Abstract

Performance of cloud computing depends on effective utilization of resources and reliability. With resource allocation algorithms such as banker's algorithm resource utilization can be done in an effective manner in cloud computing. With reliability we can estimate the fault tolerance capability of a system. Reliability improvement is largely dependent on the availability of operational profile that statistically models the pattern in which the system is more likely to be used in the operating environment. System is less reliable if it exhibits a degree of hardware and software dependency and more reliable if hardware and software failure occur independently. In Cloud computing environment, hundreds of thousands of systems are hosted that consume cloud computing services. These services have of lots of hardware, software platform and infrastructure support, each of which though carefully engineered are still capable of failure. These failure rates and complexity of database make cloud less reliable. In this paper, we have proposed a reliability model that estimates the mean time to failure and failure rate based on delayed exponential distribution. Through this model, we study the effect of older and newer systems on cloud computing reliability that consumes the cloud computing services.

References

• Thanadesh Thanakornworakij, Raja F. Nassar, Chokchai Leangsuksun, and Mihaela P?un: A Reliability Model for Cloud Computing for High Performance Computing Applications. In Springer-Verlag Berlin Heidelberg 2013, Euro-Par 2012 Workshops, LNCS 7640, pp. 474–483, 2013.
• Marshall Marshall, A. W. , Olkin, I. : A multivariate exponential distribution. Journal of the American Statistical Association 62, 30–44 (1967).
• Schroeder, B. , Gibson, G. A. : A large-scale study of failures in high-performance compu- ting systems. In: Proceedings of International Symposium on Dependable Systems and Networks, DSN, pp. 249–258. IEEE Computer Society (2006).
• Dai, Y. S. , Yang, B. , Dongarra, J. , Zhang, G. : Cloud Service Reliability: Modeling and Analysis. In: The 15th IEEE Pacific Rim International Symposium on Dependable Com- puting (2009).
• Sujata Khatri, R. S. Chhillar, V. B. Singh: Measuring Bug Complexity in Object Oriented Software System. In ACM SIGSOFT Software Engineering Notes, volume 36 Issue 6, November 2011, pages 1-8.
• Xu, J. , Kalbarczyk, Z. , Iyer, R. K. : Networked Windows NT system field failure data analysis. In: Proceedings of the 1999 Pacific Rim International Symposium on Dependable Computing, pp. 178–185 (1999).
• Gottumukkala, N. R. , Nassar, R. , Paun, M. , Leangsuksun, C. B. , Scott, S. L. : Reliability of a System of k Nodes for High Performance Computing Applications. IEEE Transactions on Reliability 59(1), 162–169 (2010).
• Jiantao Pan, Carnegie Mellon University: software Reliability. In 18-849b Dependable Embedded Systems , Spring 1999.
• en. wikipedia. org/wiki/amazon-elastic-compute-cloud#reliability.
• Hanagal, D. D. : A multivariate Weibull distribution. Economic Quality Control 11, 193–200 (1996).
• Vishwanath, K. V. , Nagappan, N. : Characterizing Cloud Computing Hardware Reliability. In: International Conference on Management of Data, pp. 193–204 (2010).
• Yi, S. , Kondo, D. , Andrzejak, A. : Reducing Costs of Spot Instances via Check pointing in the Amazon Elastic Compute Cloud. In: IEEE Cloud Computing, CLOUD, pp. 236–243 (2010).