Litcius/Paper detail

BitSystolic: A 26.7 TOPS/W 2b~8b NPU With Configurable Data Flows for Edge Devices

Qing Yang, Hai Li

2020IEEE Transactions on Circuits and Systems I Regular Papers25 citationsDOI

Abstract

Efficient deployment of deep neural networks (DNNs) emerges with the exploding demand for artificial intelligence on edge devices. Mixed-precision inference with both compressed model and reduced computation cost enlightens a way for accurate and efficient DNN deployments. Despite obtaining mixed-precision DNN models at the algorithmic level, there still lacks sufficient hardware support. In this work, we propose BitSystolic, a neural processing unit based on a systolic array structure. In BitSystolic, the numerical precision of both weights and activations can be configured in the range of 2 - 8 bits, fulfilling different requirements across mixed-precision models and tasks. Moreover, BitSystolic can support various data flows presented in different types of neural layers (e.g., convolution, fully-connected, and recurrent neural layers) and adaptive optimization of data reuse by switching between the matrix-matrix mode and vector-matrix mode. We designed and fabricated the proposed BitSystolic composed of a $16\times 16$ systolic array. Our measurement results show that BitSystolic features the unified power efficiency of up to 26.7 TOPS/W with 17.8 mW peak power consumption across various layer types.

Topics & Concepts

Computer scienceArtificial neural networkEdge deviceConvolution (computer science)ComputationMatrix (chemical analysis)Matrix multiplicationPower (physics)TOPSRange (aeronautics)Enhanced Data Rates for GSM EvolutionEdge computingAlgorithmComputer engineeringComputational scienceComputer hardwareArtificial intelligenceEngineeringCloud computingMaterials scienceQuantumComposite materialMechanical engineeringOperating systemQuantum mechanicsSpinningAerospace engineeringPhysicsAdvanced Neural Network ApplicationsAdvanced Memory and Neural ComputingFerroelectric and Negative Capacitance Devices
BitSystolic: A 26.7 TOPS/W 2b~8b NPU With Configurable Data Flows for Edge Devices | Litcius