Adaptive scheduling for parallel tasks with QoS satisfaction for hybrid cloud environments
The Journal of Supercomputing
| Hi-index | 0.00 |
Execution time prediction is an important issue in cloud computing. Predicting the execution time fast and accurately not only can help users to schedule jobs smarter, but also maximize the throughput and minimize the resource consumption of cloud platform. While hybrid cloud provides methods to federate multiple cloud platforms, different cloud platforms have different resource attributes, which will increase the difficulties to predict a job's execution time. In this paper, we exploit Rough Set Theory (RST), which is a well-known prediction technique that uses the historical data, to predict the execution time of jobs. The evaluation presents that RST can utilize the accuracy of the execution time, while the decision can be made in a short period of time.