Application of alternating decision trees in selecting sparse linear solvers

Handle URI:
http://hdl.handle.net/10754/575513
Title:
Application of alternating decision trees in selecting sparse linear solvers
Authors:
Bhowmick, Sanjukta; Eijkhout, Victor; Freund, Yoav; Fuentes, Erika; Keyes, David E. ( 0000-0002-4052-7224 )
Abstract:
The solution of sparse linear systems, a fundamental and resource-intensive task in scientific computing, can be approached through multiple algorithms. Using an algorithm well adapted to characteristics of the task can significantly enhance the performance, such as reducing the time required for the operation, without compromising the quality of the result. However, the best solution method can vary even across linear systems generated in course of the same PDE-based simulation, thereby making solver selection a very challenging problem. In this paper, we use a machine learning technique, Alternating Decision Trees (ADT), to select efficient solvers based on the properties of sparse linear systems and runtime-dependent features, such as the stages of simulation. We demonstrate the effectiveness of this method through empirical results over linear systems drawn from computational fluid dynamics and magnetohydrodynamics applications. The results also demonstrate that using ADT can resolve the problem of over-fitting, which occurs when limited amount of data is available. © 2010 Springer Science+Business Media LLC.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Applied Mathematics and Computational Science Program; Extreme Computing Research Center
Publisher:
Springer Science + Business Media
Journal:
Software Automatic Tuning
Issue Date:
2010
DOI:
10.1007/978-1-4419-6935-4_10
Type:
Book Chapter
ISBN:
9781441969347
Appears in Collections:
Applied Mathematics and Computational Science Program; Extreme Computing Research Center; Book Chapters; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorBhowmick, Sanjuktaen
dc.contributor.authorEijkhout, Victoren
dc.contributor.authorFreund, Yoaven
dc.contributor.authorFuentes, Erikaen
dc.contributor.authorKeyes, David E.en
dc.date.accessioned2015-08-24T09:54:46Zen
dc.date.available2015-08-24T09:54:46Zen
dc.date.issued2010en
dc.identifier.isbn9781441969347en
dc.identifier.doi10.1007/978-1-4419-6935-4_10en
dc.identifier.urihttp://hdl.handle.net/10754/575513en
dc.description.abstractThe solution of sparse linear systems, a fundamental and resource-intensive task in scientific computing, can be approached through multiple algorithms. Using an algorithm well adapted to characteristics of the task can significantly enhance the performance, such as reducing the time required for the operation, without compromising the quality of the result. However, the best solution method can vary even across linear systems generated in course of the same PDE-based simulation, thereby making solver selection a very challenging problem. In this paper, we use a machine learning technique, Alternating Decision Trees (ADT), to select efficient solvers based on the properties of sparse linear systems and runtime-dependent features, such as the stages of simulation. We demonstrate the effectiveness of this method through empirical results over linear systems drawn from computational fluid dynamics and magnetohydrodynamics applications. The results also demonstrate that using ADT can resolve the problem of over-fitting, which occurs when limited amount of data is available. © 2010 Springer Science+Business Media LLC.en
dc.publisherSpringer Science + Business Mediaen
dc.titleApplication of alternating decision trees in selecting sparse linear solversen
dc.typeBook Chapteren
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentApplied Mathematics and Computational Science Programen
dc.contributor.departmentExtreme Computing Research Centeren
dc.identifier.journalSoftware Automatic Tuningen
dc.contributor.institutionDepartment of Computer Science, University of Nebraska, Omaha, NE, United Statesen
dc.contributor.institutionAdvanced Computing Center, University of Texas at Austin, United Statesen
dc.contributor.institutionDepartment of Computer Science and Engineering, University of California, San Diego, United Statesen
dc.contributor.institutionMicrosoft Inc., United Statesen
dc.contributor.institutionDepartment of Applied Physics and Applied Mathematics, Columbia University, United Statesen
kaust.authorKeyes, David E.en
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.