Research Area: Cloud Computing
The exponential growth of computational power and the advent of a multitude of energy-rich fixed and mobile nodes coupled with the emergence of modern resource management and virtualization technologies, Mobile Ad-hoc Clouds (MACs) are emerging as a viable computational asset. A major challenge is to construct MACs comprising highly dynamic, mobile, heterogeneous, fractionated, and scattered resources. Conventional resource virtualization technologies do not seamlessly consolidate such resources while adequately isolating the resource layer concerns from the executing code logic, which is key for MACs to operate and provision basic services. To this end, we propose Planet Cloud, a smart MAC management platform with an intrinsic support for highly mobile, heterogeneously-composed and dynamically-configured MACs. Planet Cloud enables MACs to autonomously adapt to real-time dynamic variation in its underlying infrastructure isolating the hardware and code management concerns. Such isolation enabled seamless task deployment, execution, migration, dynamic/adaptive resource allocation, and automated failure recovery. Analysis and simulation show that the proposed platform can safely and reliably provision and maintain the needed computational power for the MAC to operate in a highly dynamic mobile work environment.
Keywords:
Author(s) Name: Ahmed Khalifa; Mohamed Azab; Mohamed Eltoweissy
Journal name:
Conferrence name: 2014 IEEE/ACM 7th International Conference on Utility and Cloud Computing
Publisher name: IEEE
DOI: 10.1109/UCC.2014.54
Volume Information: Volume 2014
