An Evolutionary approach to Standard Cell Design: a proof of concept Anil Bahuman Artificial Intelligence Center Universit...
 
 
 
Research Goal Truth Table Layout with “minimum” area “ Schematic and layout are designed in parallel” Explore possibility ...
Standard Cell Design <ul><li>Building blocks for chips. </li></ul><ul><li>Frequently used logics. </li></ul><ul><li>NAND, ...
Design Example
A Transistor in MAGIC
Error tiles indicating DRC Errors
An Inverter in MAGIC Lambda, cell limits, labels, template, inputs, output, gate, terminals, wires, poly, contacts, 3I - 1O
SPICE simulation of the Inverter
Research Goal
Research Goal – Inverter Truth Table Layout with “minimum” area “ Schematic and layout are designed in parallel” 1 0 1 0 0...
Why Is This Worth Our Efforts? <ul><li>EDA is a multi-billion dollar industry </li></ul><ul><li>On-the-fly vs. Compaction ...
Architecture
architecture MAGIC  SPICE Modified GADO (DRC ERRORS , AREA) (CORRECTNESS) DESIGN ENGINE EVALUATOR DESIGN RULE CHECKER CIRC...
Errors in a Design : 1 <ul><li>Shorted Labels </li></ul><ul><li>Overlapping transistors </li></ul><ul><li>Transistors shou...
<ul><li>Incomplete designs – broken connections </li></ul><ul><li>10+penalty for broken connections + DRC </li></ul><ul><l...
11 Building Blocks
Encoding an Object A B 14 0 Cell limits Y 15 0 Cell limits X 5 0 Cell limits Stretch Factor 2 0 0-3 Orientation 11 11 1-15...
An Individual Y X Stretch Factor Orientation Object Type
Connections b/w Transistors terminal
Key <ul><li>MAGIC </li></ul><ul><li>SPICE </li></ul>“ If you have some terminal that is not being influenced by any other ...
An Influence Check <ul><li>Domain specific rules encouraging connectivity </li></ul><ul><ul><li>The labels must not be sho...
Sample Cell Corresponding Graph
Results
Success 1
Success 2
Success 3
Evolving 1
Evolving 2
Evolving 3
Evolving 4-7 4 6 5 7
Evolving 8: Aha! Inverter
Evolving – Success 1
Limitations <ul><li>Does not always find the best solution – Is this acceptable? </li></ul><ul><li>Presently unable to des...
References 1 <ul><li>T. Lengauer.  Combinatorial Algorithms for Integrated Circuit Layout. </li></ul><ul><li>K. Rasheed.  ...
References 2 <ul><li>N. H. E. Weste, K. Eshragian.  Principles of CMOS VLSI Design. </li></ul><ul><li>C. Edwards . EDA Ven...
Merci  Danke   Sas efharisto   Mahalo  Merci  Dhanyavaad  Arigato   Vandane  Wneeweh  Shukran
Upcoming SlideShare
Loading in …5
×

Example of Automating Transistor-level Design

1,760 views

Published on

Layman Description:
Computers have been assisting humans in design of artifacts for several years now, such as design of aircraft or design of Integrated Circuits. We demonstrate one such tool that can assist engineers to build standard cells on-the-fly customized to criteria specified by them. We use genetic algorithms to guide the computer. This starts with an engineer providing the desired output signals. The computer program then randomly generates say 10,000 possible design configurations. Good designs are then Selected from this population of designs (priority given to designs that bettter meet the engineers' output requirement). Hybrid designs are developed using genetic operators.
Heres where it gets interesting. Unlike most traditional computer programs, the genetic algorithm create a hybrid design by a "Crossover" of two good "Parent" designs (creating, hopefully, a better design) or random "Mutations" of good designs. These genetic operators are used on the initial population of 10,000 designs to create a second population of another (say) 10,000 designs. What happens when you do this 1000s of times? Do you arrive at the "perfect" (optimized) design? Thats what this work is about.

ACKNOWLEDGEMENT
This work at the AI CENTER & DEPT. OF COMPUTER SCIENCE & UNI OF GEORGIA in 2001 demonstrated designs an Inverter Standard Cell on-the-fly. It was sponsored by a YAMACRAW grant from the STATE OF GEORGIA, UNITED STATES.

TECHNICAL DESCRIPTION
The problem of designing the transistor-level layout of cells in a standard cell library is a multi-objective design optimization problem. Contemporary methods are optimization or compaction engines that rely on a schematic representation
supplied by a design engineer. We have demonstrated the possibility of applying
a modified genetic algorithm (GADO) to design a cell given only a behavioral description. A working inverter is designed as a proof of concept along with other inverters with arbitrary label placements. This is an example of design without
human intervention, i.e. a computer design as opposed to a computer-aided design.

Published in: Technology
0 Comments
1 Like
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
1,760
On SlideShare
0
From Embeds
0
Number of Embeds
7
Actions
Shares
0
Downloads
27
Comments
0
Likes
1
Embeds 0
No embeds

