How to Improve Cloud Services Availability? Investigating the Impact of Power and It Subsystems Failures

Abstract

The cloud data center is a complex system composed of power, cooling, and IT subsystems. The power subsystem is crucial to feed the IT equipment. Power disruptions may result in service unavailability. This paper analyzes the impact of the power subsystem failures on IT services regarding different architecture configurations based on TIA-942 standard such as non-redundant, redundant, concurrently maintainable, and fault tolerant. We model both subsystems, power and IT, through Stochastic Petri Net (SPN). The availability results show that a fault tolerant power and IT configuration reduces the downtime from 54.1 to 34.5 hours/year when compared to a non-redundant architecture. The sensibility analysis results show that the failure and repair rates of the server component in a fault tolerant system present the highest impact on overall data center availability.