Optimal VLSI Architectural Synthesis

Although research in architectural synthesis has been conducted for over ten years it has had very little impact on industry. This in our view is due to the inability of current architectural synthesizers to provide area-delay competitive (or "optimal") architectures, that will support interfaces to analog, asynchronous, and other complex processes. They also fail to incorporate testability. The OASIC (optimal architectural synthesis with interface constraints) architectural synthesizer and the CATREE (computer aided trees) synthesizer demonstrate how these problems can be solved. Traditionally architectural synthesis is viewed as NP hard and there­ fore most research has involved heuristics. OASIC demonstrates by using an IP approach (using polyhedral analysis), that most input algo­ rithms can be synthesized very fast into globally optimal architectures. Since a mathematical model is used, complex interface constraints can easily be incorporated and solved. Research in test incorporation has in general been separate from syn­ thesis research. This is due to the fact that traditional test research has been at the gate or lower level of design representation. Nevertheless as technologies scale down, and complexity of design scales up, the push for reducing testing times is increased. On way to deal with this is to incorporate test strategies early in the design process. The second half of this text examines an approach for integrating architectural synthesis with test incorporation. Research showed that test must be considered during synthesis to provide good architectural solutions which minimize Xlll area delay cost functions.

I: Introduction.- 1. Global VLSI Design Cycle.- 1.1 VLSI Design Cycle.- 1.2 Hybrid Systems Design.- 1.3 Impact Of Technologies.- 1.4 Test Considerations.- 1.5 Bottlenecks and Open Issues.- 1.6 focus of Text.- 2. Behavioral and Structural Interfaces.- 2.1 Input to an Architectural Synthesizer.- 2.2 Interface Primitives for External Processes.- 2.3 Output Primitives from an Architectural Synthesizer.- II: Review And Background.- 3. State of the Art Synthesis.- 3.1 Terminology and Subtask Definitions.- 3.2 High Level Transformations.- 3.3 Independent Subtask Optimization.- 3.3.1 Scheduling.- 3.3.2 Resource Allocation.- 3.4 Iterative and Simultaneous Approaches.- 3.5 Mathematical Approaches.- 3.5.1 Branch and Bound.- 3.5.2 Simulated Annealing.- 3.5.3 Makespan Scheduling.- 3.5.4 Feasibility Models.- 3.6 Timing Constrained Synthesis.- 3.7 Cost Functions for Design Evaluation.- 4. Introduction to Integer Programming.- 4.1 Applications and Models.- 4.2 Solution of Unstructured IPs.- 4.3 Polyhedral Approaches To Solving Ips.- 4.4 The Node Packing Problem.- 4.5 The Knapsack Problem.- III: Optimal Architectural Synthesis With Interfaces.- 5. A Methodology for Architectural Synthesis.- 5.1 Requirements for High Level Synthesis Tools.- 5.2 High Level Methodology.- 5.3 OASIC Methodology.- 5.4 An Introduction to OASIC.- 5.5 Oasic Terminology, Assumptions, and Preprocessing.- 5.5.1 terminology.- 5.5.2 Assumptions.- 5.5.3 Preprocessing.- 6. Simultaneous Scheduling, and Selection and Allocation Of Functional Units.- 6.1 The Formal Model.- 6.2 Cost Functions.- 6.3 Functional Unit Type Selection.- 7. Oasic: Area-Delay Constrained Architectural Synthesis.- 7.1 The Precedence Constrained Scheduling Model.- 7.2 Functional Unit Allocation.- 7.3 Register Allocation.- 7.4 Bus Allocation.- 7.5 Cost Functions.- 7.6 Application Specific Tightening of Con-straints.- 8. Support for Algorithmic Constructs.- 8.1 Conditional Code.- 8.2 Loops.- 8.3 Functional Pipelining.- 9. Interface Constraints.- 9.1 General Interface: Minimum and Max-imum Timing Constraints.- 9.2 Analog Interface: Fixed Timing Con-straint.- 9.3 Asynchronous Interface.- 9.4 Unknown Unbounded Delays.- 9.5 Complex Timing Constraints.- 10. Oasic Synthesis Results.- 10.1 Elliptical Wave Filter.- 10.1.1 Structured Model.- 10.1.2 Area-delay Optimized.- 10.2 Neural Network Algorithm.- 10.3 Conditional Code Example.- 10.4 Analog and Asynchronous interface Examples.- 10.6.1 Analog Interface.- 10.6.2 Asynchronous Interface.- IV: Testable Architectural Synthesis.- 11. Testability in Architectural Synthesis.- 11.1 Design and Test.- 11.1.1 Choices in Design and Test.- 11.2 Approaches to Testability.- 11.2.1 Test Measures and Tools.- 11.2.2 Design Modifications for Testability.- 11.3 Previous Research in Design for Test.- 11.4 Approaches To Test With Synthesis.- 11.4.1 Previous Research.- 11.4.2 Commercial Systems.- 11.5 Inadequacies of Current Synthesis With Test.- 11.5.1 Feedback.- 11.5.2 Integration.- 11.5.3 Constraint Estimation.- 12. The Catree Architectural Synthesis With Testability.- 12.1 problem description.- 12.2 Comparison With Previous Research.- 12.3 Two Synthesis With Test Methodologies:Catree & Catree2.- 12.4 Catree Design Synthesis Stages.- 12.4.1 Input Specification.- 12.4.2 Design Allocation.- 12.4.3 Catree Area and Delay Estimates.- 12.4.4 Test Incorporation.- 12.4.5 Feedback.- 12.5 Catree Synthesis Results.- 12.6 Catree Discussion.- 12.7 Catree2 Design Synthesis stages.- 12.7.1 Tree Formation and Functional Unit Binding.- 12.7.2 Test Incorporation.- 12.7.3 (Test) Register and Bus Binding.- 12.7.4 Feedback.- 12.8 Catree2 Experiments.- 12.9 Catree2 Discussion.- V: Summary and Future Research.- 13. Summary and Future Research.- 13.1 Oasic Summary.- 13.2 Catree Summary.- 13.3 Future Extensions.- 13.4 Concluding Remarks.- References.