did-you-know? rent-now

Amazon no longer offers textbook rentals. We do!

did-you-know? rent-now

Amazon no longer offers textbook rentals. We do!

We're the #1 textbook rental company. Let us show you why.

9789048130306

Network-on-Chip Architectures

by ; ;
  • ISBN13:

    9789048130306

  • ISBN10:

    9048130301

  • Format: Hardcover
  • Copyright: 2009-10-01
  • Publisher: Springer Verlag
  • Purchase Benefits
  • Free Shipping Icon Free Shipping On Orders Over $35!
    Your order must be $35 or more to qualify for free economy shipping. Bulk sales, PO's, Marketplace items, eBooks and apparel do not qualify for this offer.
  • eCampus.com Logo Get Rewarded for Ordering Your Textbooks! Enroll Now
List Price: $199.99 Save up to $149.70
  • Digital
    $108.96
    Add to Cart

    DURATION
    PRICE

Supplemental Materials

What is included with this book?

Summary

The continuing reduction of feature sizes into the nanoscale regime has led to dramatic increases in transistor densities. Integration at these levels has highlighted the criticality of the on-chip interconnects. Network-on-Chip (NoC) architectures are viewed as a possible solution to burgeoning global wiring delays in many-core chips, and have recently crystallized into a significant research domain. On-chip networks instill a new flavor to communication research due to their inherently resource-constrained nature. Despite the lightweight character demanded of the NoC components, modern designs require ultra-low communication latencies in order to cope with inflating data bandwidths. The work presented in Network-on-Chip Architectures addresses these issues through a comprehensive exploration of the design space. The design aspects of the NoC are viewed through a penta-faceted prism encompassing five major issues: (1) performance, (2) silicon area consumption, (3) power/energy efficiency, (4) reliability, and (5) variability. These five aspects serve as the fundamental design drivers and critical evaluation metrics in the quest for efficient NoC implementations. The research exploration employs a two-pronged approach: (a) MICRO-architectural innovations within the major NoC components, and (b) MACRO-architectural choices aiming to seamlessly merge the interconnection backbone with the remaining system modules. These two research threads and the aforementioned five key metrics mount a holistic and in-depth attack on most issues surrounding the design of NoCs in multi-core architectures.

