Abstract
The search for availability, reliability, and quality of service has led cloud infrastructure customers to disseminate their services, contents, and data over multiple cloud data centers, often involving several Cloud service providers (CSPs). The consequence of this is that a large amount of data must be transmitted across the public Cloud. However, little is known about the bandwidth dynamics involved. To address this, we have conducted a measurement campaign for bandwidth between 18 data centers of four major CSPs. This extensive campaign allowed us to characterize the resulting time series of bandwidth as the addition of a stationary component and some infrequent excursions (typically downtimes). While the former provides a description of the bandwidth users can expect in the Cloud, the latter is closely related to the robustness of the Cloud (i.e., the occurrence of downtimes is correlated). Both components have been studied further by applying factor analysis, specifically analysis of variance, as a mechanism to formally compare data centers’ behaviors and extract generalities. The results show that the stationary process is closely related to the data center locations and CSPs involved in transfers that, fortunately, make the Cloud more predictable and allow the set of reported measurements to be extrapolated. On the other hand, although correlation in the Cloud is low, that is, only 10% of the measured pair of paths showed some correlation, we found evidence that such correlation depends on the particular relationships between pairs of data centers with little connection to more general factors. Positively, this implies that data centers either in the same area or within the same CSP do not show qualitatively more correlation than other data centers, which eases the deployment of robust infrastructures. On the downside, this metric is scarcely generalizable and, consequently, calls for exhaustive monitoring.
- Imad Abdeljaouad, Hicham Rachidi, Sherwin Fernandes, and Ahmed Karmouch. 2010. Performance analysis of modern TCP variants: A comparison of Cubic, Compound and New Reno. In Proceedings of the IEEE Symposium on Communications. 80--83.Google Scholar
Cross Ref
- Giuseppe Aceto, Alessio Botta, Walter de Donato, and Antonio Pescapé. 2013. Cloud monitoring: A survey. Comput. Netw. 57, 9 (2013), 2093--2115. Google Scholar
Digital Library
- Sharad Agarwal, John Dunagan, Navendu Jain, Stefan Saroiu, Alec Wolman, and Harbinder Bhogan. 2010. Volley: Automated data placement for geo-distributed cloud services. In Proceedings of the USENIX Symposium on Networked Systems Design and Implementation. 17--32. Google Scholar
Digital Library
- Alexa. 2016. Top Sites in the Cloud. Retrieved from http://www.alexa.com/topsites/category/Top/Computers/Internet/Cloud_Computing.Google Scholar
- Ignacio Bermudez, Stefano Traverso, Marco Mellia, and Mauricio Munafò. 2013. Exploring the cloud from passive measurements: The Amazon AWS case. In Proceedings of the IEEE International Conference on Computer Communications (INFOCOM’13). 230--234.Google Scholar
Cross Ref
- Adrian W. Bowman and Adelchi Azzalini. 1997. Applied Smoothing Techniques for Data Analysis. Oxford University Press.Google Scholar
- Rodrigo N. Calheiros, Rajiv Ranjan, Anton Beloglazov, César A. F. De Rose, and Rajkumar Buyya. 2011. CloudSim: A toolkit for modeling and simulation of cloud computing environments and evaluation of resource provisioning algorithms. Softw.: Pract. Exper. 41, 1 (2011), 23--50. Google Scholar
Digital Library
- James C. Corbett, Jeffrey Dean, Michael Epstein, Andrew Fikes, Christopher Frost, and others. 2012. Spanner: Google’s globally-distributed database. In Proceedings of the USENIX Conference on Operating Systems Design and Implementation. 251--264. Google Scholar
Digital Library
- Mark Crovella and Balachander Krishnamurthy. 2006. Internet Measurement: Infrastructure, Traffic and Applications. John Wiley 8 Sons Inc. Google Scholar
Digital Library
- Olive Jean Dunn and Virginia A. Clark. 1974. Applied Statistics: Analysis of Variance and Regression. John Wiley 8 Sons Inc.Google Scholar
- Yuan Feng, Baochun Li, and Bo Li. 2012. Jetway: Minimizing costs on inter-datacenter video traffic. In Proceedings of the ACM Conference on Multimedia. 259--268. Google Scholar
