Grant: Discovery Grants Program - Individual
University of Victoria
$48,000.00
Augmenting Network Observability with Multi-task Learning
EF 2024-25
Résumé
Network observability refers to the ability to quickly understand your network status, performance, and behaviour in real time, facilitating timely decision-making in network management. With the growing complexity of networks, especially with new network elements such as software-defined networks (SDN), cloud computing, edge computing, and IoT devices, network observability is a paramount concern in network management and engineering. Direct measurement of every relevant network segment often proves impossible due to physical or administrative constraints. Hence, there's a pressing need for innovative network tomography (NT) techniques that deduce network activities from observations of a limited number of monitoring points.
This NSERC Discovery proposal seeks to elevate network observability by addressing multiple unresolved research challenges within NT. We will investigate two types of NT problems: (1) gleaning insights about internal network behaviours, e.g., link delay, from end-to-end network measurements, e.g., path delay, and (2) extrapolating end-to-end metrics, e.g., the traffic volume between any pair of nodes, using minimal internal measurements, e.g., link loads. Our approach hinges on multi-task learning (MTL), a machine learning subset dedicated to concurrently solving interrelated tasks. By tapping into the synergies and variations between tasks, MTL offers a holistic perspective of network operations. This simultaneous learning of multiple network facets - delay, throughput, reliability, and more - allows MTL to discern intricate patterns and relationships that might go unnoticed when tasks are learned in isolation.
The potential benefits of bolstering network observability are vast. Canadian networking giants like Telus and Rogers can enhance operational efficiency, security, and overall business performance. Direct outcomes include improved network performance and a drastic reduction in downtime, as real-time oversight leads to swifter problem identification and resolution. This, in turn, propels productivity and cost-effectiveness. Furthermore, a clear view of network behaviour significantly elevates security by facilitating the prompt detection of and response to potential threats. With industries gravitating towards digital transformation, IoT, and cloud computing, heightened observability becomes indispensable for managing intricate IT networks. Academically, this proposal aligns with the emerging trend of AI for networks, representing the most challenging and rewarding research in intelligent network operation. The program's success could position Canada at the forefront of AI-driven network observability research. Plus, this research will foster the development of a diverse set of Highly Qualified Personnel (HQP), reinforcing Equity, Diversity, and Inclusion (EDI) within Computer Science and Engineering.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
British Columbia, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
RGPIN-2024-03741
Name nom
Wu, Kui
Department département
Head Office
Organization id
7
Institution établissement
University of Victoria
Province en
British Columbia
Province fr
Colombie-Britannique
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
48000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1507
Committee name en
Computer Science
Committee name fr
Sciences informatiques
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
2721
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Multimedia systems and networks
Research subject fr
Systèmes et réseaux multimédias
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
Augmenting Network Observability with Multi-task Learning
Keywords
network observability; multi-task learning; network performance; network tomography; network telemetry
Application summary
Network observability refers to the ability to quickly understand your network status, performance, and behaviour in real time, facilitating timely decision-making in network management. With the growing complexity of networks, especially with new network elements such as software-defined networks (SDN), cloud computing, edge computing, and IoT devices, network observability is a paramount concern in network management and engineering. Direct measurement of every relevant network segment often proves impossible due to physical or administrative constraints. Hence, there's a pressing need for innovative network tomography (NT) techniques that deduce network activities from observations of a limited number of monitoring points.
This NSERC Discovery proposal seeks to elevate network observability by addressing multiple unresolved research challenges within NT. We will investigate two types of NT problems: (1) gleaning insights about internal network behaviours, e.g., link delay, from end-to-end network measurements, e.g., path delay, and (2) extrapolating end-to-end metrics, e.g., the traffic volume between any pair of nodes, using minimal internal measurements, e.g., link loads. Our approach hinges on multi-task learning (MTL), a machine learning subset dedicated to concurrently solving interrelated tasks. By tapping into the synergies and variations between tasks, MTL offers a holistic perspective of network operations. This simultaneous learning of multiple network facets - delay, throughput, reliability, and more - allows MTL to discern intricate patterns and relationships that might go unnoticed when tasks are learned in isolation.
The potential benefits of bolstering network observability are vast. Canadian networking giants like Telus and Rogers can enhance operational efficiency, security, and overall business performance. Direct outcomes include improved network performance and a drastic reduction in downtime, as real-time oversight leads to swifter problem identification and resolution. This, in turn, propels productivity and cost-effectiveness. Furthermore, a clear view of network behaviour significantly elevates security by facilitating the prompt detection of and response to potential threats. With industries gravitating towards digital transformation, IoT, and cloud computing, heightened observability becomes indispensable for managing intricate IT networks. Academically, this proposal aligns with the emerging trend of AI for networks, representing the most challenging and rewarding research in intelligent network operation. The program's success could position Canada at the forefront of AI-driven network observability research. Plus, this research will foster the development of a diverse set of Highly Qualified Personnel (HQP), reinforcing Equity, Diversity, and Inclusion (EDI) within Computer Science and Engineering.