Contribution to a conference proceedings/Contribution to a book

FZJ-2026-00229

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png Content Addressable Memory Hierarchies for Computing in Memory

Manea, P.-P. (Corresponding author)FZJ* ; Leroux, N.FZJ* ; Strachan, J. P.FZJ*

2025
IEEE

This record in other databases:  

Please use a persistent id in citations: doi:10.1109/EDTM61175.2025.11040774

Abstract: Content Addressable Memories (CAMs) match inputdata to stored content, ideal for search-intensive taskslike network address lookup and similarity matching.This paper explores CAM hierarchies (similar to thestandard RAM hierarchy), focusing on two implemen-tation layers: non-volatile, analog Content AddressableMemory (aCAM) for static data patterns, and volatile,gain cell-based aCAM for dynamic tasks. Mitigatingkey non-idealities in memristor-based aCAMs increasesstorage from 2 to 4 bits per cell, enhancing performance,capacity, and energy efficiency for Compute in Memory(CIM) applications.

---

Contributing Institute(s):

1. Neuromorphic Compute Nodes (PGI-14)
2. Neuromorphic Software Eco System (PGI-15)

Research Program(s):

1. 5234 - Emerging NC Architectures (POF4-523) (POF4-523)

---