AnySP: Anytime Anywhere Anyway Signal Processing

1
AnySP Anytime Anywhere Anyway Signal Processing

• Mark Woh1, Sangwon Seo1, Scott Mahlke1,Trevor
  Mudge1,
• Chaitali Chakrabarti2, Krisztian Flautner3
• University of Michigan ACAL1
• Arizona State University2
• ARM, Ltd.3

2
The Modern Mobile Phone
The Old Mobile Phone

• Future phones are becoming more complex
• Richer applications require much more
  requirements
• How do phones handle this now?

Video Recording
Video Editing
Higher Data Rates
3D Rendering
Advanced Image Processing
Photos From - http//www.engadget.com/2009/06/10/
iphone-3g-s-supports-opengl-es-2-0-but-3g-only-sup
ports-1-1/ http//www.apple.com/iphone
3
Inside Todays Smart Phones

• Modern phones are looking like Frankenchips!
• Some cores unused and functionality duplicated

4
Cost for Multi-System Support

• Programmable Unified Architectures Provide
• Lower Cost
• Faster Time to Market
• Support for Multiple Applications (Current and
  Future)
• Bug Fixes After Manufacturing
• So where do we start?

• Supporting multiple systems is reserved for the
  most expensive phones
• Cost is in supporting all the systems that may or
  may not be used at once

Data gathered from - Ramacher, U. 2007.
Software-Defined Radio Prospects for
Multistandard Mobile Phones. Computer 40, 10
(Oct. 2007) - Finchelstein, D.F. Sze, V.
Sinangil, M.E. Koken, Y. Chandrakasan, A.P., "A
low-power 0.7-V H.264 720p video decoder,"
Solid-State Circuits Conference, 2008. A-SSCC
'08.
5
Power/Performance Requirements for Multiple
Systems
Different applications have different
power/performance characteristics! We need to
design keeping each application in mind! (Not GPP
but Domain Specific Processor)
6
The Applications

• Is there anything we can learn from the
  applications themselves?

7
H.264 Basics
T.-A. Liu, T.-M. Lin, S. -Z. Wang, et al. A
low-power dual-mode video decoder for mobile
applications, IEEE Communications Magazine,
volume 44, issue 8, pp.119-126, Aug. 2006.
8
4G Wireless Basics

• Three kernels make up the majority of the work
• FFT Extract Data from Signals
• STBC Combine Data into More Reliable Stream
• LDPC Error Correction on Data Stream

9
Mobile Signal Processing Algorithm Characteristics
Algorithm SIMD Scalar Overhead SIMD Width Amount
Algorithm Workload () Workload () Workload () (Elements) of TLP
4G FFT 75 5 20 1024 Low
4G STBC 81 5 14 4 High
4G LDPC 49 18 33 96 Low
H.264 Deblocking Filter 72 13 15 8 Medium
H.264 Intra-Prediction 85 5 10 16 Medium
H.264 Inverse Transform 80 5 15 8 High
H.264 Motion Compensation 75 5 10 8 High

• SIMD comes at a cost!
• Register File Power
• Data Movement/Alignment Cost
• SIMD architectures have to deal with this!

• Algorithms have different SIMD widths
• From very large to very small
• Though SIMD width varies all algorithms can
  exploit it
• Large percentage of work can be SIMDized
• Larger SIMD width tend to have less TLP

10
Traditional SIMD Power Breakdown

• Register File Power consumes a lot of power in
  traditional 32-wide SIMD architecture

11
Register File Access
Lots of power wasted on unneeded register file
access!

• Many of the register file access do not have to
  go back to the main register file

12
Instruction Pair Frequency
Like the Multiply-Accumulate (MAC) instruction
there is opportunity to fuse other
instructions A few instruction pairs (3-5) make
up the majority of all instruction pairs!
13
Data Alignment Problem!
Intra-Prediction
Traditional SIMD machines take too long or cost
too much to do this Good news small fixed
number patterns per kernel

• H.264 Intra-prediction has 9 different prediction
  modes
• Each prediction mode requires a specific
  permutation

14
Summary

• Conclusion about 4G and H.264
• Lots of different sized parallelism
• From 4 wide to 96 wide to 1024 wide SIMD
• Which means many different SIMD widths need to be
  supported
• Very short lived values
• Lots of potential for instruction fusings
• Limited set of shuffle patterns required for each
  kernel

15
AnySP Design
16
Traditional SIMD Architectures
32-Wide SIMD with Simple Shuffle Network
17
AnySP Architecture High Level
16 Banked Memory with SRAM-based Crossbar
8 Groups of 8-Wide Flexible Function Units
Multiple Output Adder Tree
128x128 16bit Swizzle Network
Temporary Buffer and Bypass Network
Datapath AGU and Scalar Pipeline
18
Multi-Width Support
19
AnySP FFU Datapath

• Flexible Functional Unit allows us to
• Exploit Pipeline-parallelism by joining two lanes
  together
• Handle register bypass and the temporary buffer
• Join multiple pipelines to process deeper
  subgraphs
• Fuse Instruction Pairs

20
AnySP Results
21
Simulation Environment

• Traditional SIMD architecture comparison
• SODA at 90nm technology
• AnySP
• Synthesized at 90nm TSMC
• Power, timing, area numbers were extracted
• Performance and Power for each kernel was
  generated using synthesized data on in-house
  simulator
• 4G based on a NTT DoCoMo 4G test setup
• H.264 4CIF_at_30fps

22
AnySP Speedup vs SIMD-based Architecture

• For all benchmarks we perform more than 2x better
  than a SIMD-based architecture

23
AnySP Energy-Delay vs SIMD-based Architecture

• More importantly energy efficiency is much
  better!

24
AnySP Power Breakdown

• We estimate that both H.264 and 4G wireless can
  be done in under 1 Watt at 45nm

25
Conclusion Future Work

• Conclusion
• We have presented an example architecture that
  could possibly meet the requirements of 100Mbps
  4G and HD video on the same platform
• Under the power budget and meeting the
  performance at 45nm
• Future and Ongoing Work
• Application-specific language
• Larger class of algorithms for AnySP
• Better utilization of resources for non-parallel
  kernels
• Speedup sequential parts

26
The End