Date Added: Oct 2011
The system overload is a common problem in a Database-as-a-Service (DaaS) environment because of unpredictable and bursty workloads from various clients. Due to the ser-vice delivery nature of DaaS, such system overload usually has direct economic impact on the service provider, who has to pay penalties if the system performance does not meet clients' service level agreements (SLAs). In this paper, the authors investigate techniques that prevent system overload by using admission control. They propose a pro t-oriented admission control framework, called ActiveSLA, for DaaS providers. ActiveSLA is an end-to-end framework that consists of two components.