Grant: Collaborative Research and Development Grants
University of Calgary
$24,214.00
ICs for Communication in Distributed Computer Systems
FY 2012-13
Summary
This research targets the development of integrated circuits (ICs) to enable high-speed data transfer between computing elements such as cores, microprocessors, and memories. Its goal is in line with industry needs to increase the rate at which data is transferred between computing elements while reducing the power consumed by the communication IC in so doing. This research looks at two specific problems associated with high-speed data communications: 1.) contact-less vertical signalling between three-dimensional (3D) chip stacks and 2.) efficient wired signalling between processors arranged in a plane.
The problems of heat removal, clocking limits, an industry shift towards cloud computing, plus an increasing emphasis on the computational sophistication of mobile devices stand as major developments behind the accelerating transformation of computer systems. Over the last decade these issues have forced designers away from computers based on the classic single processor (single-core) towards structures based on multiple processors (multi-core) physically distributed in space. Today, roughly 10 cores constitute a production-level general purpose processor with many more already present in specialized graphics systems. At present rates of growth a microprocessor will consist of 1000 cores in 15 years.
To fully leverage this processing potential an effective means of moving data on-chip and between chips is needed. However this is challenged by the slow growth in available chip input/output (I/O) pins. As a result a combination of more I/O paths and the ability to signal faster over those paths is needed. By targeting communication over 3D chip stacks and increasing data rates over planar interconnect while not exceeding a strict power budget this research attempts to enable economically-feasible interconnection networks to achieve those needs.
Program
Collaborative Research and Development Grants
Recipient
Location
Alberta, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
419602-2011
Name nom
Magierowski, Sebastian
Department département
Electrical and Computer Engineering
Organization id
11
Institution établissement
University of Calgary
Province en
Alberta
Province fr
Alberta
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2012
Competition year année de concours
2011
Award amount
24214
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
803
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 communications
Area of application fr
Téléinformatique
Research subject code
2504
Research subject group en
Electrical and electronic engineering
Research subject group fr
Génie électrique et électronique
Research subject en
Integrated circuits
Research subject fr
Circuits intégrés
Partie
1
Nb partie
3
Application title
ICs for Communication in Distributed Computer Systems
Keyword
NSERC
Application summary
This research targets the development of integrated circuits (ICs) to enable high-speed data transfer between computing elements such as cores, microprocessors, and memories. Its goal is in line with industry needs to increase the rate at which data is transferred between computing elements while reducing the power consumed by the communication IC in so doing. This research looks at two specific problems associated with high-speed data communications: 1.) contact-less vertical signalling between three-dimensional (3D) chip stacks and 2.) efficient wired signalling between processors arranged in a plane.
The problems of heat removal, clocking limits, an industry shift towards cloud computing, plus an increasing emphasis on the computational sophistication of mobile devices stand as major developments behind the accelerating transformation of computer systems. Over the last decade these issues have forced designers away from computers based on the classic single processor (single-core) towards structures based on multiple processors (multi-core) physically distributed in space. Today, roughly 10 cores constitute a production-level general purpose processor with many more already present in specialized graphics systems. At present rates of growth a microprocessor will consist of 1000 cores in 15 years.
To fully leverage this processing potential an effective means of moving data on-chip and between chips is needed. However this is challenged by the slow growth in available chip input/output (I/O) pins. As a result a combination of more I/O paths and the ability to signal faster over those paths is needed. By targeting communication over 3D chip stacks and increasing data rates over planar interconnect while not exceeding a strict power budget this research attempts to enable economically-feasible interconnection networks to achieve those needs.