Characterizing Cloud-to-User Latency as Perceived by AWS and Azure Users Spread over the Globe

With the growing adoption of cloud infrastructures to deliver a variety of IT services, monitoring cloud network performance has become crucial. However, cloud providers only disclose qualitative info about network performance, at most. This hinders efficient cloud adoption, resulting in no performance guarantees, uncertainties about the behavior of hosted services, and sub-optimal deployment choices. In this work, we focus on cloud-to-user latency, i.e. the latency of network paths interconnecting datacenters to worldwide-spread cloud users accessing their services. In detail, we performed a 14-day measurement campaign from 25 vantage points deployed via Planetlab infrastructure (emulating spatially- spread users) and considering services running in distinct locations on the infrastructures of the two most popular public-cloud providers, namely Amazon Web Services and Microsoft Azure. Our experimentation allows to provide an in-depth performance characterization (based on multiple probing methods and fine-grained sampling rate) of such networks as perceived by users spread worldwide. Results show the presence of both spatial and temporal latency trends. Finally, by evaluating the advantages of multi- cloud deployments, our results also provide useful guidelines to cloud customers.

[1]  Lucian Popa,et al.  What we talk about when we talk about cloud network performance , 2012, CCRV.

[2]  Jiye Shi,et al.  Use of Network Latency Profiling and Redundancy for Cloud Server Selection , 2014, 2014 IEEE 7th International Conference on Cloud Computing.

[3]  Lukas Kencl,et al.  Latency-Based Benchmarking of Cloud Service Providers , 2016, 2016 IEEE/ACM 9th International Conference on Utility and Cloud Computing (UCC).

[4]  Antonio Pescapè,et al.  On the performance of the wide-area networks interconnecting public-cloud datacenters around the globe , 2017, Comput. Networks.

[5]  Xiaowei Yang,et al.  CloudCmp: comparing public cloud providers , 2010, IMC '10.

[6]  Gwendal Simon,et al.  The brewing storm in cloud gaming: A measurement study on cloud to end-user latency , 2012, 2012 11th Annual Workshop on Network and Systems Support for Games (NetGames).

[7]  Antonio Pescapè,et al.  On the Network Performance of Amazon S3 Cloud-Storage Service , 2016, 2016 5th IEEE International Conference on Cloud Networking (Cloudnet).

[8]  Antonio Pescapè,et al.  Measuring network throughput in the cloud: The case of Amazon EC2 , 2015, Comput. Networks.

[9]  Lukas Kencl,et al.  Learning from Cloud latency measurements , 2015, 2015 IEEE International Conference on Communication Workshop (ICCW).

[10]  Cheng Huang,et al.  Estimating the performance of hypothetical cloud service deployments: A measurement-based approach , 2011, 2011 Proceedings IEEE INFOCOM.

[11]  Lisong Xu,et al.  Available bandwidth estimation in public clouds , 2018, IEEE INFOCOM 2018 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS).

[12]  David E. Culler,et al.  Operating Systems Support for Planetary-Scale Network Services , 2004, NSDI.

[13]  Rajiv Ranjan,et al.  CloudGenius: decision support for web server cloud migration , 2012, WWW.

[14]  Navid Nikaein,et al.  Providing Low Latency Guarantees for Slicing-Ready 5G Systems via Two-Level MAC Scheduling , 2018, IEEE Network.

[15]  Liang Zhu,et al.  The Need for End-to-End Evaluation of Cloud Availability , 2014, PAM.

[16]  Paul G. Crumley,et al.  Assessing Cloud Network Performance , 2018, 2018 IEEE International Conference on Communications (ICC).

[17]  Hua Chen,et al.  Pingmesh: A Large-Scale System for Data Center Network Latency Measurement and Analysis , 2015, SIGCOMM.

[18]  Lukas Kencl,et al.  Multidimensional cloud latency monitoring and evaluation , 2016, Comput. Networks.

[19]  Pedro Casas,et al.  Hi-Clust: Unsupervised Analysis of Cloud Latency Measurements Through Hierarchical Clustering , 2018, 2018 IEEE 7th International Conference on Cloud Networking (CloudNet).