Sheibanyrad / Pétrot / Jantsch | 3D Integration for NoC-based SoC Architectures | E-Book | www.sack.de
E-Book

E-Book, Englisch, 278 Seiten

Reihe: Integrated Circuits and Systems

Sheibanyrad / Pétrot / Jantsch 3D Integration for NoC-based SoC Architectures


1. Auflage 2010
ISBN: 978-1-4419-7618-5
Verlag: Springer
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 278 Seiten

Reihe: Integrated Circuits and Systems

ISBN: 978-1-4419-7618-5
Verlag: Springer
Format: PDF
Kopierschutz: 1 - PDF Watermark



This book presents the research challenges that are due to the introduction of the 3rd dimension in chips for researchers and covers the whole architectural design approach for 3D-SoCs. Nowadays the 3D-Integration technologies, 3D-Design techniques, and 3D-Architectures are emerging as interesting, truly hot, broad topics. The present book gathers the recent advances in the whole domain by renowned experts in the field to build a comprehensive and consistent book around the hot topics of three-dimensional architectures and micro-architectures. This book includes contributions from high level international teams working in this field.

Sheibanyrad / Pétrot / Jantsch 3D Integration for NoC-based SoC Architectures jetzt bestellen!

Weitere Infos & Material


1;Preface;5
2;Contents;7
3;Contributors;9
4;Part I 3DI Promises and Challenges;11
4.1;Three-Dimensional Integration of Integrated Circuits—an Introduction;12
4.1.1;1.1 Background and Introduction;12
4.1.2;1.2 Motivations and Drivers;13
4.1.2.1;1.2.1 Sustainable IC Performance Growth;13
4.1.2.2;1.2.2 Show-Stoppers and 3-D Integration as a Remedy;15
4.1.2.2.1;1.2.2.1 Transistor Scaling Barriers;15
4.1.2.2.2;1.2.2.2 On-Chip Interconnect;16
4.1.2.2.3;1.2.2.3 Off-Chip Interconnect (Memory Bandwidth Gap);17
4.1.3;1.3 Options of 3D IC;19
4.1.3.1;1.3.1 System Integration Landscape;19
4.1.3.2;1.3.2 Classification;19
4.1.3.3;1.3.3 Monolithic Approaches;20
4.1.3.4;1.3.4 Assembly Approaches;21
4.1.4;1.4 Technology Platforms and Strategies;25
4.1.4.1;1.4.1 Through Silicon Via;25
4.1.4.2;1.4.2 Cu–Cu Permanent Bonding;26
4.1.4.2.1;1.4.2.1 Wafer Preparation and Bonding Procedures;27
4.1.4.2.2;1.4.2.2 Bonding Mechanism;27
4.1.4.2.3;1.4.2.3 Surface Oxide;28
4.1.4.2.4;1.4.2.4 Process Parameters;28
4.1.4.2.5;1.4.2.5 Observation of Interfacial Voids;29
4.1.4.2.6;1.4.2.6 Non-Blanket Cu Bonding;30
4.1.4.2.7;1.4.2.7 Low Temperature Cu Bonding;31
4.1.5;1.5 Applications, Status and Outlook;32
4.1.6;References;34
4.2;The Promises and Limitations of 3-D Integration;36
4.2.1;2.1 Introduction;36
4.2.2;2.2 Computational Efficiency of Silicon;38
4.2.2.1;2.2.1 Computation;38
4.2.2.2;2.2.2 Adding Memory and Communication;39
4.2.2.3;2.2.3 Effective Computational Efficiency;40
4.2.2.4;2.2.4 Effective Computational Density;41
