Grant: Discovery Launch Supplement
University of Calgary
$12,500.00
Resilient and Sustainable Federated Learning in Cloud-Edge Service Orchestration
EF 2024-25
Résumé
Cloud-edge service orchestration is an important and challenging research area, important because it is the foundation of modern services in every aspect of our daily life, including network function virtualization, Internet of things (IoT) services, energy and utilities, and healthcare; yet challenging because of the complex nature of the layered infrastructure managed by public and private cloud-edge service providers with significant discrepancy in their availability and computing capability, leading to severe data and system heterogeneity.
A growing number of countries are enforcing privacy preservation laws with the rise of artificial intelligence (AI) extracting knowledge from massive volume of private data. Federated learning (FL), as a privacy-preserving distributed machine learning method, poses a promising distributed paradigm for performing on-premise learning without the need to transmit user data, drawing attention from both academia and industry in recent years. System heterogeneity with intermittent availability, combined with heterogeneous data on participants, remain key challenges for applying FL in real-world applications.
With FL systems gaining wider adoption for privacy-preserving machine learning in such mixture of infrastructure, the heterogeneity is expected to cause lowered performance of the trained models with longer convergence time, leading to excessive consumption of energy for both the cloud infrastructure and battery-powered edge IoT devices. Can carefully designed FL methods guide cloud-edge service orchestration to tackle these challenges? My long-term vision is to develop resilient, sustainable, and privacy-preserving distributed machine learning methods with the outlook of wider adoption of learning from distributed data repositories. Inspired by the challenge of computing node availability, this proposed program will answer the research question with three major new components that distinguish from existing studies: novel models that dynamically budget FL client participation based on availability and carbon footprints from FL, innovative algorithms for FL task replication with minimal impact of data privacy, and an original prototype service orchestration scheduler for evaluating the proposed methods in actual cloud-edge service orchestration.
This proposed program is important because it will address the fundamental requirement in the next generation of cloud and edge service orchestration from both the infrastructure and client participation levels. It will lead to high quality integrated methodologies with great potential for real-world implementation. Canadians will benefit from the resulting FL models and algorithms to address service orchestration challenges for privacy-preserving distributed machine learning in cloud and edge computing environments. This proposed program will also foster my long-term vision of achieving carbon neutrality for machine learning with cloud-edge infrastructure.
Programme
Discovery Launch Supplement
Bénéficiaire
Emplacement
Alberta, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
DGECR-2024-00511
Name nom
Drew, Steve
Department département
Department of Electrical and Software 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
2024
Competition year année de concours
2024
Award amount
12500
Program id
DGECR
Program name en
Discovery Launch Supplement
Program name fr
Supplément Tremplin vers la découverte
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1510
Committee name en
Electrical and Computer Engineering
Committee name fr
Génie électrique et informatique
Area of application code
802
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Information systems and technology
Area of application fr
Systèmes d'information et technologie de l'information
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
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Resilient and Sustainable Federated Learning in Cloud-Edge Service Orchestration
Keywords
service orchestration; federated learning; resource optimization; green computing; network function virtualization
Application summary
Cloud-edge service orchestration is an important and challenging research area, important because it is the foundation of modern services in every aspect of our daily life, including network function virtualization, Internet of things (IoT) services, energy and utilities, and healthcare; yet challenging because of the complex nature of the layered infrastructure managed by public and private cloud-edge service providers with significant discrepancy in their availability and computing capability, leading to severe data and system heterogeneity.
A growing number of countries are enforcing privacy preservation laws with the rise of artificial intelligence (AI) extracting knowledge from massive volume of private data. Federated learning (FL), as a privacy-preserving distributed machine learning method, poses a promising distributed paradigm for performing on-premise learning without the need to transmit user data, drawing attention from both academia and industry in recent years. System heterogeneity with intermittent availability, combined with heterogeneous data on participants, remain key challenges for applying FL in real-world applications.
With FL systems gaining wider adoption for privacy-preserving machine learning in such mixture of infrastructure, the heterogeneity is expected to cause lowered performance of the trained models with longer convergence time, leading to excessive consumption of energy for both the cloud infrastructure and battery-powered edge IoT devices. Can carefully designed FL methods guide cloud-edge service orchestration to tackle these challenges? My long-term vision is to develop resilient, sustainable, and privacy-preserving distributed machine learning methods with the outlook of wider adoption of learning from distributed data repositories. Inspired by the challenge of computing node availability, this proposed program will answer the research question with three major new components that distinguish from existing studies: novel models that dynamically budget FL client participation based on availability and carbon footprints from FL, innovative algorithms for FL task replication with minimal impact of data privacy, and an original prototype service orchestration scheduler for evaluating the proposed methods in actual cloud-edge service orchestration.
This proposed program is important because it will address the fundamental requirement in the next generation of cloud and edge service orchestration from both the infrastructure and client participation levels. It will lead to high quality integrated methodologies with great potential for real-world implementation. Canadians will benefit from the resulting FL models and algorithms to address service orchestration challenges for privacy-preserving distributed machine learning in cloud and edge computing environments. This proposed program will also foster my long-term vision of achieving carbon neutrality for machine learning with cloud-edge infrastructure.