Computational RAM: implementing processors in memory

TLDR

Computational RAM is a processor‑in‑memory architecture that efficiently utilizes memory bandwidth by aligning simple processing elements with memory columns, can function as either a conventional memory chip or a SIMD computer, and requires more than merely attaching logic to existing designs. The paper investigates how to integrate processing power with memory by designing an architecture that preserves and exploits memory features. The authors design an architecture that preserves and exploits memory features to integrate processing power with memory. When used as memory, computational RAM matches conventional DRAM in access time, packaging, and cost.

Abstract

Computational RAM is a processor-in-memory architecture that makes highly effective use of internal memory bandwidth by pitch-matching simple processing elements to memory columns. Computational RAM can function either as a conventional memory chip or as a SIMD (single-instruction stream, multiple-data stream) computer. When used as a memory, computational RAM is competitive with conventional DRAM in terms of access time, packaging and cost. Adding logic to memory is not a simple question of bolting together two existing designs. The paper considers how computational RAM integrates processing power with memory by using an architecture that preserves and exploits the features of memory.