4.2.3;2.3 Technology Parameter Scaling;43
4.2.3.1;2.3.1 Planar 2-D Wire Models;43
4.2.3.2;2.3.2 3-D TSV Interconnect Models;43
4.2.3.3;2.3.3 Logical Operation and DRAM Scaling;44
4.2.4;2.4 ECE Trends and Dependencies;45
4.2.4.1;2.4.1 Distributed Versus Central Memory;45
4.2.4.2;2.4.2 On-chip Versus Off-chip Memory;46
4.2.4.3;2.4.3 Size Constrained System;48
4.2.4.4;2.4.4 Power and Frequency Constrained Systems;50
4.2.5;2.5 Conclusion;51
5;Part II Technology and Circuit Design;54
5.1;Testing 3D Stacked ICs Containing Through-Silicon Vias;55
5.1.1;3.1 Introduction;55
5.1.2;3.2 3D-SICs Based on TSV Technology;56
5.1.3;3.3 3D Test Flows;60
5.1.3.1;3.3.1 Prior Art in Test Flows;60
5.1.3.2;3.3.2 Test Flows for 3D-SICs;61
5.1.3.3;3.3.3 Pre-Bond Testing for W2W Stacking;63
5.1.4;3.4 Modular Testing;64
5.1.5;3.5 3D Test Contents;66
5.1.5.1;3.5.1 New Intra-Die Defects;66
5.1.5.2;3.5.2 Test of TSV-Based Interconnects;67
5.1.6;3.6 3D Wafer Test Challenges;69
5.1.6.1;3.6.1 Pre-Bond Wafer Test Access;69
5.1.6.2;3.6.2 Post-Bond Wafer Handling;73
5.1.7;3.7 3D On-Chip DfT Architecture;73
5.1.7.1;3.7.1 Hierarchical DfT Requirements;73
5.1.7.2;3.7.2 Simplified Example DfT Architecture;74
5.1.7.3;3.7.3 Advanced DfT Techniques and Test Resource Partitioning;77
5.1.8;3.8 Conclusion;78
5.2;Design and Computer Aided Design of 3DIC;83
5.2.1;4.1 Introduction;83
5.2.2;4.2 Technology Selection;83
5.2.2.1;4.2.1 What Am I Stacking (Fig. 4.1)?;83
5.2.2.2;4.2.2 Do I Need Through Silicon Vias (Fig. 4.3)?;85
5.2.2.3;4.2.3 Through Silicon Via (TSV) Technology;85
5.2.3;4.3 3D Specific Optimizations;87
5.2.3.1;4.3.1 Example of 3D Specific Optimized Design;88
5.2.4;4.4 3D Design and CAD;91
5.2.5;4.5 Outstanding Issues in 3D Design;93
5.2.5.1;4.5.1 Cost Management;93
5.2.5.2;4.5.2 Computer Aided Design;94
5.2.5.3;4.5.3 Thermal Design and Analysis;94
5.2.5.4;4.5.4 Test and Design for Test;94
5.2.6;4.6 Conclusions;95
5.3;Physical Analysis of NoC Topologies for 3-D Integrated Systems;97
5.3.1;5.1 Introduction;97
5.3.2;5.2 Three-Dimensional On-Chip Network Topologies;98
5.3.3;5.3 Timing and Power Model for 3-D NoCs;100
5.3.3.1;5.3.1 Latency Model for 3-D NoC;100
5.3.3.2;5.3.2 Power Consumption Model for 3-D NoC;105
5.3.4;5.4 Thermal-Aware Analysis Methodology;106
5.3.5;5.5 Latency, Power, and Temperature Tradeoffs in 3-D NoC Topologies;110
5.3.5.1;5.5.1 Latency of 3-D NoCs;111
5.3.5.2;5.5.2 Power Dissipation of 3-D NoCs;113
5.3.5.3;5.5.3 Temperature in 3-D NoCs;116
5.3.6;5.6 Summary;119
5.4;Three-Dimensional Networks-on-Chip: Performance Evaluation;123
5.4.1;6.1 Introduction;123
5.4.2;6.2 3D NoC Architectures;125
5.4.2.1;6.2.1 Mesh-Based Networks;125
5.4.2.2;6.2.2 Tree-Based Networks;128
5.4.3;6.3 Performance Evaluation;128
5.4.3.1;6.3.1 Performance Metrics;128
