English
Related papers

Related papers: Investigating Warp Size Impact in GPUs

200 papers

Increasing the mini-batch size for stochastic gradient descent offers significant opportunities to reduce wall-clock training time, but there are a variety of theoretical and systems challenges that impede the widespread success of this…

This paper provides an in-depth characterization of GPU-accelerated systems, to understand the interplay between overlapping computation and communication which is commonly employed in distributed training settings. Due to the large size of…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-07-08 Seonho Lee , Jihwan Oh , Junkyum Kim , Seokjin Go , Jongse Park , Divya Mahajan

Graphics Processing Units (GPUs) are widely used by various applications in a broad variety of fields to accelerate their computation but remain susceptible to transient hardware faults (soft errors) that can easily compromise application…

Software Engineering · Computer Science 2021-03-30 Lishan Yang , Bin Nie , Adwait Jog , Evgenia Smirni

Discrete optimization is a central problem in artificial intelligence. The optimization of the aggregated cost of a network of cost functions arises in a variety of problems including (W)CSP, DCOP, as well as optimization in stochastic…

Artificial Intelligence · Computer Science 2018-01-12 Ferdinando Fioretto , Enrico Pontelli , William Yeoh , Rina Dechter

In this paper, we explore the limits of graphics processors (GPUs) for general purpose parallel computing by studying problems that require highly irregular data access patterns: parallel graph algorithms for list ranking and connected…

Distributed, Parallel, and Cluster Computing · Computer Science 2010-02-25 Frank Dehne , Kumanan Yogaratnam

Serving deep neural networks in latency critical interactive settings often requires GPU acceleration. However, the small batch sizes typical in online inference results in poor GPU utilization, a potential performance gap which GPU…

Distributed, Parallel, and Cluster Computing · Computer Science 2019-01-03 Paras Jain , Xiangxi Mo , Ajay Jain , Harikaran Subbaraj , Rehan Sohail Durrani , Alexey Tumanov , Joseph Gonzalez , Ion Stoica

Contemporary GPUs allow concurrent execution of small computational kernels in order to prevent idling of GPU resources. Despite the potential concurrency between independent kernels, the order in which kernels are issued to the GPU will…

Distributed, Parallel, and Cluster Computing · Computer Science 2015-11-26 Teng Li , Vikram K. Narayana , Tarek El-Ghazawi

Convolution is the most time-consuming operation in deep neural network operations, so its performance is critical to the overall performance of the neural network. The commonly used methods for convolution on GPU include the general matrix…

Neural and Evolutionary Computing · Computer Science 2023-06-27 Shuai Lu , Jun Chu , Luanzheng Guo , Xu T. Liu

Graphics Processing Units (GPUs) are over-stressed to accelerate High-Performance Computing applications and are used to accelerate Deep Neural Networks in several domains where they have a life expectancy of many years. These conditions…

Hardware Architecture · Computer Science 2023-10-03 Juan-David Guerrero-Balaguera , Josie E. Rodriguez Condia , Fernando F. dos Santos , Matteo Sonza , Paolo Rech

It is well known that, for most datasets, the use of large-size minibatches for Stochastic Gradient Descent (SGD) typically leads to slow convergence and poor generalization. On the other hand, large minibatches are of great practical…

Machine Learning · Computer Science 2018-06-20 Matteo Fischetti , Iacopo Mandatelli , Domenico Salvagnin

This paper explores the impact of simulator accuracy on architecture design decisions in the general-purpose graphics processing unit (GPGPU) space. We perform a detailed, quantitative analysis of the most popular publicly available GPU…

Hardware Architecture · Computer Science 2020-06-04 Mahmoud Khairy , Jain Akshay , Tor Aamodt , Timothy G. Rogers

Energy efficiency has become one of the top design criteria for current computing systems. The dynamic voltage and frequency scaling (DVFS) has been widely adopted by laptop computers, servers, and mobile devices to conserve energy, while…

Distributed, Parallel, and Cluster Computing · Computer Science 2016-10-07 Xinxin Mei , Qiang Wang , Xiaowen Chu

Memory latency, bandwidth, capacity, and energy increasingly limit performance. In this paper, we reconsider proposed system architectures that consist of huge (many-terabyte to petabyte scale) memories shared among large numbers of CPUs.…

Hardware Architecture · Computer Science 2025-09-24 Samuel Dayo , Shuhan Liu , Peijing Li , Philip Levis , Subhasish Mitra , Thierry Tambe , David Tennenhouse , H. -S. Philip Wong

Domain-specific languages that execute image processing pipelineson GPUs, such as Halide and Forma, operate by 1) dividing the image into overlapped tiles, and 2) fusing loops to improve memory locality. However, current approaches have…

Programming Languages · Computer Science 2020-09-09 Abhinav Jangda , Arjun Guha

Graph processing is typically considered to be a memory-bound rather than compute-bound problem. One common line of thought is that more available memory bandwidth corresponds to better graph processing performance. However, in this work we…

Distributed, Parallel, and Cluster Computing · Computer Science 2019-10-10 Oded Green , James Fox , Jeffrey Young , Jun Shirako , David Bader

As research and deployment of AI grows, the computational burden to support and sustain its progress inevitably does too. To train or fine-tune state-of-the-art models in NLP, computer vision, etc., some form of AI hardware acceleration is…

Hardware Architecture · Computer Science 2024-03-01 Dan Zhao , Siddharth Samsi , Joseph McDonald , Baolin Li , David Bestor , Michael Jones , Devesh Tiwari , Vijay Gadepally

The impact of the maximally possible batch size (for the better runtime) on performance of graphic processing units (GPU) and tensor processing units (TPU) during training and inference phases is investigated. The numerous runs of the…

Machine Learning · Computer Science 2019-04-04 Yuriy Kochura , Yuri Gordienko , Vlad Taran , Nikita Gordienko , Alexandr Rokovyi , Oleg Alienin , Sergii Stirenko

RISC-V GPUs present a promising path for supporting GPU applications. Traditionally, GPUs achieve high efficiency through the SPMD (Single Program Multiple Data) programming model. However, modern GPU programming increasingly relies on…

Hardware Architecture · Computer Science 2025-05-07 Huanzhi Pu , Rishabh Ravi , Shinnung Jeong , Udit Subramanya , Euijun Chung , Jisheng Zhao , Chihyo Ahn , Hyesoon Kim

Modern Graphics Processing Units (GPUs) are well provisioned to support the concurrent execution of thousands of threads. Unfortunately, different bottlenecks during execution and heterogeneous application requirements create imbalances in…

GPGPUs use the Single-Instruction-Multiple-Thread (SIMT) execution model where a group of threads-wavefront or warp-execute instructions in lockstep. When threads in a group encounter a branching instruction, not all threads in the group…

Programming Languages · Computer Science 2022-01-17 Charitha Saumya , Kirshanthan Sundararajah , Milind Kulkarni