Intel Nervana Neural Network Processor-T (NNP-T) Fused Floating Point Many-Term Dot Product
Authors:
- 1 June 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 133-136
Abstract
Intel’s Nervana Neural Network Processor for Training (NNP-T) contains at its core an advanced floating point dot product design to accelerate the matrix multiplication operations found in many AI applications. Each Matrix Processing Unit (MPU) on the Intel NNP-T can process a 32x32 BFloat16 matrix multiplication every 32 cycles, accumulating the result in single precision (FP32). To reduce hardware costs, the MPU uses a fused many-term floating point dot product design with block alignment of the many input terms during addition, resulting in a unique datapath with several interesting design trade-offs. In this paper, we describe the details of the MPU pipeline, discuss the trade-offs made in the design, and present information on the accuracy of the computation as compared to traditional FMA implementations.Keywords
This publication has 4 references indexed in Scilit:
- Experimental Analysis of Matrix Multiplication Functional Units2019
- Optimized Fused Floating-Point Many-Term Dot-Product Hardware for Machine Learning Accelerators2019
- Flexpoint: Predictive Numerics for Deep Learning2018
- A Fused Floating-Point Four-Term Dot Product UnitIEEE Transactions on Circuits and Systems I: Regular Papers, 2016