Improving Reliability of Cloud-Based Applications
With the increasing availability of various types of cloud services many organizations are becoming reliant on providers of cloud services to maintain the operation of their enterprise applications. Different types of reliability strategies designed to im
- PDF / 5,652,982 Bytes
- 13 Pages / 439.37 x 666.142 pts Page_size
- 6 Downloads / 241 Views
Faculty of Engineering and Information Technology, University of Technology, Sydney, Sydney, Australia {hongthai.tran,george.feuerlicht}@uts.edu.au 2 Unicorn College, V Kapslovně 2767/2, 130 00 Prague 3, Czech Republic 3 Department of Information Technology, University of Economics, Prague, W. Churchill Sq. 4, Prague 3, Czech Republic
Abstract. With the increasing availability of various types of cloud services many organizations are becoming reliant on providers of cloud services to maintain the operation of their enterprise applications. Different types of reliability strategies designed to improve the availability of cloud services have been proposed and implemented. In this paper we have estimated the theoretical improvements in service availability that can be achieved using the Retry Fault Tolerance, Recovery Block Fault Tolerance and Dynamic Sequential Fault Tolerance strategies, and we have compared these estimates to experimentally obtained results. The experimental results obtained using our prototype Service Consumer Framework are consistent with the theoretical predictions, and indicate significant improvements in service availability when compared to invoking cloud services directly. Keywords: Reliability of cloud services DSFT
Fault tolerance
RFT
RBFT
1 Introduction With the increasing use of cloud services the reliability of enterprise applications is becoming dependent on the reliability of consumed cloud services. In the public cloud context, service consumers do not have control over externally provided cloud services and therefore cannot guarantee the levels of security and availability that they are typically expected to provide to their users [1]. While most cloud service providers make considerable efforts to ensure the reliability of their services, cloud service consumers cannot assume continuous availability of cloud services, and are ultimately responsible for the reliable operation of their enterprise applications. In response to such concerns, hybrid cloud solutions have become popular [2]; according to Gartner Special Report on the Outlook for Cloud [3] half of large enterprises will adopt and use the hybrid cloud model by the end of 2017. Hybrid cloud solutions involve on-premise enterprise applications that utilize external cloud services, for example Paypal Payment Gateway (www.paypal.com), cloud storage service Amazon S3 (aws.amazon.com/s3), or entire SaaS (Software as a Service) applications. With a hybrid delivery model © IFIP International Federation for Information Processing 2016 Published by Springer International Publishing Switzerland 2016. All Rights Reserved M. Aiello et al. (Eds.): ESOCC 2016, LNCS 9846, pp. 235–247, 2016. DOI: 10.1007/978-3-319-44482-6_15
236
H.T. Tran and G. Feuerlicht
where enterprise applications are partially hosted on premise and partially in the cloud, enterprises can balance the benefits and drawbacks of both approaches, and decide which applications can be migrated to the cloud and which should be deployed locally to ensure hig
Data Loading...
