Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Engage Grants Program

Queen's University

$25,000.00

Efficient and scalable collective communications for the Message Passing Interface

EF 2010-11

Résumé
High-Performance Computing (HPC) is used to tackle computationally-intensive problems such as the modeling of large genomes for cancer research, crash simulations in automotive industry, modeling of the global climate system and weather forecasting, oil and gas reservoir simulation, green energy, and molecular dynamics simulations for proteins and lipids. Processes in these simulations extensively communicate with each other through the Message Passing Interface (MPI) communication library. Such communications involve frequent MPI collective communication operations, where a group of processes collectively participate in the communication operations. Performance and scalability of collective communication operations is of paramount importance to HPC applications. Therefore, efficient collective communication algorithms have to be devised that could take advantage of the multi-core architectures, high-performance interconnects, and networking topologies in the emerging highly hierarchical HPC clusters. The purpose of this Engage project is to build a new collaboration on the development of efficient and scalable multi-core and network topology aware collective communications in the MPI library between Platform Computing and the Parallel Processing Research Laboratory at Queen's University. This project will focus on adapting the traditional collective communication algorithms to hierarchical multi-core clusters, as well as the extending the leader based and multi-leader based algorithms to take into account the topology of the networks. If successful, it will significantly enhance Platform MPI competitiveness to other vendor-based and open-source MPI libraries. It will also increase the company's market share, expand its business in Canada and worldwide, and establish the company at the forefront of high performance computing software. It should be mentioned that improvements in the design and implementation of the MPI library will have reduced development costs and energy savings implications for various industrial, governmental, and academic sectors in Canada that are involved in scientific and engineering innovations through high performance computing.
Programme
Engage Grants Program
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
409302-2010
Name nom
Afsahi, Ahmad
Department département
Electrical and Computer Engineering
Organization id
29
Institution établissement
Queen's University
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2010
Competition year année de concours
2010
Award amount
25000
Program id
EGP
Program name en
Engage Grants Program
Program name fr
Programme de subventions d'engagement partenariat
Group en
Research partnerships programs
Group fr
Programmes de partenariat
Committee code
1553
Committee name en
Ontario Internal Decision Committee
Committee name fr
Comité de décision interne pour l'Ontario
Area of application code
1207
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Information, computer and communication technologies
Area of application fr
Technologie de l'information, des ordinateurs et des réseaux de communications
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
Partie
1
Nb partie
1
Application title
Efficient and scalable collective communications for the Message Passing Interface
Keyword
NSERC
Application summary
High-Performance Computing (HPC) is used to tackle computationally-intensive problems such as the modeling of large genomes for cancer research, crash simulations in automotive industry, modeling of the global climate system and weather forecasting, oil and gas reservoir simulation, green energy, and molecular dynamics simulations for proteins and lipids. Processes in these simulations extensively communicate with each other through the Message Passing Interface (MPI) communication library. Such communications involve frequent MPI collective communication operations, where a group of processes collectively participate in the communication operations. Performance and scalability of collective communication operations is of paramount importance to HPC applications. Therefore, efficient collective communication algorithms have to be devised that could take advantage of the multi-core architectures, high-performance interconnects, and networking topologies in the emerging highly hierarchical HPC clusters. The purpose of this Engage project is to build a new collaboration on the development of efficient and scalable multi-core and network topology aware collective communications in the MPI library between Platform Computing and the Parallel Processing Research Laboratory at Queen's University. This project will focus on adapting the traditional collective communication algorithms to hierarchical multi-core clusters, as well as the extending the leader based and multi-leader based algorithms to take into account the topology of the networks. If successful, it will significantly enhance Platform MPI competitiveness to other vendor-based and open-source MPI libraries. It will also increase the company's market share, expand its business in Canada and worldwide, and establish the company at the forefront of high performance computing software. It should be mentioned that improvements in the design and implementation of the MPI library will have reduced development costs and energy savings implications for various industrial, governmental, and academic sectors in Canada that are involved in scientific and engineering innovations through high performance computing.