Abstract
These days Infrastructure-as-a-Service (IaaS) clouds attract more and more customers for their flexibility and scalability. Even critical data and applications are considered for cloud deployments. However, this demands for resilient cloud infrastructures. Hence, this paper approaches adaptive and scalable high availability for IaaS clouds. First, we provide a detailed failure analysis of OpenStack showing the impact of failing services to the cloud infrastructure. Second, we analyse existing approaches for making OpenStack highly available, and pinpoint several weaknesses of current best practice. Finally, we propose and evaluate improvements to automate failover mechanisms. © 2014 IFIP Internatinal Federation for Information Processing.
Author supplied keywords
Cite
CITATION STYLE
Brenner, S., Garbers, B., & Kapitza, R. (2014). Adaptive and scalable high availability for infrastructure clouds. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8460 LNCS, pp. 16–30). Springer Verlag. https://doi.org/10.1007/978-3-662-43352-2_2
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
