Models for dependability and sustainability analysis of data center cooling architectures

The advent of cloud computing, social networks and e-commerce have demanded more computational resources from data centers. Prominent issues are sustainability, cost and dependability, which are significantly impacted by the redundant infrastructures needed to support those services. In this context, models are important to support data center designers to estimate previous issues before implementing the final architecture. This paper proposes a set of models for the integrated quantification of sustainability impact, cost and dependability of data center cooling infrastructures with the support of an environment, namely, ASTRO. Besides, a model is proposed for verifying if the energy flow does not exceed the maximum energy capacity that each component can provide (considering power devices) or the maximum cooling capacity (assuming cooling equipment). Additionally, this work presents a case study which analyzes the environmental impact, dependability metrics as well as the acquisition and operational costs of five real-world data center cooling architectures.