Evaluation of Cloud Offloading and Scheduling Mechanisms in Different Scenarios
This chapter presents a detailed discussion of the various experimental setups used to test all the solutions discussed in the previous chapters. The experimental results are then discussed and the performance of all schemes described in this book is compared with one another as well as with approaches including (1) local execution (no offloading); (2) complete offloading (all components remotely executed); (3) the non-time adaptive dynamic offloading algorithm proposed in the literature extended to applications with sequential dependency graphs; and (4) the approach where offloading takes place only via the best link at each instant of time. It is seen that performance-wise the best overall optimal solution is achieved with cognitive scheduling and cloud offloading.
- 10.E. Cuervo, A. Balasubramanian, D.-K. Cho, A. Wolman, S. Saroiu, R. Chandra, P. Bahl, MAUI: making smartphones last longer with code offload, in Proceedings of the International Conference on Mobile Systems, Applications, and Services, MobiSys (ACM, New York, 2010), pp. 49–62Google Scholar
- 23.U. Kremer, J. Hicks, J.M. Rehg, Compiler-directed remote task execution for power management, in Workshop on Compilers and Operating Systems for Low Power (2000)Google Scholar
- 49.W. Zhang, Y. Wen, D. Wu, Energy-efficient scheduling policy for collaborative execution in mobile cloud computing, in IEEE Proceedings on INFOCOM, April 2013, pp. 190–194Google Scholar