Interconnection Networks


Jose. Duato
Bok Engelsk 2002 · Electronic books.
Utgitt
Burlington : : Elsevier Science, , 2002.
Omfang
1 online resource (625 p.)
Utgave
Rev. printing
Opplysninger
Description based upon print version of record.. - Front Cover; Interconnection Networks: An Engineering Approach; Copyright Page; Contents; Foreword; Foreword to the First Printing; Preface; Chapter 1. Introduction; 1.1 Parallel Computing and Networks; 1.2 Parallel Computer Architectures; 1.3 Network Design Considerations; 1.4 Classification of Interconnection Networks; 1.5 Shared-Medium Networks; 1.6 Direct Networks; 1.7 Indirect Networks; 1.8 Hybrid Networks; 1.9 A Unified View of Direct and Indirect Networks; Chapter 2. Message Switching Layer; 2.1 Network and Router Model; 2.2 Basic Concepts; 2.3 Basic Switching Techniques. - 2.4 Virtual Channels2.5 Hybrid Switching Techniques; 2.6 Optimizing Switching Techniques; 2.7 A Comparison of Switching Techniques; 2.8 Engineering Issues; 2.9 Commented References; Exercises; Problems; Chapter 3. Deadlock, Livelock, and Starvation; 3.1 A Theory of Deadlock Avoidance; 3.2 Extensions; 3.3 Alternative Approaches; 3.4 Deadlock Avoidance in Switch-Based Networks; 3.5 Deadlock Prevention in Circuit Switching and PCS; 3.6 Deadlock Recovery; 3.7 Livelock Avoidance; 3.8 Engineering Issues; 3.9 Commented References; Exercises; Problems; Chapter 4. Routing Algorithms. - 4.1 Taxonomy of Routing Algorithms4.2 Deterministic Routing Algorithms; 4.3 Partially Adaptive Algorithms; 4.4 Fully Adaptive Algorithms; 4.5 Maximally Adaptive Routing Algorithms; 4.6 Nonminimal Routing Algorithms; 4.7 Backtracking Protocols; 4.8 Routing in MINs; 4.9 Routing in Switch-Based Networks with Irregular Topologies; 4.10 Resource Allocation Policies; 4.11 Engineering Issues; 4.12 Commented References; Exercises; Problems; Chapter 5. Collective Communication Support; 5.1 Collective Communication Services; 5.2 System Support for Collective Communication. - 5.3 Preliminary Considerations5.4 Models for Multicast Communication; 5.5 Hardware Implementations of Multicast; 5.6 Hardware Support for Barrier Synchronization and Reduction; 5.7 Software Implementations of Multicast; 5.8 Engineering Issues; 5.9 Commented References; Exercises; Problems; Chapter 6. Fault-Tolerant Routing; 6.1 Fault-Induced Deadlock and Livelock; 6.2 Channel and Network Redundancy; 6.3 Fault Models; 6.4 Fault-Tolerant Routing in SAF and VCT Networks; 6.5 Fault-Tolerant Routing in Wormhole-Switched Networks; 6.6 Fault-Tolerant Routing in PCS and Scouting Networks. - 6.7 Dynamic Fault Recovery6.8 Engineering Issues; 6.9 Commented References; Exercises; Problems; Chapter 7. Network Architectures; 7.1 Network Topology and Physical Constraints; 7.2 Router Architectures; 7.3 Engineering Issues; 7.4 Commented References; Exercises; Problems; Chapter 8. Messaging Layer Software; 8.1 Functionality of the Messaging Layer; 8.2 Impact of Message Processing Delays; 8.3 Implementation of the Messaging Layer; 8.4 Application Programming Layer: The Message Passing Interface; 8.5 Engineering Issues; 8.6 Commented References; Problems; Chapter 9. Performance Evaluation. - 9.1 Performance Metrics and Normalized Results. - The performance of most digital systems today is limited by their communication or interconnection, not by their logic or memory. As designers strive to make more efficient use of scarce interconnection bandwidth, interconnection networks are emerging as a nearly universal solution to the system-level communication problems for modern digital systems. Interconnection networks have become pervasive in their traditional application as processor-memory and processor-processor interconnect. Point-to-point interconnection networks have replaced buses in an ever widening range
Emner
Sjanger
Dewey
ISBN
1558608524

Bibliotek som har denne