Dexter: A Performance-Cost Efficient Resource Allocation Manager for Serverless Data Analytics
Leveraging serverless platforms for the efficient execution of distributed data analytics frameworks, such as Apache Spark [3], has gained substantial interest since early 2022. The elasticity, free-of-management, and on-demand scalability of serverless have motivated the effort in deploying distributed data analytics applications to serverless platforms. However, effectively auto-scaling resources for such complex workloads so that we can fully benefit from the resource elasticity of serverless remains challenging. Mis-configuration can result in severe performance and cost issues arising from resource under- and over-provisioning.
Authors
- Claudia Misale (ORCID: https://orcid.org/0000-0003-3737-7373)
- Alaa Youssef (ORCID: https://orcid.org/0000-0001-5346-7331)
- Alberto Gutierrez-Torre (ORCID: https://orcid.org/0000-0002-5548-3359)
- Josep Ll. Berral (ORCID: https://orcid.org/0000-0003-3037-3580)
- David Carrera (ORCID: https://orcid.org/0000-0003-4898-3424)
- Diego Marrón (ORCID: https://orcid.org/0000-0002-9893-6661)
- Chen Wang (ORCID: https://orcid.org/0000-0003-0204-2362)
- Anna Maria Nestorov (ORCID: https://orcid.org/0000-0003-3233-0322)
Institutions
- Barcelona Supercomputing Center (ES)
- IBM Research - Thomas J. Watson Research Center (US)
- Universitat Politècnica de Catalunya (ES)
Publication Details
- Published
- 2024-11-27
- DOI
- https://doi.org/10.1145/3652892.3700753
- Citations
- 3
- Primary Topic
- Cloud Computing and Resource Management
- Type
- article
- Field-Weighted Citation Impact
- 2.34
Funders
- Generalitat de Catalunya
- Ministerio de Ciencia e Innovación
- Agència de Gestió d'Ajuts Universitaris i de Recerca
- European Regional Development Fund
- Agencia Estatal de Investigación