A novel mechanism for the reduction of latency and cost in AWS Lambda calls

Kokkoori Anilkumar, Anilgovind (2024) A novel mechanism for the reduction of latency and cost in AWS Lambda calls. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (963kB) | Preview

Abstract

Serverless computing, especially AWS lambda revolutionised software industry. But cold start is a significant challenge, especially for latency-critical and sporadic workloads. This research addresses this issue by proposing and evaluating novel architectures. A light weight load balancer for reducing cold start and architectures for reducing cold start were implemented in this research. We analyzed multiple strategies for minimizing cold starts, including invocation-based methods using AWS EventBridge, SNS, and CloudWatch. Among these, a CloudWatchdriven architecture demonstrated superior efficiency by selectively invoking inactive Lambdas, avoiding unnecessary overhead. Additionally, the custom load balancer consistently outperformed the AWS API Gateway in test scenarios. The solution completely eliminated cold starts. It reduced latency by up to 80% and costs by 20% compared to using API Gateway for sporadic test events. The findings have broader implications for designing responsive, cost-effective serverless applications. And the solution can adapt for cross-platform deployments to unlock its full commercial potential like vendor locking.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Estrada, Giovani UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software T Technology > T Technology (General) > Information Technology > Computer software T Technology > T Technology (General) > Information Technology > Cloud computing
Divisions:	School of Computing > Master of Science in Cloud Computing
Depositing User:	Ciara O'Brien
Date Deposited:	15 Jul 2025 13:30
Last Modified:	15 Jul 2025 13:30
URI:	https://norma.ncirl.ie/id/eprint/8113

Actions (login required)

View Item