A note on "The Need for End-to-End Evaluation of Cloud Availability"

Cloud availability is a major performance parameter for cloud platforms, but there are very few measurements on commercial platforms, and most of them rely on outage reports as appeared on specialized sites, providers' dashboards, or the general press. A paper recently presented at the PAM 2014 conference by Hu et alii reports the results of a measurement campaign. In this note, the results of that paper are summarized, highlighting sources of inaccuracy and some possible improvements. In particular, the use of a low probing frequency could lead to non detection of short outages, as well as to an inaccurate estimation of the outage duration statistics. Overcoming this lack of accuracy is relevant to properly assess SLA violations and establish the basis for insurance claims.

[1]  Stefan Savage,et al.  Measuring Online Service Availability Using Twitter , 2010, WOSN.

[2]  Rajkumar Buyya,et al.  CloudSim: a toolkit for modeling and simulation of cloud computing environments and evaluation of resource provisioning algorithms , 2011, Softw. Pract. Exp..

[3]  Loretta Mastroeni,et al.  Network protection through insurance: Premium computation for the ON-OFF service model , 2011, 2011 8th International Workshop on the Design of Reliable Communication Networks (DRCN).

[4]  Rajkumar Buyya,et al.  Towards autonomic detection of SLA violations in Cloud infrastructures , 2012, Future Gener. Comput. Syst..

[5]  David E. Culler,et al.  PlanetLab: an overlay testbed for broad-coverage services , 2003, CCRV.

[6]  Emiliano Casalicchio,et al.  Mechanisms for SLA provisioning in cloud-based service providers , 2013, Comput. Networks.

[7]  Van-Anh Truong,et al.  Availability in Globally Distributed Storage Systems , 2010, OSDI.

[8]  Liam O'Brien,et al.  On a Catalogue of Metrics for Evaluating Commercial Cloud Services , 2012, 2012 ACM/IEEE 13th International Conference on Grid Computing.

[9]  Salvatore Venticinque,et al.  An SLA-based Broker for Cloud Infrastructures , 2013, Journal of Grid Computing.

[10]  Loretta Mastroeni,et al.  Violation of service availability targets in service level agreements , 2011, 2011 Federated Conference on Computer Science and Information Systems (FedCSIS).

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

[12]  Maurizio Naldi,et al.  The availability of cloud-based services: Is it living up to its promise? , 2013, 2013 9th International Conference on the Design of Reliable Communication Networks (DRCN).

[13]  Ju Wang,et al.  Windows Azure Storage: a highly available cloud storage service with strong consistency , 2011, SOSP.

[14]  Jesús Montes,et al.  GMonE: A complete approach to cloud monitoring , 2013, Future Gener. Comput. Syst..

[15]  Jack Dongarra,et al.  1 Cloud Service Reliability : Modeling and Analysis , 2010 .

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

[17]  Kashi Venkatesh Vishwanath,et al.  Characterizing cloud computing hardware reliability , 2010, SoCC '10.

[18]  Ilkka Norros,et al.  Towards risk-aware communications networking , 2013, Reliab. Eng. Syst. Saf..

[19]  Loretta Mastroeni,et al.  Compensation Policies and Risk in Service Level Agreements: A Value-at-Risk Approach under the ON-OFF Service Model , 2011, ICQT.

[20]  Lee Gillam,et al.  Adding Cloud Performance to Service Level Agreements , 2012, CLOSER.