Dyamond: Compact and Efficient 1T1C DRAM IMC Accelerator With Bit Column Addition for Memory-Intensive AI
- 13 February 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 60 (4) , 1299-1310
Abstract
This article proposes Dyamond, a one transistor, one capacitor (1T1C) dynamic random access memory (DRAM) in-memory computing (IMC) accelerator with architecture-to-circuit-level optimizations for high memory density and energy efficiency. The bit column addition (BCA) dataflow introduces output bit-wise accumulation to exploit varying accuracy and energy characteristics across different bit positions. The lower BCA (LBCA) reduces analog-to-digital converter (ADC) operations to enhance energy efficiency with inter-column analog accumulation. The higher BCA (HBCA) improves accuracy through signal enhancement and minimizes energy consumption per ADC readout with signal shift (SS). The design maximizes memory density by dedicating 1T1C cells solely to memory and integrating a compact computation circuit adjacent to the bitline sense amplifier. The memory access power is further reduced with a big-little array structure and a switchable sense amplifier (SWSA), which trades off retention time and energy consumption. Fabricated in 28-nm CMOS, Dyamond integrates 3.54-MB DRAM in a 6.48-mm2 area, achieving 27.2 TOPS/W peak efficiency and outstanding performance in advanced models such as BERT and GPT-2.Keywords
Funding Information
- Institute of Information and Communications Technology Planning and Evaluation (IITP) through the Artificial Intelligence Semiconductor Support Program to Nurture the Best Talents Grant
- Korean Government [Ministry of Science and ICT (MSIT)] (IITP-2024-RS-2023-00256472)
This publication has 19 references indexed in Scilit:
- 34.5 A 818-4094TOPS/W Capacitor-Reconfigured CIM Macro for Unified Acceleration of CNNs and Transformers2024
- DynaPlasia: An eDRAM In-Memory Computing-Based Reconfigurable Spatial Accelerator With Triple-Mode CellIEEE Journal of Solid-State Circuits, 2023
- 16.5 DynaPlasia: An eDRAM In-Memory-Computing-Based Reconfigurable Spatial Accelerator with Triple-Mode Cell for Dynamic Resource Switching2023
- A 1ynm 1.25V 8Gb 16Gb/s/Pin GDDR6-Based Accelerator-in-Memory Supporting 1TFLOPS MAC Operation and Various Activation Functions for Deep Learning ApplicationIEEE Journal of Solid-State Circuits, 2022
- A 1ynm 1.25V 8Gb, 16Gb/s/pin GDDR6-based Accelerator-in-Memory supporting 1TFLOPS MAC Operation and Various Activation Functions for Deep-Learning Applications2022
- 25.4 A 20nm 6GB Function-In-Memory DRAM, Based on HBM2 with a 1.2TFLOPS Programmable Computing Unit Using Bank-Level Parallelism, for Machine Learning Applications2021
- 16.2 eDRAM-CIM: Compute-In-Memory Design with Reconfigurable Embedded-Dynamic-Memory Array Realizing Adaptive Data Converters and Charge-Domain Computing2021
- A Programmable Heterogeneous Microprocessor Based on Bit-Scalable In-Memory ComputingIEEE Journal of Solid-State Circuits, 2020
- ELP2IM: Efficient and Low Power Bitwise Operation Processing in DRAM2020
- Deep Residual Learning for Image Recognition2016