Grant: Discovery Grants Program - Individual
The University of British Columbia
$52,000.00
Structure-Preserving Numerical Solvers for Block-Structured Linear Systems
EF 2023-24
Résumé
The recent emergence of new and exciting computing paradigms has enabled computational scientists to solve increasingly large and difficult scientific problems that were impossible to solve just a few years ago. The numerical solution of large and sparse linear systems has been a potential source of a computational bottleneck for as long as computational methods for solving scientific problems have existed; any progress in developing fast solvers for linear algebra problems has the potential to advance the entire field. Since many physical phenomena are local in nature, sparsity appears everywhere. The emergence of big--data problems further highlights the need to develop state--of--the--art methods for numerically solving large--scale linear algebra problems. This proposal concerns the development of fast algorithms for solving block--structured large and sparse linear systems. Such linear systems arise in a variety of applications in computational science and engineering. My main goal is to investigate and exploit the structural properties of problems that model the behavior of objects in the presence of multiple constraints, and in multiple physical domains with different properties and non-trivial interface conditions. My study will include exploration of the rank and sparsity structures, and other numerical properties of the matrices and underlying continuous operators. One of my primary long--term goals is to develop a comprehensive theory that facilitates the development of scalable and robust solvers for these problems. Numerical software will complement the theoretical study. My recent progress in the study of these problems and my long-time engagement in studying classical saddle--point systems have given me a broad perspective on the current state of research of these problems. They pose significant computational and algorithmic challenges and require a high level of expertise in numerical linear algebra, numerical solution of differential equations, scientific computing, and numerical analysis. The stated objectives of this proposal will provide opportunities for my students and postdoctoral fellows to gain deep knowledge and enhance their research skills. Some members of my research group are already working on these objectives and are making valuable contributions. I intend to recruit and involve three PhD students and one MSc student in steady state to pursue the proposed research objectives and to make further progress, and will also recruit a postdoctoral fellow at the early stages of the projects. I will offer a supportive environment, allowing the members of my group to gain valuable experience and prepare for their next steps in their careers. The proposed research will advance our ability to numerically solve large--scale linear systems and will hopefully generate a body of innovative new solution methods that will benefit the scientific community and society at large.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
British Columbia, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
RGPIN-2023-05244
Name nom
Greif, Chen
Department département
Computer Science
Organization id
2
Institution établissement
The University of British Columbia
Province en
British Columbia
Province fr
Colombie-Britannique
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2023
Competition year année de concours
2023
Award amount
52000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1507
Committee name en
Computer Science
Committee name fr
Sciences informatiques
Area of application code
1205
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Mathematical sciences
Area of application fr
Sciences mathématiques
Research subject code
2714
Research subject group en
Information technology
Research subject group fr
Mathematics of computing
Research subject en
Technologie de l'information
Research subject fr
Mathématiques de l'informatique
Application title
Structure-Preserving Numerical Solvers for Block-Structured Linear Systems
Keyword
double saddle-point systems; eigenvalue analysis; iterative methods; large and sparse linear systems; large-scale numerical problems; matrix decompositions; numerical linear algebra; numerical methods; preconditioning techniques; scalable algorithms
Application summary
The recent emergence of new and exciting computing paradigms has enabled computational scientists to solve increasingly large and difficult scientific problems that were impossible to solve just a few years ago. The numerical solution of large and sparse linear systems has been a potential source of a computational bottleneck for as long as computational methods for solving scientific problems have existed; any progress in developing fast solvers for linear algebra problems has the potential to advance the entire field. Since many physical phenomena are local in nature, sparsity appears everywhere. The emergence of big--data problems further highlights the need to develop state--of--the--art methods for numerically solving large--scale linear algebra problems. This proposal concerns the development of fast algorithms for solving block--structured large and sparse linear systems. Such linear systems arise in a variety of applications in computational science and engineering. My main goal is to investigate and exploit the structural properties of problems that model the behavior of objects in the presence of multiple constraints, and in multiple physical domains with different properties and non-trivial interface conditions. My study will include exploration of the rank and sparsity structures, and other numerical properties of the matrices and underlying continuous operators. One of my primary long--term goals is to develop a comprehensive theory that facilitates the development of scalable and robust solvers for these problems. Numerical software will complement the theoretical study. My recent progress in the study of these problems and my long-time engagement in studying classical saddle--point systems have given me a broad perspective on the current state of research of these problems. They pose significant computational and algorithmic challenges and require a high level of expertise in numerical linear algebra, numerical solution of differential equations, scientific computing, and numerical analysis. The stated objectives of this proposal will provide opportunities for my students and postdoctoral fellows to gain deep knowledge and enhance their research skills. Some members of my research group are already working on these objectives and are making valuable contributions. I intend to recruit and involve three PhD students and one MSc student in steady state to pursue the proposed research objectives and to make further progress, and will also recruit a postdoctoral fellow at the early stages of the projects. I will offer a supportive environment, allowing the members of my group to gain valuable experience and prepare for their next steps in their careers. The proposed research will advance our ability to numerically solve large--scale linear systems and will hopefully generate a body of innovative new solution methods that will benefit the scientific community and society at large.