Grant: Collaborative Research and Development Grants
École Polytechnique de Montréal
$42,500.00
Domain specific language integration for hardware-aware software generation
EF 2015-16
Résumé
When developing simulation software, the development team has to rely on the competence of subject matter
experts (SMEs) in different areas and with expertise from a multitude of domains, such as mechanics, power
electronics, avionics, and more. In general, expecting SMEs to translate their domain knowledge into an
imperative programming language is quite unrealistic, leading to the use of domain-specific languages. When
considering modern processors, with their increasing core count, specialized programming languages and
software development paradigms have emerged, each offering different ways on how to fully exploit their
power. Some frameworks offer programming paradigms tuned for specific hardware platforms which are well
suited for software developers. Other frameworks instead provide higher-level abstractions which trade
performance for usability to cope with software complexity. The result is that it is increasingly difficult to
extract the maximum performance from the hardware without a considerable knowledge of the hardware and
expertise in parallel programming. This is particularly important when considering the use of heterogeneous
cloud computing resources: these are not fixed and the structure of the execution platform might change from
one run to the other. The purpose of this project is to define a methodology for hiding software complexity
from SMEs by automating the software generation, mapping and optimization according to inputs from SMEs
and hardware and software experts, who describe the execution platform in sufficient detail. This methodology
is applied in particular to the aerospace industry's Full Mission Simulators (FMSs), which are applications that
are developed and maintained by multiple SMEs and are usually executed on large distributed computing
systems.
The overall result of this approach will be cost-effective and interoperable software products that will give a
competitive advantage to the aerospace industry of Canada and Quebec.
Programme
Collaborative Research and Development Grants
Bénéficiaire
Emplacement
Quebec, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
446057-2012
Name nom
Nicolescu, Gabriela
Department département
Génie informatique et génie logiciel
Organization id
64
Institution établissement
École Polytechnique de Montréal
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2015
Competition year année de concours
2012
Award amount
42500
Program id
CRDPJ
Program name en
Collaborative Research and Development Grants
Program name fr
Subventions de recherche et développement coopérative
Group en
Research partnerships programs
Group fr
Programmes de partenariat
Committee code
96
Committee name en
RPP Internal Decision Cttee
Committee name fr
Cté de décision interne - PPR
Area of application code
9999
Area of application group en
Not available
Area of application group fr
Non disponible
Area of application en
Not available
Area of application fr
Non disponible
Research subject code
9999
Research subject group en
Not available
Research subject group fr
Non disponible
Research subject en
Not available
Research subject fr
Non disponible
Partie
3
Nb partie
3
Application title
Domain specific language integration for hardware-aware software generation
Keyword
NSERC
Application summary
When developing simulation software, the development team has to rely on the competence of subject matter
experts (SMEs) in different areas and with expertise from a multitude of domains, such as mechanics, power
electronics, avionics, and more. In general, expecting SMEs to translate their domain knowledge into an
imperative programming language is quite unrealistic, leading to the use of domain-specific languages. When
considering modern processors, with their increasing core count, specialized programming languages and
software development paradigms have emerged, each offering different ways on how to fully exploit their
power. Some frameworks offer programming paradigms tuned for specific hardware platforms which are well
suited for software developers. Other frameworks instead provide higher-level abstractions which trade
performance for usability to cope with software complexity. The result is that it is increasingly difficult to
extract the maximum performance from the hardware without a considerable knowledge of the hardware and
expertise in parallel programming. This is particularly important when considering the use of heterogeneous
cloud computing resources: these are not fixed and the structure of the execution platform might change from
one run to the other. The purpose of this project is to define a methodology for hiding software complexity
from SMEs by automating the software generation, mapping and optimization according to inputs from SMEs
and hardware and software experts, who describe the execution platform in sufficient detail. This methodology
is applied in particular to the aerospace industry's Full Mission Simulators (FMSs), which are applications that
are developed and maintained by multiple SMEs and are usually executed on large distributed computing
systems.
The overall result of this approach will be cost-effective and interoperable software products that will give a
competitive advantage to the aerospace industry of Canada and Quebec.