作者: Praneeth Sakhamuri , Olivia Das
DOI: 10.1109/HASE.2017.16
关键词:
摘要: Deploying and managing a high availability tiered application in the cloud is challenging task because it requires determining buying enough number of VMs dynamically such that available. An available if working can respond timely manner for varying workloads. For given workload, we will need minimum copies each server computing power necessary to run those meeting response time requirement. Otherwise, end up with failures. In this work, assume software an replicated into one or more copy runs on separate virtual machine (VM). be different types depending their power, cost. This paper presents novel optimization model determine needed minimizes cost at same guarantees SLA (service-level agreement). The results demonstrate effective have mixture running rather than restricting single type VMs. further decision buy only cheapest not always better cost-wise.