EMARS: Efficient Management and Allocation of Resources in Serverless
2018pp. 827–830
Citations Over TimeTop 10% of 2018 papers
Abstract
We introduce EMARS, an efficient resource management system for serverless cloud computing frameworks with the goal to enhance resource (focus on memory) allocation among containers. We have built our prototype on top of an open-source serverless platform, OpenLambda. It is based upon application workloads and serverless functions' memory needs. As a background motivation we analyzed the latencies and memory requirements of functions running on AWS lambda. The memory limits also lead to variations in number of containers spawned on OpenLambda. We use memory limit settings to propose a model of predictive efficient memory management.
Related Papers
- → Grid Resource Allocation and Management Algorithm Based on Optimized Multi-task Target Decision(2019)9 cited
- → Iterative cost update method of generalized Kelly mechanism for fair utility resource allocation(2014)2 cited
- Resource allocation and load management considering the assessed quality of the provided service and the use of agent technologies(2013)
- → A note on federation management: autonomic resource allocation with economic-enhanced agents(2012)
- Resource Allocation in Project Portfolio Management: Practice in the Baltic States(2014)