Cloud computing, with its seemingly unlimited storage capacity and computational capabilities, has become during the years the de-facto computing paradigm that ensures scalability and performance where the limited hardware of commodity devices fail. In the early days, datacenters were sparsely deployed at distant locations from the end-users and were delivering high end-to-end communication latency. However, today’s cloud datacenters have become more geographically spread, the speed and bandwidth of the networks keep increasing, hence pushing the end-users latency down. In this paper, we provide a state-of-the-art cloud reachability study as we perform extensive client-to-cloud latency measurements towards 200 datacenters deployed globally from the major cloud providers. We leverage the well- known measurements platform RIPE Atlas, involving in our study up to 12000 probes deployed in heterogeneous environments, e.g., home and offices. In order to evaluate and quantify the current state of cloud computing, we compare our latency results against three known timing thresholds, namely, human reaction time, perceivable latency and motion-to-photons. These three thresholds provide a good meter to check whether novel applications, e.g., augmented reality, can be supported by the current cloud infrastructure. We find that a good portion of the world’s population can access cloud datacenters even within the motion-to-photon, the most stringent threshold.

The Web Conference is announcing latest news and developments biweekly or on a monthly basis. We respect The General Data Protection Regulation 2016/679.