Performance evaluation of a SaaS cloud under different levels of workload computational demand variability and tardiness bounds

摘要

As the paradigm shift toward Software as a Service (SaaS) continues to gain momentum, there is a growing focus on the performance of SaaS clouds from both academia and industry. In this paper, we evaluate the performance of a SaaS cloud under various tardiness bounds and different levels of workload computational demand variability. The workload consists of bag-of-tasks jobs, which are scheduled on the underlying virtualized host environment. The jobs have soft deadlines and different levels of variability in their computational demands. A multi-tier SLA is employed, which imposes a soft and a hard tardiness bound on each job. Furthermore, the employed pricing scheme is based on the provided level of Quality of Service (QoS). The performance of the SaaS cloud is evaluated by simulation, in an attempt to shed light on how it is affected by the tardiness bound and the computational demand variability of the workload.