APIM: An Antiferromagnetic MRAM-Based Processing-In-Memory System for Efficient Bit-Level Operations of Quantized Convolutional Neural Networks
- 1 March 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 43 (8) , 2405-2410
Abstract
Quantized Convolutional Neural Network (QCNN) is an attractive approach that reduces hardware overheads, especially for energy-constrained systems. However, existing QCNNs still require non-trivial hardware resources and memory capacity in order not to compromise model accuracy. To address this issue, we propose an antiferromagnetic magnetic random-access memory (ARAM)-based processing-in-memory (PIM) system, leveraging bit-level sparsity. Three optimization techniques are proposed to optimize hardware resource utilization while preserving CNN accuracy. Firstly, the ARAM-based memory subsystem allows dynamic adaptation of variable bit-width across CNN layers. Secondly, the bit-level accelerator employs the bit-fusion format engineered for processing data from the ARAM subsystem. Thirdly, a customized data path within the RISC-V core guarantees efficient instruction processing to the ARAM-based memory subsystem and bit-level accelerator, enabling optimal bit-level data transmission and computation. Experimental results demonstrate that this design remarkably reduces data movement by 50%-83% across existing CNNs. Compared to state-of-the-art designs, it enhances throughput and latency by an average of 5x and 10x, respectively. In addition, this design achieves speedups between 1.63x and 2.96x, outstripping other designs in AlexNet, VGG16, and ResNet18 benchmarks.Keywords
Funding Information
- National Natural Science Foundation of China (52121001, 61971024, 62004011, 62122008, 92164206)
- Tencent Foundation through the XPLORER PRIZE
- National Key Research and Development Program of China (2022YFb300200)
- Project Funded by China Postdoctoral Science Foundation (2022M720344)
This publication has 14 references indexed in Scilit:
- IMGA: Efficient In-Memory Graph Convolution Network Aggregation With Data Flow OptimizationsIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2023
- Electrical manipulation and detection of antiferromagnetism in magnetic tunnel junctionsNature Electronics, 2023
- AppCiP: Energy-Efficient Approximate Convolution-in-Pixel Scheme for Neural Network AccelerationIEEE Journal on Emerging and Selected Topics in Circuits and Systems, 2023
- A Survey of MRAM-Centric Computing: From Near Memory to In MemoryIEEE Transactions on Emerging Topics in Computing, 2022
- Structured Term Pruning for Computational Efficient Neural Networks InferenceIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2022
- Accelerating Graph-Connected Component Computation With Emerging Processing-In-Memory ArchitectureIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2022
- Optimizing Temporal Convolutional Network Inference on FPGA-Based AcceleratorsIEEE Journal on Emerging and Selected Topics in Circuits and Systems, 2020
- FPGA-based High-performance CNN Accelerator Architecture with High DSP Utilization and Efficient Scheduling Mode2020
- Design and Implementation of a Low-Power, Embedded CNN Accelerator on a Low-end FPGA2019
- Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference2018