Grant: Engage Grants Program
University of Toronto
$24,940.00
Applications exploiting heterogeneous processor architectures for improved throughput and energy-efficiency
FY 2010-11
Summary
High transistor density in deep sub-micron ICs has created a ""power wall"", limiting the rate of clock frequency
scaling in general purpose microprocessors. The rate of increase in processor speed has tapered off in recent
years and the two main microprocessor vendors, Intel and AMD, now incorporate multiple processor cores on a
single chip to improve performance, rather than a single core with increased speed and/or complexity. Today's
multi-core processors are homogeneous in the sense that they comprise multiple copies of a single type of
processor core. Future multi-core processors will contain billions of transistors and will be heterogeneous,
comprising multiple different types of cores - each type of core being tailored for certain applications.
AMD has recently announced their Fusion architecture which integrates a graphics processing unit (GPU) on
the same chip as a general-purpose processor. Graphics processors excel at executing mathematical
computations in parallel, such as those needed for displaying advanced computer graphics. Recently, software
developers have begun exploiting GPUs for general-purpose (non-graphics) applications, notably scientific
computing applications. With the close-coupling of GPUs and CPUs on a single chip, new opportunities exist
for application development and architectural enhancements. This NSERC Engage project takes the first steps
towards creating new applications and architectures for next generation heterogeneous multi-core processors.
Applications that contain graphics-oriented and general purpose computations will be created. A framework
will be developed for simulating the execution of such applications on future heterogeneous architectures, with
the goal of permitting early evaluation of processor speed, power consumption and cost.
Program
Engage Grants Program
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
408126-2010
Name nom
EnrightJerger, Natalie
Department département
Electrical and Computer Engineering, Edward S. Rogers Sr. Department of
Organization id
31
Institution établissement
University of Toronto
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
24940
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
2719
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Computer architecture and design
Research subject fr
Architecture d'ordinateur et conception
Partie
1
Nb partie
1
Application title
Applications exploiting heterogeneous processor architectures for improved throughput and energy-efficiency
Keyword
NSERC
Application summary
High transistor density in deep sub-micron ICs has created a ""power wall"", limiting the rate of clock frequency
scaling in general purpose microprocessors. The rate of increase in processor speed has tapered off in recent
years and the two main microprocessor vendors, Intel and AMD, now incorporate multiple processor cores on a
single chip to improve performance, rather than a single core with increased speed and/or complexity. Today's
multi-core processors are homogeneous in the sense that they comprise multiple copies of a single type of
processor core. Future multi-core processors will contain billions of transistors and will be heterogeneous,
comprising multiple different types of cores - each type of core being tailored for certain applications.
AMD has recently announced their Fusion architecture which integrates a graphics processing unit (GPU) on
the same chip as a general-purpose processor. Graphics processors excel at executing mathematical
computations in parallel, such as those needed for displaying advanced computer graphics. Recently, software
developers have begun exploiting GPUs for general-purpose (non-graphics) applications, notably scientific
computing applications. With the close-coupling of GPUs and CPUs on a single chip, new opportunities exist
for application development and architectural enhancements. This NSERC Engage project takes the first steps
towards creating new applications and architectures for next generation heterogeneous multi-core processors.
Applications that contain graphics-oriented and general purpose computations will be created. A framework
will be developed for simulating the execution of such applications on future heterogeneous architectures, with
the goal of permitting early evaluation of processor speed, power consumption and cost.