English
Related papers

Related papers: Enabling and Exploiting Partition-Level Parallelis…

200 papers

Efficient parallelism is necessary for achieving low-latency, high-throughput inference with large language models (LLMs). Tensor parallelism (TP) is the state-of-the-art method for reducing LLM response latency, however GPU communications…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-01-27 Mert Hidayetoglu , Aurick Qiao , Michael Wyatt , Jeff Rasley , Yuxiong He , Samyam Rajbhandari

Memory-centric computing aims to enable computation capability in and near all places where data is generated and stored. As such, it can greatly reduce the large negative performance and energy impact of data access and data movement, by…

Hardware Architecture · Computer Science 2024-12-30 Onur Mutlu , Ataberk Olgun , Geraldo F. Oliveira , Ismail Emir Yuksel

Memory-centric computing aims to enable computation capability in and near all places where data is generated and stored. As such, it can greatly reduce the large negative performance and energy impact of data access and data movement, by…

Hardware Architecture · Computer Science 2023-09-15 Onur Mutlu

The increasing prevalence and growing size of data in modern applications have led to high costs for computation in traditional processor-centric computing systems. Moving large volumes of data between memory devices (e.g., DRAM) and…

Hardware Architecture · Computer Science 2022-06-01 Geraldo F. Oliveira , Juan Gómez-Luna , Saugata Ghose , Onur Mutlu

Massive off-chip accesses in GPUs are the main performance bottleneck, and we divided these accesses into three types: (1) Write, (2) Data-Read, and (3) Read-Only. Besides, We find that many writes are duplicate, and the duplication can be…

Hardware Architecture · Computer Science 2024-08-20 Wei Zhao , Dan Feng , Wei Tong , Xueliang Wei , Bing Wu

Processing in memory (PIM) moves computation into memories with the goal of improving throughput and energy-efficiency compared to traditional von Neumann-based architectures. Most existing PIM architectures are either general-purpose but…

Hardware Architecture · Computer Science 2019-07-23 Oscar Castañeda , Maria Bobbett , Alexandra Gallyas-Sanhueza , Christoph Studer

Processing-in-memory (PIM) has emerged as a promising solution for accelerating memory-intensive workloads as they provide high memory bandwidth to the processing units. This approach has drawn attention not only from the academic community…

Hardware Architecture · Computer Science 2024-09-11 Dongjae Lee , Bongjoon Hyun , Taehun Kim , Minsoo Rhu

Several manufacturers have already started to commercialize near-bank Processing-In-Memory (PIM) architectures. Near-bank PIM architectures place simple cores close to DRAM banks and can yield significant performance and energy improvements…

Hardware Architecture · Computer Science 2022-05-24 Christina Giannoula , Ivan Fernandez , Juan Gómez-Luna , Nectarios Koziris , Georgios Goumas , Onur Mutlu

This paper summarizes the idea of Adaptive-Latency DRAM (AL-DRAM), which was published in HPCA 2015, and examines the work's significance and future potential. AL-DRAM is a mechanism that optimizes DRAM latency based on the DRAM module and…

Hardware Architecture · Computer Science 2018-05-09 Donghyuk Lee , Yoongu Kim , Gennady Pekhimenko , Samira Khan , Vivek Seshadri , Kevin Chang , Onur Mutlu

Pipeline parallelism (PP) is widely used for training large language models (LLMs), yet its scalability is often constrained by high activation memory consumption as the number of in-flight microbatches grows with the degree of PP. In this…

Machine Learning · Computer Science 2025-07-01 Xinyi Wan , Penghui Qi , Guangxing Huang , Min Lin , Jialin Li

The continued growth of the computational capability of throughput processors has made throughput processors the platform of choice for a wide variety of high performance computing applications. Graphics Processing Units (GPUs) are a prime…

Hardware Architecture · Computer Science 2018-05-01 Rachata Ausavarungnirun

Deep learning (DL) models are piquing high interest and scaling at an unprecedented rate. To this end, a handful of tiled accelerators have been proposed to support such large-scale training tasks. However, these accelerators often…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-06-07 Jiahao Fang , Huizheng Wang , Qize Yang , Dehao Kong , Xu Dai , Jinyi Deng , Yang Hu , Shouyi Yin

As the size of artificial intelligence and machine learning (AI/ML) models and datasets grows, the memory bandwidth becomes a critical bottleneck. The paper presents a novel extended memory hierarchy that addresses some major memory…

Hardware Architecture · Computer Science 2025-05-20 Jordi Altayo , Paul Delestrac , David Novo , Simey Yang , Debjyoti Bhattacharjee , Francky Catthoor

Many modern workloads such as neural network inference and graph processing are fundamentally memory-bound. For such workloads, data movement between memory and CPU cores imposes a significant overhead in terms of both latency and energy. A…

Hardware Architecture · Computer Science 2023-04-04 Juan Gómez-Luna , Izzat El Hajj , Ivan Fernandez , Christina Giannoula , Geraldo F. Oliveira , Onur Mutlu

Massive multi-threading in GPU imposes tremendous pressure on memory subsystems. Due to rapid growth in thread-level parallelism of GPU and slowly improved peak memory bandwidth, the memory becomes a bottleneck of GPU's performance and…

Hardware Architecture · Computer Science 2019-06-17 Bing Li , Mengjie Mao , Xiaoxiao Liu , Tao Liu , Zihao Liu , Wujie Wen , Yiran Chen , Hai , Li

Modern commercial-off-the-shelf (COTS) multicore processors have advanced memory hierarchies that enhance memory-level parallelism (MLP), which is crucial for high performance. To support high MLP, shared last-level caches (LLCs) are…

Hardware Architecture · Computer Science 2025-07-23 Connor Sullivan , Alex Manley , Mohammad Alian , Heechul Yun

The configurable building blocks of current FPGAs -- Logic blocks (LBs), Digital Signal Processing (DSP) slices, and Block RAMs (BRAMs) -- make them efficient hardware accelerators for the rapid-changing world of Deep Learning (DL).…

Hardware Architecture · Computer Science 2021-10-01 Aman Arora , Bagus Hanindhito , Lizy K. John

In modern Commercial Off-The-Shelf (COTS) multicore systems, each core can generate many parallel memory requests at a time. The processing of these parallel requests in the DRAM controller greatly affects the memory interference delay…

Distributed, Parallel, and Cluster Computing · Computer Science 2014-07-29 Heechul Yun

This paper discusses recent research that aims to enable computation close to data, an approach we broadly call processing-in-memory (PIM). PIM places computation mechanisms in or near where the data is stored (i.e., inside memory chips or…

Hardware Architecture · Computer Science 2025-02-07 Onur Mutlu , Saugata Ghose , Juan Gómez-Luna , Rachata Ausavarungnirun , Mohammad Sadrosadati , Geraldo F. Oliveira

Graphics Processing Units (GPUs) have traditionally relied on the host CPU to initiate access to the data storage. This approach is well-suited for GPU applications with known data access patterns that enable partitioning of their dataset…