Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

University of Waterloo

$36,000.00

Building Low-Latency Serverless Platforms with Predictable Performance

FY 2023-24

Summary
Serverless computing is an emerging execution model that offers significant operational and cost benefits over traditional cloud computing models. These benefits include elastic scaling, simplified pricing, and deployments that do not require any server or virtual machine management. With these advantages, we should expect large-scale migrations of existing cloud applications to serverless computing platforms. Yet adoption has been slow, and the most successful serverless applications are simple and latency-insensitive with limited communication between functions. This is in large part due to the high latency and unpredictable performance offered by current serverless platforms.  The overarching goal of my research program is to understand the root causes of these performance problems and to design new serverless platforms and infrastructure services that provide low latency and predictable performance. With predictable performance, serverless providers can offer performance-based Service Level Agreements (SLAs) to applications, which can enable new classes of applications to be deployed on serverless platforms. Providing low-latency and predictable performance requires improvements to every major layer in the serverless platform stack. This includes new resource scheduling and provisioning strategies that reduce the worst case latency to improve predictability, elastic network resource allocation to prevent slowdowns due to network congestion, and infrastructure services with dynamic durability guarantees that can offer different tiers of predictable performance. By solving the performance problems at each layer, we can create serverless platforms that offer latency and performance predictability that are similar to traditional cloud computing models without needing explicit resource reservations.
Program
Discovery Grants Program - Individual
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
RGPIN-2023-03781
Name nom
Wong, Bernard
Department département
David R. Cheriton School of Computer Science
Organization id
33
Institution établissement
University of Waterloo
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2023
Competition year année de concours
2023
Award amount
36000
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
801
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 software
Area of application fr
Logiciel
Research subject code
9999
Research subject group en
Not available
Research subject group fr
Not available
Research subject en
Non disponible
Research subject fr
Non disponible
Application title
Building Low-Latency Serverless Platforms with Predictable Performance
Keyword
Cloud computing; Cloud infrastructure; Predictable performance; Resource scheduling; Serverless computing; Service level agreements; Virtual network embedding
Application summary
Serverless computing is an emerging execution model that offers significant operational and cost benefits over traditional cloud computing models. These benefits include elastic scaling, simplified pricing, and deployments that do not require any server or virtual machine management. With these advantages, we should expect large-scale migrations of existing cloud applications to serverless computing platforms. Yet adoption has been slow, and the most successful serverless applications are simple and latency-insensitive with limited communication between functions. This is in large part due to the high latency and unpredictable performance offered by current serverless platforms.  The overarching goal of my research program is to understand the root causes of these performance problems and to design new serverless platforms and infrastructure services that provide low latency and predictable performance. With predictable performance, serverless providers can offer performance-based Service Level Agreements (SLAs) to applications, which can enable new classes of applications to be deployed on serverless platforms. Providing low-latency and predictable performance requires improvements to every major layer in the serverless platform stack. This includes new resource scheduling and provisioning strategies that reduce the worst case latency to improve predictability, elastic network resource allocation to prevent slowdowns due to network congestion, and infrastructure services with dynamic durability guarantees that can offer different tiers of predictable performance. By solving the performance problems at each layer, we can create serverless platforms that offer latency and performance predictability that are similar to traditional cloud computing models without needing explicit resource reservations.