Affiliation:
1. North Carolina State University, USA
2. Massachusetts Institute of Technology, USA
Abstract
A daunting challenge faced by program performance autotuning is input sensitivity, where the best autotuned configuration may vary with different input sets. This paper presents a novel two-level input learning algorithm to tackle the challenge for an important class of autotuning problems, algorithmic autotuning. The new approach uses a two-level input clustering method to automatically refine input grouping, feature selection, and classifier construction. Its design solves a series of open issues that are particularly essential to algorithmic autotuning, including the enormous optimization space, complex influence by deep input features, high cost in feature extraction, and variable accuracy of algorithmic choices. Experimental results show that the new solution yields up to a 3x speedup over using a single configuration for all inputs, and a 34x speedup over a traditional one-level method for addressing input sensitivity in program optimizations.
Funder
U.S. Department of Energy
National Science Foundation
Publisher
Association for Computing Machinery (ACM)
Subject
Computer Graphics and Computer-Aided Design,Software
Reference46 articles.
1. Government’s open data. http://www.data.org/. Government’s open data. http://www.data.org/.
2. UCI data sets. http://archive.ics.uci.edu/ml/datasets. UCI data sets. http://archive.ics.uci.edu/ml/datasets.
3. Using Machine Learning to Focus Iterative Optimization
4. Finding effective compilation sequences
Cited by
18 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. The politics of digital technologies: Reimagining social participation in the digital age;Proceedings of the 16th International Conference on Theory and Practice of Electronic Governance;2023-09-26
2. Analysing the Impact of Workloads on Modeling the Performance of Configurable Software Systems;2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE);2023-05
3. HPAC;Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis;2021-11-13
4. Bliss: auto-tuning complex applications using a pool of diverse lightweight learning models;Proceedings of the 42nd ACM SIGPLAN International Conference on Programming Language Design and Implementation;2021-06-18
5. Automatic Optimization of Matrix Implementations for Distributed Machine Learning and Linear Algebra;Proceedings of the 2021 International Conference on Management of Data;2021-06-09