Measuring the impact of data center failures on a cloud-based emergency medical call system

摘要

Emergency call services are expected to be highly available in order to minimize the loss ofurgent calls and, as a consequence, minimize loss of life due to lack of timely medical response.This service availability depends heavily on the cloud data center on which it is hosted. However,availability information alone cannot provide sufficient understanding of how failures impact theservice and users' perception. In this paper, we evaluate the impact of failures on an emergencycall system, considering service-levelmetrics such as the number of affected calls per failure andthe time an emergency service takes until it recovers from a failure. We analyze a real data setfrom an emergency call center for a large Brazilian city. From stochastic models that representa cloud data center, we evaluate different data center architectures to observe the impact offailures on the emergency call service. Results show that changing data center's architecturein order to improve availability from two to three nines cannot decrease the average numberof affected calls per failure. On the other hand, it can decrease the probability to affect aconsiderable number of calls at the same time.