SSA-over-array (SSoA): A stacked DRAM architecture for near-memory computing
Abstract Aiming to enhance the bandwidth in near-memory computing, this paper proposes a SSA-over-array (SSoA) architecture. By relocating the secondary sense amplifier (SSA) from dynamic random access memory (DRAM) to the logic die and repositioning the DRAM-to-logic stacking interface closer to the DRAM core, the SSoA overcomes the layout and area limitations of SSA and master DQ (MDQ), leading to improvements in DRAM data-width density and frequency, significantly enhancing bandwidth density. The quantitative evaluation results show a 70.18 times improvement in bandwidth per unit area over the baseline, with a maximum bandwidth of 168.296 Tbps/Gb. We believe the SSoA is poised to redefine near-memory computing development strategies.
Authors
- Wenwu Xiao (ORCID: https://orcid.org/0000-0003-0387-7206)
- Fujun Bai (ORCID: https://orcid.org/0000-0002-7971-5986)
- Xiping Jiang (ORCID: https://orcid.org/0009-0004-0269-8753)
- Fengguo Zuo
- Jianguo Yang (ORCID: https://orcid.org/0000-0002-3387-1238)
- Yixin Guo (ORCID: https://orcid.org/0000-0001-8887-595X)
- Ming Liu (ORCID: https://orcid.org/0000-0002-0937-7547)
- Song Wang (ORCID: https://orcid.org/0000-0002-7682-5987)
- Yubing Wang (ORCID: https://orcid.org/0000-0003-2444-1030)
Institutions
- University of Science and Technology of China (CN)
- Chinese Academy of Sciences (CN)
- Institute of Microelectronics (CN)
- Xi'an UniIC Semiconductors (China) (CN)
- University of Chinese Academy of Sciences (CN)
Publication Details
- Journal
- Journal of Semiconductors
- Published
- 2024-10-01
- DOI
- https://doi.org/10.1088/1674-4926/24050004
- Citations
- 2
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.42