PROJECT TITLE :

Cost Aware Service Placement and Load Dispatching in Mobile Cloud Systems

ABSTRACT:

With proliferation of sensible phones and an increasing range of services provisioned by clouds, it's commonplace for users to request cloud services from their mobile devices. Accessing services directly from the Web information centers inherently incurs high latency because of long RTTs and potential congestions in WAN. To lower the latency, some researchers propose to 'cache’ the services at edge clouds or smart routers in the access network which are closer to end users than the Internet cloud. Although 'caching’ is a promising technique, inserting the services and dispatching users’ requests during a manner that may minimize the users’ access delay and repair providers’ value has not been addressed therefore so much. During this paper, we tend to study the joint optimization of service placement and cargo dispatching within the mobile cloud systems. We tend to show this drawback is distinctive to both the traditional caching downside in mobile networks and therefore the content distribution drawback in content distribution networks. We tend to develop a set of economical algorithms for service suppliers to attain numerous trade-offs among the average latency of mobile users’ requests, and the value of service providers. Our resolution utilizes user's mobility pattern and services access pattern to predict the distribution of user's future requests, and then adapt the service placement and load dispatching online primarily based on the prediction. We tend to conduct extensive trace driven simulations. Results show our answer not solely achieves a lot of lower latency than directly accessing service from remote clouds, but additionally outperforms different classical benchmark algorithms in term of the latency, price and algorithm running time.