Auto-scaling microservices on IaaS under SLA with cost-effective framework
Conference proceedings article
Authors/Editors
Strategic Research Themes
No matching items found.
Publication Details
Author list: Prachitmutita I., Aittinonmongkol W., Pojjanasuksakul N., Supattatham M., Padungweang P.
Publisher: Hindawi
Publication year: 2018
Start page: 583
End page: 588
Number of pages: 6
ISBN: 9781538643624
ISSN: 0146-9428
eISSN: 1745-4557
Languages: English-Great Britain (EN-GB)
Abstract
Nowadays, the application usage continuously fluctuates depending on the behavior of users that affect the number of request to each service behind the application. Moreover, application development has started to change from a Monolithic architecture to a Micro services architecture for many services. Thus, it is hard for the administrators to maintain each service as per the Service Level Agreement (SLA) in a cost effective way. This paper proposes a new auto-scaling framework based on predicted workload, with artificial neural network, recurrent neural network and resource scaling optimization algorithm to create an automated system for managing the whole application via scale-out / in with Infrastructure as a Service (IaaS). The experimental result of each module is evaluated with real workload history - FIFA World Cup 98 website. Results show that our framework can automatically scale server in advance in order to guarantee services under SLA and have appropriate cost of Infrastructure. ฉ 2018 IEEE.
Keywords
Auto-scaling, IaaS, Machine-Learning, MicroSerivce, Predictable
Contact Information