Table of Contents

Introductionp. 1
The Diminishing Returns of Instruction-Level Parallelismp. 1
The Dawn of the Communication-Centric Revolutionp. 2
The Global Wiring Challengep. 2
The Network-on-Chip (NoC) Solutionp. 4
Overview of Researchp. 6
Legend for Figs. 1.4 and 1.5p. 7
A Baseline NoC Architecturep. 13
MICRO-Architectural Exploration
ViChaR: A Dynamic Virtual Channel Regulator for NoC Routers [39]p. 19
Importance of Buffer Size and Organizationp. 19
Related Work in Buffer Designp. 22
The Proposed Dynamic Virtual Channel Regulator (ViChaR)p. 24
Variable Number of Virtual Channelsp. 27
ViChaR Component Analysisp. 30
Simulation Resultsp. 35
Simulation Platformp. 35
Analysis of Resultsp. 35
Chapter Summaryp. 39
RoCo: The Row-Column Decoupled Router - A Gracefully Degrading and Energy-Efficient Modular Router Architecture for On-Chip Networks [40]p. 41
Introduction and Motivationp. 41
Related Work in Partitioned Router Architecturesp. 43
The Proposed Row-Column (RoCo) Decoupled Routerp. 44
Row-Column Switchp. 44
Blocking Delayp. 48
Concurrency Control for High-Contention Environmentsp. 50
Flexible and Reusable On-Chip Communicationp. 51
Fault-Tolerance Through Hardware Recyclingp. 51
Performance Evaluationp. 56
Simulation Platformp. 56
Energy Modelp. 57
A Performance, Energy, and Fault-Tolerance (PEF) Metricp. 57
Performance Resultsp. 58
Chapter Summaryp. 62
Exploring Faulto Tolerant Network-on-Chip Architectures [37]p. 65
Introduction and Motivationp. 65
Simulation Platform Preliminariesp. 67
Handling Link Soft Faultsp. 68
Flit-Based HBH Retransmission Schemep. 69
Deadlock Recoveryp. 72
Handling Soft Errors in Intra-Router Logicp. 77
Virtual Channel Arbiter Errorsp. 78
Routing Computation Unit Errorsp. 80
Switch Allocator Errorsp. 81
Crossbar Errorsp. 83
Retransmission Buffer Errorsp. 83
Handshaking Signal Errorsp. 83
Handling Hard Faultsp. 83
Proximity-Aware (PA) Fault-Tolerant Routing Algorithmp. 84
Extension of PA Routing for Hot-Spot Avoidancep. 86
Service-Oriented Networking (SON)p. 88
SON - Direction Lookup Table (DLT) and Service Information Provider (SIP)p. 89
Chapter Summaryp. 91
On the Effects of Process Variation in Network-on-Chip Architectures [45]p. 93
Introduction and Motivationp. 93
Related Work in Process Variation (PV)p. 94
The Impact of PV on NoC Architecturesp. 95
Evaluation Platformp. 95
PV Effects on Router Componentsp. 98
The Proposed SturdiSwitch Architecturep. 105
IntelliBuffer: A Leakage-Aware Elastic Buffer Structurep. 106
VA Compaction Mechanismp. 108
SA Folding Mechanismp. 111
Chapter Summaryp. 115
Macro-Architectural Exploration
The Quest for Scalable On-Chip Interconnection Networks: Bus/NoC Hybridization [15]p. 119
Introduction and Motivationp. 119
Exploration of Existing On-Chip Bus Architecturesp. 121
Traditional Bus Architecturesp. 121
TDMA Buses and Hybrid Interconnectsp. 123
Constraints of Traditional Busesp. 124
CDMA Interconnectsp. 125
The Dynamic Time-Division Multiple-Access (dTDMA) Busp. 127
Operation of the Dynamic Timeslot Allocationp. 128
Implementation of the dTDMA Busp. 129
Comparison with a Traditional Bus Architecturep. 130
dTDMA Bus Performancep. 132
Comparison with Networks-on-Chipp. 138
Experimental Setupp. 138
Resultsp. 139
Interconnect Hybridizationp. 141
Affinity Groupingp. 142
Simulation Methodologyp. 143
Hybridization Resultsp. 144
Chapter Summaryp. 146
Design and Management of 3D Chip Multiprocessors Using Network-In-Memory (NetInMem) [43]p. 147
Introduction and Motivationp. 149
Backgroundp. 149
NUCA Architecturesp. 150
Network-In-Memory (NetInMem)p. 150
Three-Dimensional (3D) Design and Architecturesp. 151
A 3D NetInMem Architecturep. 153
The dTDMA Bus as a Communication Pillarp. 155
CPU Placementp. 157
3D L2 Cache Managementp. 162
Processors and L2 Cache Organizationp. 162
Cache Management Policiesp. 163
Experimental Evaluationp. 164
Methodologyp. 164
Resultsp. 166
Chapter Summaryp. 169
A Novel Dimensionally-Decomposed Router for On-Chip Communication in 3D Architectures [44]p. 171
Introduction and Motivationp. 173
Three-Dimensional Network-on-Chip Architecturesp. 176
A 3D Symmetric NoC Architecturep. 176
The 3D NoC-Bus Hybrid Architecturep. 177
A True 3D NoC Routerp. 179
A Partially-Connected 3D NoC Router Architecturep. 182
The Proposed 3D Dimensionally-Decomposed (DimDe) NoC Router Architecturep. 182
Performance Evaluationp. 190
Simulation Platformp. 190
Energy Modelp. 191
Performance Resultsp. 191
Chapter Summaryp. 196
Digest of Additional NoC MACRO-Architectural Researchp. 199
A Distributed Multi-Point Network Interface for Low-Latency, Deadlock-Free On-Chip Interconnects [42]p. 199
Design of a Dynamic Priority-Based Fast Path Architecture for On-Chip Interconnects [46]p. 201
Exploring the Effects of Data Compression in NoC Architectures [47]p. 203
Conclusions and Future Workp. 207
Referencesp. 211
Table of Contents provided by Ingram. All Rights Reserved.

Supplemental Materials

What is included with this book?

The New copy of this book will include any supplemental materials advertised. Please check the title of the book to determine if it should include any access cards, study guides, lab manuals, CDs, etc.

The Used, Rental and eBook copies of this book are not guaranteed to include any supplemental materials. Typically, only the book itself is included. This is true even if the title states it includes any access cards, study guides, lab manuals, CDs, etc.

Rewards Program