Work

CASA: A Framework for SLO- and Carbon-Aware Autoscaling and Scheduling in Serverless Cloud Computing

Year: 2024

Type: conference-paper

Abstract: Serverless cloud computing is an emerging cloud computing paradigm that can reduce user costs but has stringent performance requirements, due to the need to execute short-duration functions in a timely manner and a growing carbon footprint. Traditional carbon-reducing techniques in serverless cloud platforms such as shutting down idle containers can cause higher violation rates of service level objectives (SLOs). Conversely, traditional latency-reduction methods of prewarming containers can improve performance but increase the associated carbon footprint of serverless cloud platforms. We propose a novel carbon-and SLO-aware framework called CASA to schedule and autoscale containers in a serverless cloud cluster. Experimental results indicate that CASA reduces the operational carbon footprint of a serverless cluster by up to 2.6× while reducing the average SLO violation rate by up to 1.4× compared to the state-of-the-art.

DOI: 10.1109/igsc64514.2024.00010

Authors: Sirui Qi, Hayden Moore, Ninad Hogade, Dejan Milojicic, Cullen Bash, Sudeep Pasricha

Institutions: Colorado State University, Hewlett-Packard (United States)

Language: English

Topic: Cloud Computing and Resource Management

Subfield: Information Systems

Field: Computer Science

Domain: Physical Sciences

SDG: Climate action

Open Access status: green

Funder: National Science Foundation

Awards: EAGER: Exploring Multi-Modal Deep Learning Systems for Sustainable Connected and Autonomous Vehicles, DESC:Type I: Sustainable Serverless Computing