Approximate Computing on FPGA using Neural Acceleration

13 slides
2.33 MB

Similar Presentations

Presentation Transcript


Approximate Computing on FPGA using Neural AccelerationPresented By: Mikkel Nielsen, Nirvedh Meshram, Shashank Gupta, Kenneth Siu


Approximate ComputingInvolves computations that do not need to be exact (tolerance to quality degradation) Neural Network’s (NN) speed can be exploited Optimization (performance and energy efficiency) in favor of accuracy Implement a NN accelerator that interacts with the CPU Useful in many computer vision and image processing applications like edge detection


Motivation To combine approaches of specialized logic (accelerator) and approximate computing for enhanced performance and energy efficiencyTop Level System Design


Architecture Design of NPUTop Level Diagram of NPU


Architecture and FeaturesTotal of 8 Processing Elements in one Processing Unit (in initial design) Weights needed for Neural Processing loaded into the weight FIFO at time of configuration A Scheduling Buffer is configured in configuration phase and use to generate control signals used for Input, Output, Sigmoid and Accumulator FIFO, PE input selection and Sigmoid Function After this Inputs are loaded into input FIFO (using enqd instruction) Inputs & Weights are 16 bit wide Fixed with 7 fractional bits. NPU supports 32 bit integers and single precision Floating Points. Input interface does required format conversion


Architecture and FeaturesCompute Unit: Performs Multiplication and Addition operation State Machine: Controls &configures NPU – stall due to insufficient input/Push output to FIFO Accumulator FIFO: Stores intermediate results when No of Inputs > No of PE Sigmoid Function Unit: Current NPU supports tan sigmoid and linear functions Output FIFO: Holds output of NPU


Software for ConfigurationWeights can be generated through custom MATLAB code or through a compiler A perl based compiler which expects weights and the structure of the neural network as input The compiler will then generate a sequence of instructions which will be loaded into the NPU These instructions will load values in the weight buffers as well as the scheduling buffer


Zedboard ImplementationUsed Vivado tools to set programmable logic and generate a bitstream for gates Implements bitstream as a First-stage boot loader by wrapping bitstream with boot files On Zedboard boot, programmable logic is loaded with design Driver interfaces C code with programmable logic(NPU) Comparison between C code runs native on Digilent Linux on Zedboard to test ARM core


Zedboard ChallengesConfiguring Vivado to generate bitstream. Synthesis/Implementation Debugging/Errors Creating appropriate wrapper so Zedboard does not crash on boot


BenchmarksSobel Edge Detection Good program for approximate computing Uses convolution of a 3x3 matrix to find edges Took 0.4 ms for 512x512 image


AxBench BenchmarksUsing AxBench Utilizes software NN (FANN Library) Need a hardware NN to fully utilize efficiency Benchmarks run both with and without NN


In ProgressCompare the performance against another Processing Unit with 16 PEs and check speedup gains Build an NPU with 2 Processing units with 8 PEs each and again compare the performance & speedup Modify the scheduler to remove stalls due to unavailable data More benchmarks


References[1] H. Esmaeilzadeh, A. Sampson, L. Ceze, and D. Burger. Neural acceleration for general- purpose approximate programs. MICRO, 2012. [2] D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning internal representations by error propagation,” in Parallel Distributed Processing: Explorations in the Microstructure of Cognition. MIT Press, 1986,vol.1, pp. 318–362. [3] Marc de Kruijf and K. Sankaralingam, “Exploring the synergy of emerging workloads and silicon reliability trends” in SELSE, 2009. [4] Thierry Moreau, Mark Wyse, Jacob Nelson, Adrian Sampson, Hadi Esmaeilzadeh, Luis Ceze, Mark Oskin, SNNAP: Approximate computing on programmable SoCs via neural acceleration. HPCA 2015: 603-14.

Browse More Presentations

Last Updated: 8th March 2018