Digital Library
- Sally Floyd and Vern Paxson. 2001. Difficulties in simulating the Internet. IEEE/ACM Trans. Netw. 9, 4 (2001), 392--403. Google Scholar
Digital Library
- Forrester Research. 2016. The Future of data center wide-area networking. Retrieved from http://www.forrester.com.Google Scholar
- José L. García-Dorado. 2015. Bandwidth in the Cloud. arXiv. Retrieved from http://arxiv.org/abs/1512.01129Google Scholar
- José L. García-Dorado, José A. Hernández, Javier Aracil, Jorge E. López de Vergara, and Sergio Lopez-Buedo. 2011. Characterization of the busy-hour traffic of IP networks based on their intrinsic features. Comput. Netw. 55, 9 (2011), 2111--2125. Google Scholar
Digital Library
- José L. García-Dorado and Sanjay G. Rao. 2015. Cost-aware multi data-center bulk transfers in the Cloud from a customer-side perspective. IEEE Trans. Cloud Comput. (2015).Google Scholar
- Gene V. Glass, Percy D. Peckham, and James R. Sanders. 1972. Consequences of failure to meet assumptions underlying the fixed effects analysis of variance and covariance. Rev. Educ. Res. 42, 3 (1972), 237--288.Google Scholar
Cross Ref
- Mohammad Hajjat, Ruiqi Liu, Yiyang Chang, T. S. Eugene Ng, and Sanjay G. Rao. 2015. Application-specific configuration selection in the cloud: Impact of provider policy and potential of systematic testing. In Proceedings of the IEEE International Conference on Computer Communications (INFOCOM’15). 873--881.Google Scholar
- High Scalability. 2016. Latency is everywhere and it costs you sales—How to crush it. Retrieved from http://highscalability.com/latency-everywhere-and-it-costs-you-sales-how-crush-it/.Google Scholar
- Manish Jain and Constantinos Dovrolis. 2002. End-to-end available bandwidth: Measurement methodology, dynamics, and relation with TCP throughput. In Proceedings of the ACM Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication (SIGCOMM’02). 295--308. Google Scholar
Digital Library
- Avinash Lakshman and Prashant Malik. 2010. Cassandra: A decentralized structured storage system. SIGOPS Operat. Syst. Rev. 44, 2 (2010), 35--40. Google Scholar
Digital Library
- Nikolaos Laoutaris, Michael Sirivianos, Xiaoyuan Yang, and Pablo Rodriguez. 2011. Inter-datacenter bulk transfers with NetStitcher. In Proceedings of the ACM Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication (SIGCOMM’11). 74--85. Google Scholar
Digital Library
- Sung-Ju Lee, Puneet Sharma, Sujata Banerjee, Sujoy Basu, and Rodrigo Fonseca. 2005. Measuring bandwidth between planetlab nodes. In Proceedings of the Passive and Active Network Measurement Conference. 292--305. Google Scholar
Digital Library
- Philipp Leitner and Jürgen Cito. 2016. Patterns in the chaos--A study of performance variation and predictability in public IaaS clouds. ACM Trans. Internet Technol. 16, 3 (2016), 15:1--15:23. Google Scholar
Digital Library
- Ang Li, Xiaowei Yang, Srikanth Kandula, and Ming Zhang. 2010. CloudCmp: Comparing public cloud providers. In Proceedings of the ACM Conference on Internet Measurement. 1--14. Google Scholar
Digital Library
- Felipe Mata, Piotr Żuraniewski, Michel Mandjes, and Marco Mellia. 2014. Anomaly detection in diurnal data. Comput. Netw. 60 (2014), 187--200. Google Scholar
Digital Library
- Norman H. Nie, Dale H. Bent, and C. Hadlai Hull. 1975. SPSS: Statistical Package for the Social Sciences. Vol. 227. McGraw-Hill.Google Scholar
- Valerio Persico, Alessio Botta, Pietro Marchetta, and Antonio Pescapé. 2017. On the performance of the wide-area networks interconnecting public-cloud datacenters around the globe. Comput. Netw. 112, C (2017), 67--83. Google Scholar
Digital Library
- Valerio Persico, Pietro Marchetta, Alessio Botta, and Antonio Pescapé. 2015a. Measuring network throughput in the cloud: The case of Amazon EC2. Comput. Netw. 93, 3 (2015), 408--422. Google Scholar
Digital Library
- Valerio Persico, Pietro Marchetta, Alessio Botta, and Antonio Pescapé. 2015b. On network throughput variability in Microsoft Azure cloud. In Proceedings of the IEEE Global Communications Conference (GLOBECOM’15). 1--6.Google Scholar
Cross Ref
