the Belief Propagation Decoding Algorithm
- QC-LDPC Codes
- Belief Propagation Algorithm
- Parallel Computations Using GPUs
- GPU Architecture
- Implementation of a LDPC Decoder on GPUs
- Data Structure to Represent the Messages
- Decoding Procedures in GPU with the Use of Shared Memory
- Results and Discussions
- Simulation Results
- Conclusion
- References
- ICA3PP 2011 Short Papers
- A Combined Arithmetic Logic Unit and Memory Element for the Design of a Parallel Computer
- Introduction
- Theoretical Basis for Parallel Self-Timed Adder
- Design of ARAM Memory-cum-Logic Units
- Logical Operations
- Memory Access Operations
- Control and Floating Point Operations
- Instruction Cycles
- Implementation of ARAM
- Performance Evaluation
- Conclusion
- References
- Parallel Implementation of External Sort and Join Operations on a Multi-core Network-Optimized System on a Chip
- Introduction
- Algorithm Implementation
- External Sort Algorithm Implementation
- External Hash Join Algorithm Implementation
- DB2 Sort and Hash Join Accelerator
- Simulation Results
- Conclusion
- References
- STM with Transparent API Considered Harmful
- Introduction
- Benchmarks for STMs
- JWormBench: A Port of WormBench to Java
- Annotations to Avoid Over-Instrumentation
- Over-Instrumented Tasks
- New Java Annotations for the Deuce API
- Performance Evaluation
- Related Work
- Conclusions
- References
- A Global Snapshot Collection Algorithm with Concurrent Initiators with Non-FIFO Channel
- Introduction
- Related Works
- Assumption and System Model
- Concurrent Snapshot Collection Strategy
- Algorithm
- Data Structure
- Messages
- Algorithms
- Proof of Correctness
- Complexity Analysis
- Conclusion
- References
- An Approach for Code Compression in Run Time for Embedded Systems - A Preliminary Results
- Introduction
- Architectures for Code Compression
- Code Compression in Run Time
- Algorithm for Compression/Decompression with MIC Method
- Simulations with Benchmark MiBench
- Related Work
- Conclusions and Future Work
- References
- Optimized Two Party Privacy Preserving Association Rule Mining Using Fully Homomorphic Encryption
- Introduction
- Privacy Preserving Association Rule Mining
- Motivation and Problem Formulation
- Background
- Notations
- Some Binary Operations
- Fully Homomorphic Encryption (FHE)
- Proposed Solution
- ARM with Privacy Preservation
- Performance and Security
- Performance Analysis
- Security Analysis
- Conclusion
- References
- SLA-Based Resource Provisioning for Heterogeneous Workloads in a Virtualized Cloud Datacenter
- Introduction
- Related Work
- System Model
- Datacenter Model
- SLA and Application Models
- Admission Control and Scheduling Policy
- Forecasting Model
- Admission Control and Scheduling
- SLA Enforcement and Rescheduling of VMs
- Performance Evaluation
- Workload Data
- Performance Metrics
- Analysis of Results
- Conclusions and Future Directions
- References
- SC: A Programming Model and Language for Embedded Manycores
- Introduction
- Related Work
- The SC Programming Model
- Components
- Behavior
- Designing for Execution Guarantees
- The SC Programming Language
- Components
- System Agents
- Input / Output
- Software Architecture
- A Sketch of the SC Compilation Process
- Evaluation
- Conclusion
- References
- Provisioning Spot Market Cloud Resources to Create Cost-Effective Virtual Clusters
- Introduction
- Related Work
- Cloud-Based Virtual Clusters
- Use of Variable Pricing Resources
- System Model of a Cloud-Based Virtual Cluster
- Cost-Effective Resource Provisioning and Scheduling Policy
- Estimating Job Runtimes
- Estimation Correction and Rescheduling
- Job Moldability and Speedup Considerations
- Performance Evaluation
- Comparison with Best-Case and Worst-Case Scenarios
- Effect of Different Runtime Estimation Methods
- Conclusions and Future Work
- References
- A Principled Approach to Grid Middleware
- Introduction
- Background and Motivation
- Grid Vision and Grid Practice
- The Cloud Vision - A Better Grid?
- Overview of the Minimum Intrusion Grid
- System Architecture
- MiG Design Principles and Rationale
- Browser-Based Interface
- How MiG Executes Grid Jobs
- Software Deployment
- MiG Features beyond the ``Job-Shop''
- VGrids: Virtual Organisations in MiG
- Resource Management
- Storage in MiG
- Advanced VGrid Features for Group Collaboration
- Current Status and Future Directions
- Conclusions
- References
- Performance Analysis of Preemption-Aware Scheduling in Multi-cluster Grid Environments
- Introduction
- Analytical Queuing Model
- Preemption-Aware Scheduling Policy
- Performance Evaluation
- Experimental Setup
- Experimental Results
- Related Work
- Conclusions and Future Work
- References
- Performance Evaluation of Open Source Seismic Data Processing Packages
- Introduction
- Related Work
- Sequence of Seismic Functions
- Performance Evaluation
- SU and Madagascar Seismic Data Format
- Seismic Data Conversion
- Tests and Results
- Conclusion
- References
- Reputation-Based Resource Allocation in Market-Oriented Distributed Systems
- Introduction
- Related Work
- Models
- System Model
- Application Model
- Market Model
- Reputation-Based Resource Allocation
- Formation of Resource Reputation
- The Suitability between Resource and Task
- Incorporation of Data Staging
- Experimental Methodology
- Experimental Settings
- Performance Metrics
- Experimental Results
- Impact of Scheduling with Variability of Network Capacity
- Comparing Market-Based Resource Allocation Approaches
- Conclusion
- References
- Cooperation-Based Trust Model and Its Application in Network Security Management
- Introduction
- Related Work
- User Cooperation Trust Model (UCTM)
- Design Principle
- The Reputation Model
- Simulation Detail
- Simulation Results
- Conclusion
- References
- Performance Evaluation of the Three-Dimensional Finite-Difference Time-Domain(FDTD) Method on Fermi Architecture GPUs
- Introduction
- Background
- A Brief Overview of the FDTD Method
- Parallelization of FDTD Method on Fermi Architecture GPUs
- Method
- Simulation Model
- Implementation in CUDA
- Performance Analysis and Comparison between Using Shared Memory and L1 Cache
- Conclusions and Future Work
- References
- The Probability Model of Peer-to-Peer Botnet Propagation
- Introduction
- Related Work
- Theoretical Propagation Model
- Propagation Ability and Quarantine Ability
- Simulation Experiments
- Conclusion and Future Work
- References
- A Parallelism Extended Approach for the Enumeration of Orthogonal Arrays
- Introduction
- Major Contribution
- Background
- Serial MCS Algorithm
- Two Key Properties of Serial MCS Algorithm
- A Step by Step Extending Parallelism Approach for EOA
- Parallel Computing for Each Small Extending Step
- Drawback of Previous Work with Master-Slave Load Balancing Method
- Convert MCS Algorithm by Using Stack Data Structure
- Work Sharing Method for Dynamic Load Balancing
- MPI Implementation
- Experimental Evaluation
- Experiments for Small Extending Step
- Dynamic Load Balancing
- Impact of Duration Time d and Chunk Size c
- Conclusion and Future Work
- References
- Author Index
