On the Network Performance of Amazon S3 Cloud-Storage Service

The advances in networking technologies and the increase in the need for storage resources have prompted many companies to outsource their storage needs. Cloud-storage providers offer clean and simple file-system interfaces, abstracting away the complexities of direct hardware management. At the same time, however, such services eliminate the direct oversight of performance that final users with high service-level requirements traditionally expect. While several works in literature have addressed security-related issues (such as privacy, integrity, availability, etc.) few of them have targeted the network performance of this kind of services. In this work we propose the analysis of the performance of the network associated to the storage service offered by Amazon: S3. Thanks to a large-scale distributed campaign performed by leveraging the Bismark measurement platform, we have characterized how the performance of the network may impact the quality of service experienced by final users on the basis of their location and the configuration of services. We found how performance heavily changes (up to 1553 KiB/s) according to the location of the customers and the cloud region they rely on (up to 2117 KiB/s), also deriving a number of usage guidelines for the customers. In addition we characterize the impact of leveraging the Amazon CDN service to distribute contents, finding that while it guarantees up to a 275-percent performance improvement, cases exist for which additional costs may lead to worse performance.

[1]  Gabriel Antoniu,et al.  A performance evaluation of Azure and Nimbus clouds for scientific applications , 2012, CloudCP '12.

[2]  Costin Raiciu,et al.  Opening Up Black Box Networks with CloudTalk , 2012, HotCloud.

[3]  Antonio Pescapè,et al.  A First Look at Public-Cloud Inter-Datacenter Network Performance , 2016, 2016 IEEE Global Communications Conference (GLOBECOM).

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

[5]  Lukas Kencl,et al.  CLAudit: Planetary-scale cloud latency auditing platform , 2013, 2013 IEEE 2nd International Conference on Cloud Networking (CloudNet).

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

[7]  Matei Ripeanu,et al.  Amazon S3 for science grids: a viable solution? , 2008, DADC '08.

[8]  Antonio Pescapè,et al.  CloudSurf: A platform for monitoring public-cloud networks , 2016, 2016 IEEE 2nd International Forum on Research and Technologies for Society and Industry Leveraging a better tomorrow (RTSI).

[9]  Huan Liu,et al.  Web Server Farm in the Cloud: Performance Evaluation and Dynamic Architecture , 2009, CloudCom.

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

[11]  Keqiang He,et al.  Next stop, the cloud: understanding modern web service deployment in EC2 and azure , 2013, Internet Measurement Conference.

[12]  Steve Uhlig,et al.  Let Latency Guide You: Towards Characterization of Cloud Application Performance , 2015, 2015 IEEE 7th International Conference on Cloud Computing Technology and Science (CloudCom).

[13]  Jie Li,et al.  Early observations on the performance of Windows Azure , 2010, HPDC '10.

[14]  Jorge-Arnulfo Quiané-Ruiz,et al.  Runtime measurements in the cloud , 2010, Proc. VLDB Endow..

[15]  Marco Mellia,et al.  Exploring the cloud from passive measurements: The Amazon AWS case , 2013, 2013 Proceedings IEEE INFOCOM.

[16]  Randy H. Katz,et al.  A view of cloud computing , 2010, CACM.

[17]  Alessio Botta,et al.  On Network Throughput Variability in Microsoft Azure Cloud , 2014, GLOBECOM 2014.

[18]  Simson L. Garfinkel,et al.  An Evaluation of Amazon's Grid Computing Services: EC2, S3, and SQS , 2007 .

[19]  T. S. Eugene Ng,et al.  The Impact of Virtualization on Network Performance of Amazon EC2 Data Center , 2010, 2010 Proceedings IEEE INFOCOM.

[20]  Nick Feamster,et al.  Broadband internet performance , 2011, SIGCOMM 2011.