English
Related papers

Related papers: Tackling the Matrix Multiplication Micro-kernel Ge…

200 papers

Sparse general matrix-matrix multiplication (SpGEMM) is a critical operation in many applications. Current multithreaded implementations are based on Gustavson's algorithm and often perform poorly on large matrices due to limited cache…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-04-11 Jordi Wolfson-Pou , Jan Laukemann , Fabrizio Petrini

The widespread adoption of large language models such as ChatGPT and Bard has led to unprecedented demand for these technologies. The burgeoning cost of inference for ever-increasing model sizes coupled with hardware shortages has limited…

Machine Learning · Computer Science 2023-05-24 Vishvak Murahari , Ameet Deshpande , Carlos E. Jimenez , Izhak Shafran , Mingqiu Wang , Yuan Cao , Karthik Narasimhan

Linear-scaling electronic-structure techniques, also called O(N) techniques, rely heavily on the multiplication of sparse matrices, where the sparsity arises from spatial cut-offs. In order to treat very large systems, the calculations must…

Materials Science · Physics 2009-10-31 D. R. Bowler , T. Miyazaki , M. J. Gillan

General Matrix Multiplication (GEMM) is the cornerstone of HPC workloads and Deep Learning. State-of-the-art vendor libraries tune tensor layouts, parallelization schemes, and cache blocking to minimize data movement across the memory…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-04-09 Evangelos Georganas , Alexander Heinecke , Pradeep Dubey

Modern GPGPUs provide massive arithmetic throughput, yet many scientific kernels remain limited by memory bandwidth. In particular, repeatedly loading precomputed auxiliary data wastes abundant compute resources while stressing the memory…

Performance · Computer Science 2025-11-04 Zijian Cao , Qiao Sun , Tiangong Zhang , Huiyuan Li

Fourier Neural Operators (FNO) are widely used for learning partial differential equation solution operators. However, FNO lacks architecture-aware optimizations,with its Fourier layers executing FFT, filtering, GEMM, zero padding, and iFFT…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-04-17 Shixun Wu , Yujia Zhai , Huangliang Dai , Hairui Zhao , Yue Zhu , Haiyang Hu , Zizhong Chen

Kernel methods provide an elegant and principled approach to nonparametric learning, but so far could hardly be used in large scale problems, since na\"ive implementations scale poorly with data size. Recent advances have shown the benefits…

Machine Learning · Computer Science 2020-11-30 Giacomo Meanti , Luigi Carratino , Lorenzo Rosasco , Alessandro Rudi

This study was aimed at simultaneously achieving sufficient accuracy and high performance for general matrix multiplications. Recent architectures, such as NVIDIA GPUs, feature high-performance units designed for low-precision matrix…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-04-29 Yuki Uchino , Katsuhisa Ozaki , Toshiyuki Imamura

General sparse matrix-matrix multiplication (SpGEMM) is a fundamental building block for numerous applications such as algebraic multigrid method (AMG), breadth first search and shortest path problem. Compared to other sparse BLAS routines,…

Mathematical Software · Computer Science 2015-09-15 Weifeng Liu , Brian Vinter

Among ML operators today, GEneralMatrix Multiplication (GEMM)-based operators are known to be key operators that build the main backbone of ML models. As their computational overhead dominates the overall execution time (e.g., 42.8% - 96.6%…

Hardware Architecture · Computer Science 2025-04-17 Rachid Karami , Sheng-Chun Kao , Hyoukjun Kwon

The level of abstraction at which application experts reason about linear algebra computations and the level of abstraction used by developers of high-performance numerical linear algebra libraries do not match. The former is conveniently…

Mathematical Software · Computer Science 2020-08-10 Henrik Barthels , Christos Psarras , Paolo Bientinesi

Many useful tasks in data science and machine learning applications can be written as simple variations of matrix multiplication. However, users have difficulty performing such tasks as existing matrix/vector libraries support only a…

Programming Languages · Computer Science 2023-05-17 Junyoung Kim , Kenneth Ross , Eric Sedlar , Lukas Stadler

Factorization and multiplication of dense matrices and tensors are critical, yet extremely expensive pieces of the scientific toolbox. Careful use of low rank approximation can drastically reduce the computation and memory requirements of…

Performance · Computer Science 2023-11-15 Sameer Deshmukh , Rio Yokota , George Bosilca

Despite advances in scalable models, the inference tools used for Gaussian processes (GPs) have yet to fully capitalize on developments in computing hardware. We present an efficient and general approach to GP inference based on Blackbox…

Machine Learning · Computer Science 2021-07-01 Jacob R. Gardner , Geoff Pleiss , David Bindel , Kilian Q. Weinberger , Andrew Gordon Wilson

Data movement is the dominating factor affecting performance and energy in modern computing systems. Consequently, many algorithms have been developed to minimize the number of I/O operations for common computing patterns. Matrix…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-01-26 Johannes de Fine Licht , Grzegorz Kwasniewski , Torsten Hoefler

Tile-based many-Processing Element (PE) accelerators can achieve competitive performance on General Matrix Multiplication (GEMM), but they are extremely hard to program, as their optimal software mapping is deeply coupled with hardware…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-12-16 Aofeng Shen , Chi Zhang , Yakup Budanaz , Alexandru Calotoiu , Torsten Hoefler , Luca Benini

Support for lower precision computation is becoming more common in accelerator hardware due to lower power usage, reduced data movement and increased computational performance. However, computational science and engineering (CSE) problems…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-09-06 Jennifer A. Loe , Christian A. Glusa , Ichitaro Yamazaki , Erik G. Boman , Sivasankaran Rajamanickam

In biological data, it is often the case that observed data are available only for a subset of samples. When a kernel matrix is derived from such data, we have to leave the entries for unavailable samples as missing. In this paper, we make…

Machine Learning · Computer Science 2007-05-23 Koji Tsuda , Shotaro Akaho , Kiyoshi Asai

Big data and the Internet of Things era continue to challenge computational systems. Several technology solutions such as NoSQL databases have been developed to deal with this challenge. In order to generate meaningful results from large…

Data Structures and Algorithms · Computer Science 2016-11-11 Vijay Gadepally , Jake Bolewski , Dan Hook , Dylan Hutchison , Ben Miller , Jeremy Kepner

Scaling up the sparse matrix-vector multiplication kernel on modern Graphics Processing Units (GPU) has been at the heart of numerous studies in both academia and industry. In this article we present a novel non-parametric, self-tunable,…

Numerical Analysis · Computer Science 2012-12-24 Xintian Yang , Srinivasan Parthasarathy , Ponnuswamy Sadayappan