Mobile computing systems are made up of different components among which Mobile Support Stations (MSSs) play a key role. This paper proposes an efficient MSS-based non-blocking coordinated check pointing scheme for mobile computing environment. In the scheme suggested nearly all aspects of check pointing and their related overheads are forwarded to the MSSs and as a result the workload of Mobile Hosts (MHs) will reduce substantially. Moreover, the total amount of exchanging checkpoint requests will be decreased in order to have a batch transmission of such requests. The scheme is also enhanced using a simple data structure to have fewer propagating checkpoint requests and avoid the avalanche effect in the system. Simulation results show that compared to other existing algorithms, in the proposed scheme the average number of propagating requests and checkpoints, the average elapsed time for each check pointing process, and the size of system messages are significantly lower and smaller, respectively. Thus considering its distinguishing features, the proposed approach would be efficient and suitable for mobile computing environment.

By switching the computational load from mobile devices to the cloud, Mobile Cloud Computing (MCC) allows mobile devices to offer a wider range of functionalities. There are several issues in using mobile devices as resource providers, including unstable wireless connections, limited energy capacity, and frequent location changes. Fault tolerance and reliable resource allocation are among the challenges encountered by mobile service providers in MCC. In this paper, a new reliable resource allocation and fault tolerance mechanism is proposed in order to apply a fully distributed resource allocation algorithm without exploiting any central component. The objective is to improve the reliability of mobile resources. The proposed approach involves two steps: (1) Predicting device status by gathering contextual information and applying TOPSIS to prevent faults caused by volatility of mobile devices, and (2) Adapting replication and Checkpointing methods to fault tolerance. A context-aware reliable offloading middleware is developed to collect contextual information and manage the offloading process. To evaluate the proposed method, several experiments are run in a real environment. The results indicate improvements in success rates, completion time, and energy consumption for tasks with high computational loads.