- Valerio Persico, Pietro Marchetta, Alessio Botta, and Antonio Pescapé. 2016. A first look at public-cloud inter-datacenter network performance. In Proceedings of the IEEE Global Communications Conference (GLOBECOM’16). 1--7.Google Scholar
Cross Ref
- Ingmar Poese, Steve Uhlig, Mohamed Ali Kaafar, Benoit Donnet, and Bamba Gueye. 2011. IP geolocation databases: Unreliable? ACM SIGCOMM Comput. Commun. Rev. 41, 2 (2011), 53--56. Google Scholar
Digital Library
- Joel Scheuner, Jürgen Cito, Philipp Leitner, and Harald Gall. 2015. Cloud workbench: Benchmarking IaaS providers based on infrastructure-as-code. In Proceedings of the IW3C2 Conference on World Wide Web. 239--242. Google Scholar
Digital Library
- P. N. Shankaranarayanan, Ashiwan Sivakumar, Sanjay G. Rao, and Mohit Tawarmalani. 2014. Performance sensitive replication in geo-distributed cloud datastores. In Proceedings of the IEEE/IFIP Conference on Dependable Systems and Networks. 240--251. Google Scholar
Digital Library
- Ajay Tirumala, Mark Gates, Feng Qin, Jon Dugan, and Jim Ferguson. 2016. iPerf--The TCP/UDP bandwidth measurement tool. Retrieved from https://github.com/esnet/iperf. (2016).Google Scholar
- Remco van de Meent, Michel Mandjes, and Aiko Pras. 2006. Gaussian traffic everywhere? In Proceedings of the IEEE International Conference on Communications (ICC’06). 573--578.Google Scholar
Cross Ref
- Edward Walker. 2008. Benchmarking Amazon EC2 for high-performance scientific computing. Login: The Magazine of USENIX 8 SAGE 33, 5 (2008), 18--23.Google Scholar
- Guohui Wang and T. S. Eugene Ng. 2010. The impact of virtualization on network performance of Amazon EC2 data center. In Proceedings of the IEEE International Conference on Computer Communications (INFOCOM’10). 1--9. Google Scholar
Digital Library
- Timothy Wood, K. K. Ramakrishnan, Prashant Shenoy, and Jacobus van der Merwe. 2011. CloudNet: Dynamic pooling of cloud resources by live WAN migration of virtual machines. ACM SIGPLAN Not. 46, 7 (2011), 121--132. Google Scholar
Digital Library
- Zhe Wu, Michael Butkiewicz, Dorian Perkins, Ethan Katz-Bassett, and Harsha V. Madhyastha. 2013. SPANStore: Cost-effective geo-replicated storage spanning multiple cloud services. In Proceedings of the ACM Symposium on Operating Systems Principles. 292--308. Google Scholar
Digital Library
- Zhe Wu and Harsha V. Madhyastha. 2013. Understanding the latency benefits of multi-cloud webservice deployments. ACM SIGCOMM Comput. Commun. Rev. 43, 2 (2013), 13--20. Google Scholar
Digital Library
- Rostyslav Zabolotnyi, Philipp Leitner, Waldemar Hummer, and Schahram Dustdar. 2015. JCloudScale: Closing the gap between IaaS and PaaS. ACM Trans. Internet Technol. 15, 3 (2015), 10:1--10:20. Google Scholar
Digital Library
- Eyal Zohar, Israel Cidon, and Osnat (Ossi) Mokryn. 2011. The power of prediction: Cloud bandwidth and cost reduction. In Proceedings of the ACM Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication (SIGCOMM’11). 86--97. Google Scholar
Digital Library
Index Terms
Bandwidth Measurements within the Cloud: Characterizing Regular Behaviors and Correlating Downtimes
Recommendations
Iris: An Inter-cloud Resource Integration System for Elastic Cloud Data Centers
CLOSER 2014: Proceedings of the 4th International Conference on Cloud Computing and Services ScienceThis paper proposes a new cloud computing service model, Hardware as a Service (HaaS), that is based on the idea of implementing ``elastic data centers'' that provide a data center administrator with resources located at different data centers as demand ...
An inter-cloud bridge system for heterogeneous cloud platforms
Over the years, more cloud computing systems have been developed providing flexible interfaces for inter-cloud interaction. This work approaches the concept of inter-cloud by utilizing APIs, open source specifications and exposed interfaces from cloud ...
Cloud SLA relationships in multi-cloud environment: models and practices
ICCMS '17: Proceedings of the 8th International Conference on Computer Modeling and SimulationIn the past few years, cloud computing has been realized and has achieved advancement. Sincemany cloud systems and providers have been deployed in this world, various models and platforms have been developed to support multiple cloud (i.e., multi-cloud) ...






Comments