PLATFORM: Parallel Linear Algebra Tool FOr Reduced Modeling
With advances in the scope of computational modeling methodologies, an increased focus is being placed on the application of data-driven techniques to increasingly complex problems. Due to the associated scale of the application, processing large datasets has emerged as a development bottleneck in p...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-02-01
|
| Series: | SoftwareX |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352711023000092 |
| _version_ | 1811166137414582272 |
|---|---|
| author | Nicholas Arnold-Medabalimi Christopher R. Wentland Cheng Huang Karthik Duraisamy |
| author_facet | Nicholas Arnold-Medabalimi Christopher R. Wentland Cheng Huang Karthik Duraisamy |
| author_sort | Nicholas Arnold-Medabalimi |
| collection | DOAJ |
| description | With advances in the scope of computational modeling methodologies, an increased focus is being placed on the application of data-driven techniques to increasingly complex problems. Due to the associated scale of the application, processing large datasets has emerged as a development bottleneck in practical applications of data-driven methods. While large-scale partial differential equation solvers are optimized for sparse linear algebra, many data-decomposition techniques (e.g. the singular value decomposition) require dense linear algebra operations. This work presents the tool PLATFORM which has enabled the application of modal decomposition and data-driven reduced-order modeling techniques for moderate (giga-) and large (tera-) scale data processing. The I/O and computing strategies and priorities are described. Most importantly, this framework uses abstraction techniques which allow users with limited understanding of distributed linear algebra computations and I/O to flexibly prototype and test methods on memory-intensive problems that are demanding in memory for the scripting environment. |
| first_indexed | 2024-04-10T15:48:47Z |
| format | Article |
| id | doaj.art-a4954bf99c5a42f5989c5e876c96ac40 |
| institution | Directory Open Access Journal |
| issn | 2352-7110 |
| language | English |
| last_indexed | 2024-04-10T15:48:47Z |
| publishDate | 2023-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | SoftwareX |
| spelling | doaj.art-a4954bf99c5a42f5989c5e876c96ac402023-02-12T04:15:12ZengElsevierSoftwareX2352-71102023-02-0121101313PLATFORM: Parallel Linear Algebra Tool FOr Reduced ModelingNicholas Arnold-Medabalimi0Christopher R. Wentland1Cheng Huang2Karthik Duraisamy3University of Michigan, Department of Aerospace Engineering, United States of America; Corresponding author.University of Michigan, Department of Aerospace Engineering, United States of AmericaUniverstiy of Kansas, Department of Aerospace Engineering, United States of AmericaUniversity of Michigan, Department of Aerospace Engineering, United States of AmericaWith advances in the scope of computational modeling methodologies, an increased focus is being placed on the application of data-driven techniques to increasingly complex problems. Due to the associated scale of the application, processing large datasets has emerged as a development bottleneck in practical applications of data-driven methods. While large-scale partial differential equation solvers are optimized for sparse linear algebra, many data-decomposition techniques (e.g. the singular value decomposition) require dense linear algebra operations. This work presents the tool PLATFORM which has enabled the application of modal decomposition and data-driven reduced-order modeling techniques for moderate (giga-) and large (tera-) scale data processing. The I/O and computing strategies and priorities are described. Most importantly, this framework uses abstraction techniques which allow users with limited understanding of distributed linear algebra computations and I/O to flexibly prototype and test methods on memory-intensive problems that are demanding in memory for the scripting environment.http://www.sciencedirect.com/science/article/pii/S2352711023000092Model reductionModal decompositionDistributed linear algebra |
| spellingShingle | Nicholas Arnold-Medabalimi Christopher R. Wentland Cheng Huang Karthik Duraisamy PLATFORM: Parallel Linear Algebra Tool FOr Reduced Modeling SoftwareX Model reduction Modal decomposition Distributed linear algebra |
| title | PLATFORM: Parallel Linear Algebra Tool FOr Reduced Modeling |
| title_full | PLATFORM: Parallel Linear Algebra Tool FOr Reduced Modeling |
| title_fullStr | PLATFORM: Parallel Linear Algebra Tool FOr Reduced Modeling |
| title_full_unstemmed | PLATFORM: Parallel Linear Algebra Tool FOr Reduced Modeling |
| title_short | PLATFORM: Parallel Linear Algebra Tool FOr Reduced Modeling |
| title_sort | platform parallel linear algebra tool for reduced modeling |
| topic | Model reduction Modal decomposition Distributed linear algebra |
| url | http://www.sciencedirect.com/science/article/pii/S2352711023000092 |
| work_keys_str_mv | AT nicholasarnoldmedabalimi platformparallellinearalgebratoolforreducedmodeling AT christopherrwentland platformparallellinearalgebratoolforreducedmodeling AT chenghuang platformparallellinearalgebratoolforreducedmodeling AT karthikduraisamy platformparallellinearalgebratoolforreducedmodeling |