A Survey on the Expanding Scope and Interdisciplinary Opportunities for Processing-in-Memory Techniques

Jan 1, 2026 by Kazi Asifuzzaman, Yuan He, Tianyun Zhang, Eric Tang, Narasinga Rao Miniskar, Keita Teranishi, Jeffrey S. Vetter (IEEE Access)

DOI 10.1109/ACCESS.2026.3659051

We mapped PIM’s messy, exciting landscape — from DRAM and NVM integration to analog compute bricks — and show how those diverse primitives are already reshaping workloads from generative AI to genome analysis while calling out the messy real-world barriers (manufacturing, power, thermal, software and consistency) that still block widespread adoption. This survey is for people who want a single, structured read on where PIM actually works, where it’s promising, and what problems the community still needs to solve to make it mainstream.

source S2, crossref