Grant: Engage Grants Program
University of British Columbia
$22,750.00
Leveraging dynamic co-processor platforms for high-performance applications
EF 2011-12
Résumé
Secodix Corporation has developed a dynamic co-processor for accelerating signal processing and other
high-performance computations. However, legacy applications that want to use the co-processor need to be
rewritten to use a special library. This imposes a significant burden on the part of the programmer. The main
goal of this proposal is to explore the feasibility of automatically transforming legacy applications to use the
library without requiring any support from the programmer. In this proposal, we will establish the principles
and guidelines for performing such transformations automatically, with the goal of exploring compiler support
in the future. In so doing, we will also develop a performance model for estimating the potential speedup of the
application on the platform, and come up with heuristics to optimally choose a sequence of library calls.
Based on the results of this project, Secodix will be able to seamlessly port legacy codes to their platform, thus
significantly expanding their applicability and reach. In turn, this will lead to enhanced speedups for many
important classes of high-performance applications, which are used in a wide range of critical domains. The
students trained in this project will acquire significant expertise in transforming applications to co-processor
platforms, which is an important skill as the industry embraces co-processors in the near future.
Programme
Engage Grants Program
Bénéficiaire
Emplacement
British Columbia, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
417722-2011
Name nom
Pattabiraman, Karthik
Department département
Electrical and Computer Engineering
Organization id
2
Institution établissement
University of British Columbia
Province en
British Columbia
Province fr
Colombie-Britannique
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2011
Competition year année de concours
2011
Award amount
22750
Program id
EGP
Program nane 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
1554
Committee name en
Pacific Internal Decision Committee
Committee name fr
Comité de décision interne pour le Pacifique
Area of application code
801
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Computer software
Area of application fr
Logiciel
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
Leveraging dynamic co-processor platforms for high-performance applications
Keyword
NSERC
Application summary
Secodix Corporation has developed a dynamic co-processor for accelerating signal processing and other
high-performance computations. However, legacy applications that want to use the co-processor need to be
rewritten to use a special library. This imposes a significant burden on the part of the programmer. The main
goal of this proposal is to explore the feasibility of automatically transforming legacy applications to use the
library without requiring any support from the programmer. In this proposal, we will establish the principles
and guidelines for performing such transformations automatically, with the goal of exploring compiler support
in the future. In so doing, we will also develop a performance model for estimating the potential speedup of the
application on the platform, and come up with heuristics to optimally choose a sequence of library calls.
Based on the results of this project, Secodix will be able to seamlessly port legacy codes to their platform, thus
significantly expanding their applicability and reach. In turn, this will lead to enhanced speedups for many
important classes of high-performance applications, which are used in a wide range of critical domains. The
students trained in this project will acquire significant expertise in transforming applications to co-processor
platforms, which is an important skill as the industry embraces co-processors in the near future.