English
Related papers

Related papers: Guaranteed DGEMM Accuracy While Using Reduced Prec…

200 papers

Tucker decomposition is one of the SOTA CNN model compression techniques. However, unlike the FLOPs reduction, we observe very limited inference time reduction with Tucker-compressed models using existing GPU software such as cuDNN. To this…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-01-06 Lizhi Xiang , Miao Yin , Chengming Zhang , Aravind Sukumaran-Rajam , P. Sadayappan , Bo Yuan , Dingwen Tao

FPGA is appropriate for fix-point neural networks computing due to high power efficiency and configurability. However, its design must be intensively refined to achieve high performance using limited hardware resources. We present an…

Hardware Architecture · Computer Science 2022-01-03 Qingyang Yi , Heming Sun , Masahiro Fujita

Modern datacenters increasingly rely on low-power, single-slot inference accelerators to balance performance, energy efficiency, and rack density constraints. The NVIDIA T4 GPU has become widely deployed due to strong performance per watt…

Performance · Computer Science 2026-05-07 Kathiravan Palaniappan

In this paper, we explore the acceleration of tensor product operations in finite element methods, leveraging the computational power of the NVIDIA A100 GPU Tensor Cores. We provide an accessible overview of the necessary mathematical…

Mathematical Software · Computer Science 2024-07-16 Cu Cui

Large matrix multiplication is a cornerstone of modern machine learning workloads, yet traditional approaches suffer from cubic computational complexity (e.g., $\mathcal{O}(n^3)$ for a matrix of size $n\times n$). We present Low-Rank GEMM,…

Performance · Computer Science 2025-11-25 Alfredo Metere

Large language models (LLMs) have been widely applied but face challenges in efficient inference. While quantization methods reduce computational demands, ultra-low bit quantization with arbitrary precision is hindered by limited GPU Tensor…

Machine Learning · Computer Science 2025-03-14 Shaobo Ma , Chao Fang , Haikuo Shao , Zhongfeng Wang

This paper lays out insights and opportunities for implementing higher-precision matrix-matrix multiplication (GEMM) from (in terms of) lower-precision high-performance GEMM. The driving case study approximates double-double precision…

Mathematical Software · Computer Science 2024-12-31 Devangi N. Parikh , Robert A. van de Geijn , Greg M. Henry

Transformers are at the core of modern AI nowadays. They rely heavily on matrix multiplication and require efficient acceleration due to their substantial memory and computational requirements. Quantization plays a vital role in reducing…

Hardware Architecture · Computer Science 2026-04-03 Ahmed J. Abdelmaksoud , Cristian Sestito , Shiwei Wang , Themis Prodromakis

Tensor decomposition has been widely used in machine learning and high-volume data analysis. However, large-scale tensor factorization often consumes huge memory and computing cost. Meanwhile, modernized computing hardware such as tensor…

Optimization and Control · Mathematics 2022-09-12 Zi Yang , Junnan Shan , Zheng Zhang

The proliferation of high-throughput sequencing machines ensures rapid generation of up to billions of short nucleotide fragments in a short period of time. This massive amount of sequence data can quickly overwhelm today's storage and…

Distributed, Parallel, and Cluster Computing · Computer Science 2018-11-13 Subho S. Banerjee , Mohamed El-Hadedy , Jong Bin Lim , Zbigniew T. Kalbarczyk , Deming Chen , Steve Lumetta , Ravishankar K. Iyer

Modern processors deliver higher throughput for lower-precision arithmetic than for higher-precision arithmetic. For matrix multiplication, the Ozaki scheme exploits this performance gap by splitting the inputs into lower-precision…

Mathematical Software · Computer Science 2026-04-01 Shota Kawakami , Daisuke Takahashi

Recent studies from several hyperscalars pinpoint to embedding layers as the most memory-intensive deep learning (DL) algorithm being deployed in today's datacenters. This paper addresses the memory capacity and bandwidth challenges of…

Machine Learning · Computer Science 2019-08-27 Youngeun Kwon , Yunjae Lee , Minsoo Rhu

Arbitrary-precision integer multiplication is the core kernel of many applications in simulation, cryptography, etc. Existing acceleration of arbitrary-precision integer multiplication includes CPUs, GPUs, FPGAs, and ASICs. Among these…

Hardware Architecture · Computer Science 2023-09-22 Zhuoping Yang , Jinming Zhuang , Jiaqi Yin , Cunxi Yu , Alex K. Jones , Peipei Zhou

This study explores the use of INT8-based emulation for accelerating traditional FP64-based HPC workloads on modern GPU architectures. Through SCILIB-Accel automatic BLAS offload tool for cache-coherent Unified Memory Architecture, we…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-04-01 Hang Liu , Junjie Li , Yinzhi Wang , Niraj K. Nepal , Yang Wang

Modern GPUs are equipped with tensor cores (TCs) that are commonly used for matrix multiplication in artificial intelligence workloads. However, because they have high computational throughput, they can lead to significant performance gains…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-09-01 Brian Curless , Michael Gowanlock

General matrix-matrix multiplication (GEMM) is a cornerstone of AI computations, making tensor processing engines (TPEs) increasingly critical in GPUs and domain-specific architectures. Existing architectures primarily optimize dataflow or…

Hardware Architecture · Computer Science 2025-03-11 Qizhe Wu , Huawen Liang , Yuchen Gui , Zhichen Zeng , Zerong He , Linfeng Tao , Xiaotian Wang , Letian Zhao , Zhaoxi Zeng , Wei Yuan , Wei Wu , Xi Jin

Low precision data representation is important to reduce storage size and memory access for convolutional neural networks (CNNs). Yet, existing methods have two major limitations: (1) requiring re-training to maintain accuracy for deep…

Signal Processing · Electrical Eng. & Systems 2020-03-10 Chen Wu , Mingyu Wang , Xinyuan Chu , Kun Wang , Lei He

Tensor Cores have been an important unit to accelerate Fused Matrix Multiplication Accumulation (MMA) in all NVIDIA GPUs since Volta Architecture. To program Tensor Cores, users have to use either legacy wmma APIs or current mma APIs.…

Hardware Architecture · Computer Science 2022-11-29 Wei Sun , Ang Li , Tong Geng , Sander Stuijk , Henk Corporaal

Many recent computational accelerators provide non-standard (e.g., reduced precision) arithmetic operations to enhance performance for floating-point matrix multiplication. Unfortunately, the properties of these accelerators are not widely…

Hardware Architecture · Computer Science 2025-02-25 Benjamin Valpey , Xinyi Li , Sreepathi Pai , Ganesh Gopalakrishnan

Recent hardware acceleration advances have enabled powerful specialized accelerators for finite element computations, spiking neural network inference, and sparse tensor operations. However, existing approaches face fundamental limitations:…

Hardware Architecture · Computer Science 2026-01-09 Chuanzhen Wang , Leo Zhang , Eric Liu