A Runtime Model Based Monitoring Approach for Cloud
CLOUD '10 Proceedings of the 2010 IEEE 3rd International Conference on Cloud Computing
CloudScale: elastic resource scaling for multi-tenant cloud systems
Proceedings of the 2nd ACM Symposium on Cloud Computing
AutoScale: Dynamic, Robust Capacity Management for Multi-Tier Data Centers
ACM Transactions on Computer Systems (TOCS)
| Hi-index | 0.00 |
Elastic resource management is one of the key characteristics of cloud computing systems. Existing elastic approaches focus mainly on single resource consumption such as CPU consumption, rarely considering comprehensively various features of applications. Applications deployed on a PaaS are usually heterogeneous. While sharing the same resource, these applications are usually quite different in resource consuming. How to deploy these heterogeneous applications on the smallest size of hardware thus becomes a new research topic. In this paper, we take into consideration application's CPU consumption, I/O consumption, consumption of other server resources and application's request rate, all of which are defined as application features. This paper proposes a practical and effective elasticity approach based on the analysis of application features. The evaluation experiment shows that, compared with traditional approach, our approach can save up to 32.8% VMs without significant increase of average response time and SLA violation.