A full spectrum of computing-in-memory technologies

Z Sun, S Kvatinsky, X Si, A Mehonic, Y Cai… - Nature Electronics, 2023‏ - nature.com
Computing in memory (CIM) could be used to overcome the von Neumann bottleneck and to
provide sustainable improvements in computing throughput and energy efficiency …

[HTML][HTML] A Comprehensive Review of Processing-in-Memory Architectures for Deep Neural Networks

R Kaur, A Asad, F Mohammadi - Computers, 2024‏ - mdpi.com
This comprehensive review explores the advancements in processing-in-memory (PIM)
techniques and chiplet-based architectures for deep neural networks (DNNs). It addresses …

Neupims: Npu-pim heterogeneous acceleration for batched llm inferencing

G Heo, S Lee, J Cho, H Choi, S Lee, H Ham… - Proceedings of the 29th …, 2024‏ - dl.acm.org
Modern transformer-based Large Language Models (LLMs) are constructed with a series of
decoder blocks. Each block comprises three key components:(1) QKV generation,(2) multi …

Functionally-complete boolean logic in real dram chips: Experimental characterization and analysis

İE Yüksel, YC Tuğrul, A Olgun… - … Symposium on High …, 2024‏ - ieeexplore.ieee.org
Processing-using-DRAM (PuD) is an emerging paradigm that leverages the analog
operational properties of DRAM circuitry to enable massively parallel in-DRAM computation …

MIMDRAM: An end-to-end processing-using-DRAM system for high-throughput, energy-efficient and programmer-transparent multiple-instruction multiple-data …

GF Oliveira, A Olgun, AG Yağlıkçı… - … Symposium on High …, 2024‏ - ieeexplore.ieee.org
Processing-using-DRAM (PUD) is a processing-in-memory (PIM) approach that uses a
DRAM array's massive internal parallelism to execute very-wide (eg, 16,384-262,144-bit …

Pathfinding future pim architectures by demystifying a commercial pim technology

B Hyun, T Kim, D Lee, M Rhu - 2024 IEEE International …, 2024‏ - ieeexplore.ieee.org
Processing-in-memory (PIM) has been explored for decades by computer architects, yet it
has never seen the light of day in real-world products due to its high design overheads and …

pLUTo: Enabling massively parallel computation in DRAM via lookup tables

JD Ferreira, G Falcao, J Gómez-Luna… - 2022 55th IEEE/ACM …, 2022‏ - ieeexplore.ieee.org
Data movement between the main memory and the processor is a key contributor to
execution time and energy consumption in memory-intensive applications. This data …

Swiftrl: Towards efficient reinforcement learning on real processing-in-memory systems

K Gogineni, SS Dayapule… - … Analysis of Systems …, 2024‏ - ieeexplore.ieee.org
Reinforcement Learning (RL) is the process by which an agent learns optimal behavior
through interactions with experience datasets, all of which aim to maximize the reward …

Evaluating homomorphic operations on a real-world processing-in-memory system

H Gupta, M Kabra, J Gómez-Luna… - 2023 IEEE …, 2023‏ - ieeexplore.ieee.org
Computing on encrypted data is a promising approach to reduce data security and privacy
risks, with homomorphic encryption serving as a facilitator in achieving this goal. In this work …

PULSAR: Simultaneous many-row activation for reliable and high-performance computing in off-the-shelf DRAM chips

IE Yuksel, YC Tugrul, F Bostanci, AG Yaglikci… - arxiv preprint arxiv …, 2023‏ - arxiv.org
Data movement between the processor and the main memory is a first-order obstacle
against improving performance and energy efficiency in modern systems. To address this …