Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Discovery Grants Program - Individual

University of Toronto

$55,000.00

Harnessing Computation Irregularity for Performance Optimization

EF 2024-25

Résumé
The main goal of my research program over the next five years will be to build high-performance and easy-to-use systems that significantly improve the performance of simulations that operate on very large data sets. To enhance the scalability and efficiency of large-scale simulations in machine learning and scientific computing, the adoption of data compression techniques, such as sparsification and quantization, as well as algorithms designed to process compressed data, has gained significant traction. Unfortunately, the compression process results in code with irregular memory accesses, creating hurdles for effective code optimization. This limitation renders many existing software systems inadequate for accelerating scientific and data analytics simulations. My research is dedicated to addressing this critical issue by pioneering novel algorithms as well as software techniques that automatically convert code with irregular computation patterns into efficient and scalable code. Over the next five years, I will develop scalable and easy-to-deploy systems for large-scale data processing. My research will be centred around three parallel and complementary tasks: (1) Inventing functionality to enable general compilers to optimize irregular codes; (2) Co-designing compression algorithms and systems for machine learning; (3) Inventing novel communication collectives and abstractions for distributed execution of large machine learning applications. The proposed research program is set to advance scientific and machine learning simulations via enabling the efficient processing of large amounts of data. It will also improve programmer productivity and foster both cost and energy savings.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
RGPIN-2024-06018
Name nom
Mehri Dehnavi, Maryam
Department département
Computer Science
Organization id
31
Institution établissement
University of Toronto
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
55000
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
802
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Information systems and technology
Area of application fr
Systèmes d'information et technologie de l'information
Research subject code
2704
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Distributed and parallel processing
Research subject fr
Traitement réparti et simultané
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Harnessing Computation Irregularity for Performance Optimization
Keywords
parallel computing; vectorization; code generators; compilers; sparse computations; performance optimization; machine learning; systems; high performance computing; optimization methods
Application summary
The main goal of my research program over the next five years will be to build high-performance and easy-to-use systems that significantly improve the performance of simulations that operate on very large data sets. To enhance the scalability and efficiency of large-scale simulations in machine learning and scientific computing, the adoption of data compression techniques, such as sparsification and quantization, as well as algorithms designed to process compressed data, has gained significant traction. Unfortunately, the compression process results in code with irregular memory accesses, creating hurdles for effective code optimization. This limitation renders many existing software systems inadequate for accelerating scientific and data analytics simulations. My research is dedicated to addressing this critical issue by pioneering novel algorithms as well as software techniques that automatically convert code with irregular computation patterns into efficient and scalable code. Over the next five years, I will develop scalable and easy-to-deploy systems for large-scale data processing. My research will be centred around three parallel and complementary tasks: (1) Inventing functionality to enable general compilers to optimize irregular codes; (2) Co-designing compression algorithms and systems for machine learning; (3) Inventing novel communication collectives and abstractions for distributed execution of large machine learning applications. The proposed research program is set to advance scientific and machine learning simulations via enabling the efficient processing of large amounts of data. It will also improve programmer productivity and foster both cost and energy savings.