Related papers: MultPIM: Fast Stateful Multiplication for Processi…

Matrix-matrix multiplication is a fundamental operation of great importance to scientific computing and, increasingly, machine learning. It is a simple enough concept to be introduced in a typical high school algebra course yet in practice…

Single instruction, multiple data (SIMD) is a popular design style of in-memory computing (IMC) architectures, which enables memory arrays to perform logic operations to achieve low energy consumption and high parallelism. To implement a…

In this paper, we propose PIM-LLM, a hybrid architecture developed to accelerate 1-bit large language models (LLMs). PIM-LLM leverages analog processing-in-memory (PIM) architectures and digital systolic arrays to accelerate low-precision…

Memristive in-memory sorting has been proposed recently to improve hardware sorting efficiency. Using iterative in-memory min computations, data movements between memory and external processing units can be eliminated for improved latency…

In this paper, we present FLiMS, a highly-efficient and simple parallel algorithm for merging two sorted lists residing in banked and/or wide memory. On FPGAs, its implementation uses fewer hardware resources than the state-of-the-art…

Our ISCA 2015 paper provides a new programmable processing-in-memory (PIM) architecture and system design that can accelerate key data-intensive applications, with a focus on graph processing workloads. Our major idea was to completely…

Spiking Neural Networks (SNNs) have emerged as a biologically inspired alternative to conventional deep networks, offering event-driven and energy-efficient computation. However, their throughput remains constrained by the serial update of…

Processing-using-DRAM (PUD) is a processing-in-memory (PIM) approach that uses a DRAM array's massive internal parallelism to execute very-wide data-parallel operations, in a single-instruction multiple-data (SIMD) fashion. However, DRAM…

Compute-in-memory (CIM) accelerators for spiking neural networks (SNNs) are promising solutions to enable $\mu$s-level inference latency and ultra-low energy in edge vision applications. Yet, their current lack of flexibility at both the…

Modern data-intensive applications demand memory solutions that deliver high-density, low-power, and integrated computational capabilities to reduce data movement overhead. This paper presents the use of Gain-Cell embedded DRAM (GC-eDRAM) -…

Transformers have emerged as a powerful tool for natural language processing (NLP) and computer vision. Through the attention mechanism, these models have exhibited remarkable performance gains when compared to conventional approaches like…

In-memory computing (IMC) can eliminate the data movement between processor and memory which is a barrier to the energy-efficiency and performance in Von-Neumann computing. Resistive RAM (RRAM) is one of the promising devices for IMC…

Memristors have been suggested as a novel route to neuromorphic computing based on the similarity between neurons (synapses and ion pumps) and memristors. The D.C. action of the memristor is a current spike, which we think will be fruitful…

Stochastic computing (SC) offers hardware simplicity but suffers from low throughput, while high-throughput Digital Computing-in-Memory (DCIM) is bottlenecked by costly adder logic for matrix-vector multiplication (MVM). To address this…

Triangle counting (TC) is a fundamental problem in graph analysis and has found numerous applications, which motivates many TC acceleration solutions in the traditional computing platforms like GPU and FPGA. However, these approaches suffer…

Emerging machine learning (ML) models (e.g., transformers) involve memory pin bandwidth-bound matrix-vector (MV) computation in inference. By avoiding pin crossings, processing in memory (PIM) can improve performance and energy for…

With the rapid growth of deep neural networks (DNNs), compute-in-memory (CIM) has emerged as a promising energy-efficient paradigm for accelerating multiply-and-accumulate (MAC) operations. Yet, current CIM architectures are largely limited…

With the rapid advent of generative models, efficiently deploying these models on specialized hardware has become critical. Tensor Processing Units (TPUs) are designed to accelerate AI workloads, but their high power consumption…

This paper evaluates the efficacy of recent commercial processing-in-memory (PIM) solutions to accelerate fast Fourier transform (FFT), an important primitive across several domains. Specifically, we observe that efficient implementations…

DRAM scaling has exacerbated the RowHammer vulnerability. To counter this, JEDEC recently introduced Per Row Activation Counting (PRAC) with the Alert Back-Off protocol as an optional DDR5 feature. While promising, PRAC requires per-row…