Title :
COCA: Computation Offload to Clouds Using AOP
Author :
Chen, Hsing-Yu ; Lin, Yue-Hsun ; Cheng, Chen-Mou
Author_Institution :
Intel-NTU Connected Context Comput. Center, Nat. Taiwan Univ., Taipei, Taiwan
Abstract :
In this paper, we describe COCA -- Computation Offload to Clouds using AOP (aspect-oriented programming). COCA is a programming framework that allows smart phones application developers to offload part of the computation to servers in the cloud easily. COCA works at the source level. By harnessing the power of AOP, COCA inserts appropriate offloading code into the source code of the target application based on the result of static and dynamic profiling. As a proof of concept, we integrate COCA into the Android development environment and fully automate the new build process, making application programming and software maintenance easier. With COCA, mobile applications can now automatically offload part of the computation to the cloud, achieving better performance and longer battery life. Smart phones such as iPhone and Android phones can now easily leverage the immense computing power of the cloud to achieve tasks that were considered difficult before, such as having a more complicated artificial-intelligence engine.
Keywords :
aspect-oriented programming; cloud computing; mobile computing; smart phones; software maintenance; AOP; Android development environment; Android phone; application programming; aspect-oriented programming; cloud; computation offload; dynamic profiling; iPhone; mobile application; offloading code; programming framework; smart phones application; software maintenance; source code; static profiling; Artificial intelligence; IEEE 802.11 Standards; Java; Programming; Servers; Smart phones; Software; Android operating system; aspect oriented programming; computation offload;
Conference_Titel :
Cluster, Cloud and Grid Computing (CCGrid), 2012 12th IEEE/ACM International Symposium on
Conference_Location :
Ottawa, ON
Print_ISBN :
978-1-4673-1395-7
DOI :
10.1109/CCGrid.2012.98