Optimization Techniques for GPU Programming
54
TL;DR: In this article , a survey discusses various optimization techniques found in 450 articles published in the last 14 years and analyzes the optimizations from different perspectives which shows that the various optimizations are highly interrelated, explaining the need for techniques such as auto-tuning.
read more
Abstract: In the past decade, Graphics Processing Units have played an important role in the field of high-performance computing and they still advance new fields such as IoT, autonomous vehicles, and exascale computing. It is therefore important to understand how to extract performance from these processors, something that is not trivial. This survey discusses various optimization techniques found in 450 articles published in the last 14 years. We analyze the optimizations from different perspectives which shows that the various optimizations are highly interrelated, explaining the need for techniques such as auto-tuning.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Unleashing the potential: AI empowered advanced metasurface research
Yunlai Fu,Xuxi Zhou,Yiwan Yu,Jiawang Chen,Shuming Wang,Shining Zhu,Zhenlin Wang +6 more
TL;DR: AI-powered advanced metasurface research explores the intersection of AI and metasurfaces, leveraging AI's computational power to design, analyze, and optimize metasurfaces for various applications.
GPU acceleration of Levenshtein distance computation between long strings
TL;DR: In this paper , a GPU implementation of the WFA algorithm and a new optimization that can halve the elements to be computed, providing additional performance gains, are presented, which is the best ever reported.
7
Progress and Opportunities of Foundation Models in Bioinformatics
Qing Li,Zhihang Hu,Yixuan Wang,Lei Li,Yimin Fan,Irwin King,Le Song,Yu Li +7 more
TL;DR: A systematic investigation and summary of FMs in bioinformatics, tracing their evolution, current research status, and the methodologies employed, aiming to guide the research community in choosing appropriate FMs for their research needs.
6
Sustainable Optimizing Performance and Energy Efficiency in Proof of Work Blockchain: A Multilinear Regression Approach
Meennapa Rukhiran,Songwut Boonsong,Paniti Netinant +2 more
TL;DR: The results reveal that strategically adjusting GPU hardware, software, and configuration can preserve substantial energy while preserving computational efficiency, and offer practical recommendations for optimizing the feature configurations of GPUs to reduce energy consumption, mitigate the environmental impacts of blockchain operations, and contribute to the current research on performance in PoW blockchain applications.
6
References
General transformations for GPU execution of tree traversals
Michael Goldfarb,Youngjoon Jo,Milind Kulkarni +2 more
- 17 Nov 2013
TL;DR: This work argues that there are general-purpose techniques for implementing irregular algorithms on GPUs that exploit similarities in algorithmic structure rather than application-specific knowledge, and demonstrates these techniques on several tree traversal algorithms, achieving speedups of up to 38× over 32-thread CPU versions.
High-performance code generation for stencil computations on GPU architectures
Justin Holewinski,Louis-Noël Pouchet,P. Sadayappan +2 more
- 25 Jun 2012
TL;DR: This paper develops compiler algorithms for automatic generation of efficient, time-tiled stencil code for GPU accelerators from a high-level description of the stencil operation, and shows that the code generation scheme can achieve high performance on a range of GPU architectures, including both nVidia and AMD devices.
An Efficient Parallelization Strategy for Dynamic Programming on GPU
Karl-Eduard Berger,Francois Galea +1 more
- 20 May 2013
TL;DR: It is shown that parametrizing the solver parallelism according to the hardware allows better performance and provides good acceleration compared to a standard GPU parallel strategy on a dynamic programming-based implementation of the knapsack problem.
WolfGraph: The edge-centric graph processing on GPU
TL;DR: WolfGraph adopts the edge-centric processing, which iterates over the edges rather than vertices, and develops a new method, called Concatenated Edge List (CEL), to process a graph that is bigger than the global memory of GPU.
Fast sparse matrix-vector multiplication on GPUs: implications for graph mining
Xintian Yang,Srinivasan Parthasarathy,P. Sadayappan +2 more
- 01 Jan 2011
TL;DR: Using real web graph data, it is shown how a novel non-parametric, self-tunable, approach to data representation for computing this kernel, particularly targeting sparse matrices representing power-law graphs, can yield significant benefits over current state of the art GPU efforts.