5.4.3.2;6.3.2 Performance Analysis of 3D Mesh-Based NoCs;130
5.4.3.3;6.3.3 Performance Analysis of 3D Tree-Based NoCs;132
5.4.3.4;6.3.4 Simulation Methodology;133
5.4.3.5;6.3.5 Experimental Results for Mesh-Based Networks;135
5.4.3.6;6.3.6 Experimental Results for Tree-Based Networks;138
5.4.3.7;6.3.7 Effects of Traffic Localization;141
5.4.3.8;6.3.8 Effects of Wire Delay on Latency and Bandwidth;143
5.4.3.9;6.3.9 Network Aspect Ratio;144
5.4.3.10;6.3.10 Multi-Layer IPs;146
5.4.4;6.4 Heat Dissipation Profile of 3D NoCs;147
5.4.4.1;6.4.1 Temperature Analysis;147
5.4.4.2;6.4.2 The Relationship Between Temperature and Energy;148
5.4.4.3;6.4.3 Simulation Methodology;148
5.4.4.4;6.4.4 Experimental Results;149
5.4.5;6.5 Conclusion;151
6;Part III System and Architecture Design;154
6.1;Asynchronous 3D-NoCs Making Use of Serialized Vertical Links;155
6.1.1;7.1 Introduction;155
6.1.2;7.2 3D-Integration Technology;156
6.1.2.1;7.2.1 TSV Technology Challenges;157
6.1.3;7.3 NoC and 3D-Design Contribution;158
6.1.4;7.4 Asynchronous Circuit Exploitation;160
6.1.5;7.5 Serialized Vertical Links;163
6.1.5.1;7.5.1 Implementation;165
6.1.6;7.6 Conclusion;170
6.2;Design of Application-Specific 3D Networks-on-Chip Architectures;172
6.2.1;8.1 Introduction;172
6.2.2;8.2 Related Work;174
6.2.3;8.3 Design Flow;175
6.2.3.1;8.3.1 Floorplanning;175
6.2.3.2;8.3.2 3D Networks-on-Chip Synthesis;176
6.2.3.3;8.3.3 NoC Objectives and Constraints;176
6.2.3.4;8.3.4 NoC Design Parameters;177
6.2.3.5;8.3.5 Detailed Design;178
6.2.4;8.4 Problem Description and Formulation;178
6.2.4.1;8.4.1 Problem Description;178
6.2.4.2;8.4.2 Problem Formulation;179
6.2.5;8.5 3D Design Models;180
6.2.5.1;8.5.1 3D Interconnect Modelling;180
6.2.5.2;8.5.2 Modelling Routers;182
6.2.6;8.6 Design Algorithms;183
6.2.6.1;8.6.1 Initial Network Construction;183
6.2.6.2;8.6.2 Flow Ripup and Rerouting;185
6.2.6.3;8.6.3 Router Merging;188
6.2.6.4;8.6.4 Complexity of the Algorithm;190
6.2.7;8.7 Deadlock Considerations;190
6.2.8;8.8 Results;190
6.2.8.1;8.8.1 Experimental Setup;190
6.2.8.2;8.8.2 Comparison of Results;193
6.2.9;8.9 Conclusions;194
6.3;3D Network on Chip Topology Synthesis: Designing Custom Topologies for Chip Stacks;197
6.3.1;9.1 Introduction;197
6.3.1.1;9.1.1 3D-Stacking;197
6.3.1.2;9.1.2 Networks on Chips for 3D ICs;199
6.3.1.3;9.1.3 Designing NoCs for 3D ICs;201
6.3.2;9.2 3D Architecture and Design Flow;204
6.3.3;9.3 Design Flow Assumptions;205
6.3.4;9.4 Design Approach;206
6.3.5;9.5 Algorithm;208
6.3.5.1;9.5.1 Phase 1;209
6.3.5.2;9.5.2 Phase 2;212
6.3.5.3;9.5.3 Find Paths;213
6.3.5.4;9.5.4 Switch Position Computation;215
6.3.6;9.6 Experiments and Case Studies;221
6.3.6.1;9.6.1 Multimedia SoC Case Study;222
6.3.6.2;9.6.2 Impact of Inter-layer Link Constraint and Comparisons with Mesh;224