No notes for slide

Example of Automating Transistor-level Design

  1. 1. An Evolutionary approach to Standard Cell Design: a proof of concept Anil Bahuman Artificial Intelligence Center University of Georgia July 2001
  2. 5. Research Goal Truth Table Layout with “minimum” area “ Schematic and layout are designed in parallel” Explore possibility of automating the design of simple logic standard cells by exploring design spaces not considered by human designers 1 0 1 0 0 1 Out In
  3. 6. Standard Cell Design <ul><li>Building blocks for chips. </li></ul><ul><li>Frequently used logics. </li></ul><ul><li>NAND, full adder, latch etc. </li></ul><ul><li>Costly (time + money) to redesign. </li></ul><ul><li>Can we design on-the-fly? </li></ul>Standard cells are building blocks of frequently used logics employed in VLSI Design. Examples include NAND, Full Adder, Latch and Inverter. These libraries are typically designed by hand in a very costly and time consuming process. One of the major challenges is the migration of standard cell layouts every time there is a change in the process. There is a strong need for automation alternatives.
  4. 7. Design Example
  5. 8. A Transistor in MAGIC
  6. 9. Error tiles indicating DRC Errors
  7. 10. An Inverter in MAGIC Lambda, cell limits, labels, template, inputs, output, gate, terminals, wires, poly, contacts, 3I - 1O
  8. 11. SPICE simulation of the Inverter
  9. 12. Research Goal
  10. 13. Research Goal – Inverter Truth Table Layout with “minimum” area “ Schematic and layout are designed in parallel” 1 0 1 0 0 1 Out In
  11. 14. Why Is This Worth Our Efforts? <ul><li>EDA is a multi-billion dollar industry </li></ul><ul><li>On-the-fly vs. Compaction </li></ul><ul><li>Traditional methods are schematic-dependent </li></ul><ul><li>In our knowledge GAs have not been used for the design of standard cells </li></ul>
  12. 15. Architecture
  13. 16. architecture MAGIC SPICE Modified GADO (DRC ERRORS , AREA) (CORRECTNESS) DESIGN ENGINE EVALUATOR DESIGN RULE CHECKER CIRCUIT SIMULATOR Fitness module (CIRCUIT FILE)
  14. 17. Errors in a Design : 1 <ul><li>Shorted Labels </li></ul><ul><li>Overlapping transistors </li></ul><ul><li>Transistors should only touch Poly or Diff </li></ul><ul><li>Poly Contacts should NOT touch diffusion or diff contact </li></ul><ul><li>Poly should NOT touch diffusion contact </li></ul><ul><li>Pdiff Contact should not touch Ndiff contact </li></ul>FITNESS FUNCTION PENALIZES MAX_PENALTY FOR…
  15. 18. <ul><li>Incomplete designs – broken connections </li></ul><ul><li>10+penalty for broken connections + DRC </li></ul><ul><li>Complete designs with DRC errors and/or Circuit simulation errors </li></ul><ul><li>Penalty for “incorrect simulation” + DRC errors </li></ul>Errors in a Design : 2 FITNESS FUNCTION PENALIZES…
  16. 19. 11 Building Blocks
  17. 20. Encoding an Object A B 14 0 Cell limits Y 15 0 Cell limits X 5 0 Cell limits Stretch Factor 2 0 0-3 Orientation 11 11 1-15 Object Type B A VALUE PARAMETER
  18. 21. An Individual Y X Stretch Factor Orientation Object Type
  19. 22. Connections b/w Transistors terminal
  20. 23. Key <ul><li>MAGIC </li></ul><ul><li>SPICE </li></ul>“ If you have some terminal that is not being influenced by any other terminal, we want to know how close it is to some terminal that can influence it. ”
  21. 24. An Influence Check <ul><li>Domain specific rules encouraging connectivity </li></ul><ul><ul><li>The labels must not be shorted </li></ul></ul><ul><ul><li>Every input must influence at least one output </li></ul></ul><ul><ul><li>Every output must be influenced by at least one input </li></ul></ul><ul><ul><li>The gate of a transistor must be influenced by at least one input </li></ul></ul><ul><ul><li>One terminal of the transistor must be influenced by an input </li></ul></ul><ul><ul><li>Other terminal of the transistor must influence an output </li></ul></ul>
  22. 25. Sample Cell Corresponding Graph
  23. 26. Results
  24. 27. Success 1
  25. 28. Success 2
  26. 29. Success 3
  27. 30. Evolving 1
  28. 31. Evolving 2
  29. 32. Evolving 3
  30. 33. Evolving 4-7 4 6 5 7
  31. 34. Evolving 8: Aha! Inverter
  32. 35. Evolving – Success 1
  33. 36. Limitations <ul><li>Does not always find the best solution – Is this acceptable? </li></ul><ul><li>Presently unable to design more complex cells </li></ul><ul><ul><li>Alternative representation </li></ul></ul><ul><ul><li>Starting from similar designs </li></ul></ul><ul><li>Speed – almost linear speedup [Mazumder] </li></ul>
  34. 37. References 1 <ul><li>T. Lengauer. Combinatorial Algorithms for Integrated Circuit Layout. </li></ul><ul><li>K. Rasheed. GADO: A Genetic Algorithm for Continuous Design Optimization, PhD. Thesis. http://www.cs.uga.edu/~khaled </li></ul><ul><li>P. Mazumder and E. M. Rudnick. Genetic Algorithms for VLSI Design, Layout & Test Automation. </li></ul><ul><li>D. E. Goldberg. Genetic Algorithms in Search, Optimization and Machine Learning. </li></ul><ul><li>J. Rabaey. Digital Integrated Circuits: A Design Perspective. </li></ul>
  35. 38. References 2 <ul><li>N. H. E. Weste, K. Eshragian. Principles of CMOS VLSI Design. </li></ul><ul><li>C. Edwards . EDA Vendors Rethink Standard-Cell Libraries, Electronics Times, June 2000. </li></ul><ul><li>D. Pietromonaco. Automating Cost-Effective Library Creation, Integrated System Design, November 2000. </li></ul><ul><li>http://www.research.compaq.com/wrl/projects/ magic / </li></ul><ul><li>http://bwrc.eecs.berkeley.edu/Classes/IcBook/ SPICE / </li></ul><ul><li>http://ece.www.colorado.edu/~ecen4228/ spice /spice.htm </li></ul>
  36. 39. Merci Danke Sas efharisto Mahalo Merci Dhanyavaad Arigato Vandane Wneeweh Shukran

×