Deploying Distributed Cloud Infrastructures: Who and at What Cost?

摘要

Academics and industry experts are now advocating for going from large-centralized Cloud Computing (CC) infrastructures to smaller ones massively distributed at the edge of the network. Referred to as "fog/edge/local computing", such a dawning paradigm is attracting growing interest as it improves the whole services agility in addition to bringing computing resources closer to end-users. While several initiatives investigate how such Distributed Cloud Computing (DCC) infrastructures can be operated, the economical viability of such solutions is still questionable, especially if the objective is to propose attractive prices in comparison to those proposed by giant actors such as Amazon, Microsoft and Google. In this article, we go beyond the state of the art of the current cost model of DCC infrastructures. First, we provide a classification of the different ways of deploying DCC platforms. Then, we propose a versatile cost model that can help new actors evaluate the viability of deploying a DCC solution. We illustrate the relevance of our proposal by instantiating it over three use-cases and compare them according to similar computation capabilities provided by the AWS solution. Such a study clearly shows that deploying a DCC infrastructure makes sense for telecom operators as well as new actors willing to enter the game.