6.3.7;9.7 Conclusions;226
6.4;3-D NoC on Inductive Wireless Interconnect;228
6.4.1;10.2 Wireless Interconnect with Inductive Coupling;229
6.4.2;10.3 Wireless 3-D NoC for Building Block 3-D ICs;229
6.4.2.1;10.3.1 Toward Building Block 3-D ICs;231
6.4.2.1.1;10.3.1.1 Application Chip Design;232
6.4.2.1.2;10.3.1.2 Service Chip Design;232
6.4.2.2;10.3.2 Wireless 3-D Crossbar Architecture;232
6.4.2.2.1;10.3.2.1 Time-Division Multiplex Transfer;232
6.4.2.2.2;10.3.2.2 Static TDM Scheme;233
6.4.2.2.3;10.3.2.3 Dynamic TDM Scheme;234
6.4.2.2.4;10.3.2.4 Hardware Structure;235
6.4.2.2.5;10.3.2.5 TX Module Architecture;235
6.4.2.2.6;10.3.2.6 RX Module Architecture;236
6.4.2.2.7;10.3.2.7 Vertical Crossbar Architecture;236
6.4.2.2.8;10.3.2.8 Communication Protocol;237
6.4.2.2.9;10.3.2.9 Downstream Transfer;237
6.4.2.2.10;10.3.2.10 Upstream Transfer;238
6.4.3;10.4 An Implementation Example: MuCCRA-Cube;239
6.4.3.1;10.4.1 PE Array Structure in a Die;239
6.4.3.1.1;10.4.1.1 PE Structure;240
6.4.3.2;10.4.2 Homogeneous Stacking;242
6.4.3.3;10.4.3 3-D Inductive Connection Channels;244
6.4.3.4;10.4.4 Three-Phase Interleaving Scheme;245
6.4.3.5;10.4.5 Prototype Chip;246
6.4.3.6;10.4.6 Communication Link;247
6.4.3.7;10.4.7 Execution Performance;248
6.4.4;10.5 Research Perspective;249
6.5;Influence of Stacked 3D Memory/Cache Architectures on GPUs;252
6.5.1;11.1 Introduction;252
6.5.2;11.2 Background;253
6.5.2.1;11.2.1 How Do GPUs Work?;253
6.5.2.2;11.2.2 GPU Hardware Architecture;254
6.5.2.3;11.2.3 An Example of a Modern GPU: NVIDIA GeForce GTX280;257
6.5.2.4;11.2.4 3D Technology;258
6.5.2.5;11.2.5 MRAM;259
6.5.3;11.3 Employing 3D Technology in GPUs;262
6.5.3.1;11.3.1 Why 3D Technology?;262
6.5.3.2;11.3.2 Area Estimation Model;263
6.5.3.3;11.3.3 Cost Model;263
6.5.3.4;11.3.4 Power Model and Thermal Profiling;265
6.5.4;11.4 Evaluating 3D-Based GPU;265
6.5.4.1;11.4.1 The Design Space;266
6.5.4.2;11.4.2 Simulation Environment of GPU Architecture;266
6.5.4.3;11.4.3 Experiments and Results;267
6.5.4.4;11.4.4 Iso-Cost Designs;269
6.5.4.5;11.4.5 Power and Thermal Constraints;271
6.5.4.6;11.4.6 MRAM as an Alternative to SRAM;272
6.5.5;11.5 Summary;273
7;Index;275



Ihre Fragen, Wünsche oder Anmerkungen
Vorname*
Nachname*
Ihre E-Mail-Adresse*
Kundennr.
Ihre Nachricht*
Lediglich mit * gekennzeichnete Felder sind Pflichtfelder.
Wenn Sie die im Kontaktformular eingegebenen Daten durch Klick auf den nachfolgenden Button übersenden, erklären Sie sich damit einverstanden, dass wir Ihr Angaben für die Beantwortung Ihrer Anfrage verwenden. Selbstverständlich werden Ihre Daten vertraulich behandelt und nicht an Dritte weitergegeben. Sie können der Verwendung Ihrer Daten jederzeit widersprechen. Das Datenhandling bei Sack Fachmedien erklären wir Ihnen in unserer Datenschutzerklärung.