Hamming Transcoders for Power Reduction on Internal Buses

1
Hamming Transcoders for Power Reduction on
Internal Buses

• Victor Wen
• Jan. 13, 2000
• University of California, Berkeley

2
Outline

• Motivations
• Related Work
• Initial Approaches
• Transition Code Technique
• Preliminary Results
• Future Work/Conclusion

3
Power reduction through coding
Encoded Version

• Can we encode information in a way that takes
  less power?
• Do this on chip?!

4
Reasoning

• Increasing importance of wires relative to
  transistors
• Spend transistors to drive wires more
  efficiently?
• Try to reduce transitions over wires
• Orthogonal to other power-saving techniques
• I.e. voltage reduction, low-swing drive
• clock gating
• Parallelism (like vectors!)
• Portable devices more important

5
Related Work

• Bus Invert Coding, by M. R. Stan and W. P.
  Burleson
• Reduce peak power by 50, avg by up to 25
• Work-zone Encoding, by E. Musoll et al.
• Compare favorably with other techniques
• Test Vector Ordering, by P. Girard et al.
• Result 8.2 to 54.1 less activities
• Minimizing Power consumption, by A. Chandrakasan
  and R. Broderson

6
Huffman-based Compression

• Variable bit length problem!
• Possible soln macro clock
• Less bits ! less transitions

7
Hamming Weight

• Find a map function to minimize transition
• Search space is large 256! (For 8-bit bus)
• Leads to transition code idea

8
Hamming Transcoder
State Transition Diagram
256x256 table for 8 bit bus
Code 0xFF Freq 10
Code 0x00 Freq 2620

• Most frequent arc assigned low-weight codes
• Use output codes to XOR transmission line
• Every 1 in coded version causes transistion
• Most frequent arcs cause least number of
  transitions

9
Hamming Transcoder (cont)

• Only transitions matter, not absolute value
• Recognize more frequent transitions assign
  low-weight code to them
• Guarantees more frequent transitions have less
  bits changes on the wire

10
Transition Code Setup
9
8
8
Coder
Decoder
11
Simulation Setup

• Sun offering processor descriptions in Verilog
• picoJava (for now)
• UltraSparc (soon)

12
Simulation Results (1)

• Savings
• Rank 9 saves 79.52
• Rank 256 saves 79.68

• 9th bit overhead
• Rank 1 23
• Rank 9 0.29

13
Simulation Results (2)
Number of transitions drops quickly as ranks
increases 256x256 table might not be
necessary Other trace files show similar trends
Note icu_data connects between instruction cache
unit and integer unit. A fairly long bus
according to picoJavas floorplan
14
Conclusion Future Work

• Conclusion
• Transition coding attacks the root of the problem
• Minimal change to existing circuits
• Orthogonal to other low power techniques
• Future work
• Simulate SPEC on Sparc UltraSparc RTL
• Build adaptability into coder/decoder
• Use of more history
• Implement actual hardware