Prioritization of Overflow Tasks to Improve Performance of Mobile Cloud
Abstract
In this Prioritization of Overflow Tasks to Improve Performance of Mobile Cloud paper we present two such mechanisms that use prioritization: one in which forged tasks are given full priority over newly arrived tasks, and another in which a threshold is established to control priority so that full priority is given to forked tasks if their number exceeds a predefined threshold.
System Configuration
Platform : cloud computing
Conclusion
We have developed two priority schemes for resource allocation of on-demand job requests in a cloud server pool based on giving different priorities to overflow tasks with or without a pre-defined threshold. Unlike many of the existing works, our model is not sacrificing the complexity of the problem of offloading just to make it solvable. Instead, complexity is addressed through the use of two interacting stochastic models that are solved through fixed point iteration to achieve any desired level of error.
