"TaScaaS: A Multi-Tenant Serverless Task Scheduler and Load Balancer as a Service" is a new scientific work published in cooperation with the AI-SPRINT project. The work hightights the benefits brought by TaScaaS, a service that can endure workload distribution across multiple infrastructures.
Paper Overview
- Highlights the need for a service capable to handle workload distribution across multiple infrastructures to mitigate unpredictable performance fluctuations.
- Presents TaScaaS, a highly scalable and completely serverless service deployed on AWS to distribute loosely coupled jobs among several computing infrastructures, and load balances them using a completely asynchronous approach to cope with the performance fluctuations with minimum impact in the execution time.
- Shows how TaScaaS is not only capable of handling fluctuations efficiently, achieving a reduction in execution times up to 45% in the experiments, but can split the jobs to be computed to meet the user-defined execution time.
Authors
- Vicent Gimenez-Alventosa, Universitat Politècnica de València, Spain
- German Molto, Universitat Politècnica de València, Spain
- Damian Segrelles, Universitat Politècnica de València, Spain
Keywords
Cloud computing, heterogeneous computing, load balance, serverless