LOG-CIM: An Energy-Efficient Logarithmic Quantization Computing-In-Memory Processor With Exponential Parallel Data Mapping and Zero-Aware 6T Dual-WL Cell
- 30 May 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 59 (10) , 3330-3341
Abstract
This article introduces the logarithmic quantization computing-in-memory processor (LOG-CIM), an advanced processor that extends the capabilities of conventional linear quantization (INT) by integrating logarithmic quantization (LOGQ) into computing-in-memory (CIM) architectures. Emphasizing three key features, the LOG-CIM significantly achieves high energy efficiency and memory density using LOGQ: 1) exponential parallel data mapping (EPDM) not only facilitates efficient shifted alignment and data accumulation, optimizing the processing pipeline for LOGQ complex computations but also achieves a 51.5% improvement in energy efficiency as a result; 2) the implementation of a zero-aware dual-word-line (ZADU) cell, which not only enhances parallelism but also reduces computation power by lowering read power 88.9% for 89.3% bit sparsity; and 3) shift-addition peripheral (SAP) logic, which maximizes throughput by performing the addition of different bit positions from two data read by the ZADU cell before they enter the adder tree, thereby enhancing 24.3% energy efficiency. The LOG-CIM is fabricated in 28-nm CMOS technology and occupies a 5.832 mm 2 with a total of 48-kB on-chip CIMs. It achieves 192 TOPS/W using conventional data mapping at 4-b INT weight and 4-b input activation and 116.4 TOPS/W using EPDM at 4-b LOGQ weight and 8-b INT input activation at 120-MHz clock and 0.85-V supply voltage with the ResNet-50 @ ImageNet dataset. When using the 4-b LOGQ weight for inference, the energy efficiency was 3.54 times higher than the 8-b INT weight with similar accuracy. During training, using 4-b LOGQ weight to reach the same accuracy increases epoch compared with using only 8-b INT weight, but consequentially saves 68.7% in relative total energy. Furthermore, it achieves 599-kb/mm 2 memory density.Keywords
Funding Information
- Samsung Electronics Company Ltd (IO201207-07799-01)
- Institute of Information and Communications Technology Planning and Evaluation (IITP) through the Artificial Intelligence Semiconductor
- Nurture the Best Talents Grant funded by the Korean Government (IITP-2024-RS-2023-00256472)
This publication has 13 references indexed in Scilit:
- LOG-CIM: A 116.4 TOPS/W Digital Computing-In-Memory Processor Supporting a Wide Range of Logarithmic Quantization with Zero-Aware 6T Dual-WL Cell2023
- DynaPlasia: An eDRAM In-Memory Computing-Based Reconfigurable Spatial Accelerator With Triple-Mode CellIEEE Journal of Solid-State Circuits, 2023
- Neuro-CIM: ADC-Less Neuromorphic Computing-in-Memory Processor With Operation Gating/Stopping and Digital–Analog NetworksIEEE Journal of Solid-State Circuits, 2023
- ReDCIM: Reconfigurable Digital Computing- In -Memory Processor With Unified FP/INT Pipeline for Cloud AI AccelerationIEEE Journal of Solid-State Circuits, 2022
- A 161.6 TOPS/W Mixed-mode Computing-in-Memory Processor for Energy-Efficient Mixed-Precision Deep Neural Networks2022
- DeepShift: Towards Multiplication-Less Neural Networks2021
- A 43.1TOPS/W Energy-Efficient Absolute-Difference-Accumulation Operation Computing-In-Memory With Computation ReuseIEEE Transactions on Circuits and Systems II: Express Briefs, 2021
- Cascade Current Mirror to Improve Linearity and Consistency in SRAM In-Memory ComputingIEEE Journal of Solid-State Circuits, 2021
- A 65nm 4Kb algorithm-dependent computing-in-memory SRAM unit-macro with 2.3ns and 55.8TOPS/W fully parallel product-sum operation for binary DNN edge processors2018
- 17.3 A 28nm 256kb 6T-SRAM with 280mV improvement in VMIN using a dual-split-control assist scheme2015