Cloud computing, with the on-demand feature to dynamically provision resource to meet workload variation, promises customers the ability to maintain the availability of their applications to the largest extent. To respond the workload, different dynamic scaling strategies (changing the partition of resources inside a VM or adjusting the amount of VM instances) are employed. However, how and where to add or remove resource has impact on the satisfaction of availability requirements, which has not been widely discussed yet.
In Wenting’s work, a novel availability-based computing model was presented to describe availability attribute of one application in the hierarchical structured cloud. Moreover, an availability-aware placement algorithm of dynamic scaling was proposed to achieve the objectives of meeting availability demands and minimizing the total communication cost. Her paper is published at The 9th IEEE International Conference on Autonomic and Trusted Computing (ATC 2012), Fukuoka, Japan.
The above picture was shot when she was presenting her work at the conference. If you are interested in her research, feel free to send your comments and opinion to her.