English
Related papers

Related papers: Toward a Universal GPU Instruction Set Architectur…

200 papers

Efficiently exploiting GPUs is increasingly essential in scientific computing, as many current and upcoming supercomputers are built using them. To facilitate this, there are a number of programming approaches, such as CUDA, OpenACC and…

Performance · Computer Science 2017-11-07 G. D. Balogh , I. Z. Reguly , G. R. Mudalige

Understanding GPU topology is essential for performance-related tasks in HPC or AI. Yet, unlike for CPUs with tools like hwloc, GPU information is hard to come by, incomplete, and vendor-specific. In this work, we address this gap and…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-11-11 Stepan Vanecek , Manuel Walter Mussbacher , Dominik Groessler , Urvij Saroliya , Martin Schulz

As large language models continue to scale, training demands on compute and system capacity grow rapidly, making single-vendor homogeneous clusters insufficient. This paper presents a technical solution for heterogeneous mixed training in…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-02-23 Jon Hu , Thomas Jia , Jing Zhu , Zhendong Yu

As scientific codes are ported between GPU platforms, continuous testing is required to ensure numerical robustness and identify numerical differences. Compiler-induced numerical differences occur when a program is compiled and run on…

Numerical Analysis · Mathematics 2024-10-15 Anwar Hossain Zahid , Ignacio Laguna , Wei Le

Parallel computing using accelerators has gained widespread research attention in the past few years. In particular, using GPUs for general purpose computing has brought forth several success stories with respect to time taken, cost, power,…

Domain specific accelerators present new challenges and opportunities for code generation onto novel instruction sets, communication fabrics, and memory architectures. In this paper we introduce an intermediate representation (IR) which…

Distributed, Parallel, and Cluster Computing · Computer Science 2018-10-24 Matthew Sotoudeh , Anand Venkat , Michael Anderson , Evangelos Georganas , Alexander Heinecke , Jason Knight

SMEs increasingly seek alternatives to cloud LLM APIs, which raise data privacy concerns. Dedicated cloud GPU instances offer improved privacy but with limited guarantees and ongoing costs, while professional on-premise hardware (A100,…

Machine Learning · Computer Science 2026-01-15 Jonathan Knoop , Hendrik Holtmann

We present a cross-architecture evaluation of production LLM inference on AMD Instinct MI325X GPUs, benchmarking four models spanning 235B to 1 trillion parameters across three architectural families (MoE+MLA, Dense+GQA, MoE+GQA) on an…

Hardware Architecture · Computer Science 2026-03-12 Athos Georgiou

We investigate GPU-based parallelization of Iterative-Deepening A* (IDA*). We show that straightforward thread-based parallelization techniques which were previously proposed for massively parallel SIMD processors perform poorly due to warp…

Artificial Intelligence · Computer Science 2017-05-09 Satoru Horie , Alex Fukunaga

Deep learning and hardware for it has garnered immense academic and industry interest in the past 5 years, with many novel proposals. However, the state-of-art remains NVIDIA's TensorCore-based systems that provide top-of-line performance…

Hardware Architecture · Computer Science 2023-01-12 Michael Davies , Karthikeyan Sankaralingam

Efficiently scaling deep neural networks across GPU clusters requires navigating complex trade-offs between computational throughput, memory utilization, and synchronization overhead. This paper presents a unified empirical evaluation of…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-01-06 Md Sultanul Islam Ovi

A major bottleneck in scenario-based Sample Average Approximation (SAA) for stochastic programming (SP) is the cost of solving an exact second-stage problem for every scenario, especially when each scenario contains an NP-hard combinatorial…

Optimization and Control · Mathematics 2026-05-12 Jingyi Zhao , Linxin Yang , Haohua Zhang , Qile He , Tian Ding

The growth of data to be processed in the Oil & Gas industry matches the requirements imposed by evolving algorithms based on stencil computations, such as Full Waveform Inversion and Reverse Time Migration. Graphical processing units…

Distributed, Parallel, and Cluster Computing · Computer Science 2020-08-05 Vitor Hugo Mickus Rodrigues , Lucas Cavalcante , Maelso Bruno Pereira , Fabio Luporini , István Reguly , Gerard Gorman , Samuel Xavier de Souza

For NVIDIA GPUs, CUDA is the primary interface through which applications orchestrate GPU execution, yet much of the logic that realizes CUDA operations resides in NVIDIA's closed-source userspace driver. As a result, the translation from…

Performance · Computer Science 2026-04-30 Yuang Yan , Ian Karlin , Ryan Grant

This paper presents a comprehensive evaluation of Intel Gaudi NPUs as an alternative to NVIDIA GPUs, which is currently the de facto standard in AI system design. First, we create a suite of microbenchmarks to compare Intel Gaudi-2 with…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-03-25 Yunjae Lee , Juntaek Lim , Jehyeon Bang , Eunyeong Cho , Huijong Jeong , Taesu Kim , Hyungjun Kim , Joonhyung Lee , Jinseop Im , Ranggi Hwang , Se Jung Kwon , Dongsoo Lee , Minsoo Rhu

Graphics Processing Units (GPUs) are high performance co-processors originally intended to improve the use and quality of computer graphics applications. Once, researchers and practitioners noticed the potential of using GPU for general…

Numerical Analysis · Computer Science 2016-07-12 K. Parand , Saeed Zafarvahedian , Sayyed A. Hossayni

Due to the recent announcement of the Frontier supercomputer, many scientific application developers are working to make their applications compatible with AMD architectures (CPU-GPU), which means moving away from the traditional CPU and…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-11-11 Matthew Leinhauser , René Widera , Sergei Bastrakov , Alexander Debus , Michael Bussmann , Sunita Chandrasekaran

The proliferation of IoT devices and advancements in network technologies have intensified the demand for real-time data processing at the network edge. To address these demands, low-power AI accelerators, particularly GPUs, are…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-08-13 Abhinaba Chakraborty , Wouter Tavernier , Akis Kourtis , Mario Pickavet , Andreas Oikonomakis , Didier Colle

The proliferation of extensive neural network architectures, particularly deep learning models, presents a challenge in terms of resource-intensive training. GPU memory constraints have become a notable bottleneck in training such sizable…

Machine Learning · Computer Science 2025-02-07 Cevat Volkan Karadağ , Nezih Topaloğlu

We present Hawkeye, a system for analyzing and reproducing GPU-level arithmetic operations. Using our framework, anyone can re-execute on a CPU the exact matrix multiplication operations underlying a machine learning model training or…

Cryptography and Security · Computer Science 2026-05-19 Erez Badash , Dan Boneh , Ilan Komargodski , Megha Srivastava