Heterogeneous Memory Design Exploration for AI Accelerators with a Gain Cell Memory Compiler

Xinxin Wang; Lixian Yan; Shuhan Liu; Luke Upton; Zhuoqi Cai; Yiming Tan; Shengman Li; Koustav Jana; Peijing Li; Jesse Cirimelli-Low; Thierry Tambe; Matthew Guthaus; H. -S. Philip Wong

Heterogeneous Memory Design Exploration for AI Accelerators with a Gain Cell Memory Compiler

Hardware Architecture 2026-02-26 v1 Systems and Control Systems and Control

Authors: Xinxin Wang , Lixian Yan , Shuhan Liu , Luke Upton , Zhuoqi Cai , Yiming Tan , Shengman Li , Koustav Jana , Peijing Li , Jesse Cirimelli-Low , Thierry Tambe , Matthew Guthaus , H. -S. Philip Wong

View on arXiv ↗ PDF ↗

Abstract

As memory increasingly dominates system cost and energy, heterogeneous on-chip memory systems that combine technologies with complementary characteristics are becoming essential. Gain Cell RAM (GCRAM) offers higher density, lower power, and tunable retention, expanding the design space beyond conventional SRAM. To this end, we create an OpenGCRAM compiler supporting both SRAM and GCRAM. It generates macro-level designs and layouts for commercial CMOS processes and characterizes area, delay, and power across user-defined configurations. The tool enables systematic identification of optimal heterogeneous memory configurations for AI tasks under specified performance metrics.

Keywords

processing-in-memory memory hierarchy computer architecture

Cite

@article{arxiv.2602.21278,
  title  = {Heterogeneous Memory Design Exploration for AI Accelerators with a Gain Cell Memory Compiler},
  author = {Xinxin Wang and Lixian Yan and Shuhan Liu and Luke Upton and Zhuoqi Cai and Yiming Tan and Shengman Li and Koustav Jana and Peijing Li and Jesse Cirimelli-Low and Thierry Tambe and Matthew Guthaus and H. -S. Philip Wong},
  journal= {arXiv preprint arXiv:2602.21278},
  year   = {2026}
}

Heterogeneous Memory Design Exploration for AI Accelerators with a Gain Cell Memory Compiler

Abstract

Keywords

Cite

Related papers