Reliability Analysis of Power System Employed by High Power Consumption Data Center

Abstract

In this paper, we analyze the reliability of a power system employed in large data centers with high power demand. As an emergency response system, it is crucial to ensure that the data center retains its functionality in the event of an emergency. This assessment aims to have a seamless transition with high reliability. Continuous Markov process (CMP) and Monte Carlo simulation (MCS) provide an exponential index of reliability based on the number of nines in the result. The model under test uses a four-state matrix/equation based on the availability and failure states for CMP and MCS. Finally, a discussion provide insight in how these two methods achieve similar results with different mathematical approaches.