Compute Intensive Code Offloading in Mobile Device Cloud IEEE TENCON 2016 — Technologies for Smart Nation Paper ID # 533 Authors Sajeeb Saha, Md. Ahsan Habib, Md. Abdur Razzaque Green Networking Research Group Department of Computer Science and Engineering, University of Dhaka, Dhaka-1000, Bangladesh 2 • Introduction • Research Challenges • Contributions of this Work • State of the art Solutions • Code Offloading Architecture • Code Offloading Decision • Experimental Testbed • Simulation Results • Conclusion Dept. of Computer Science and Engineering, University of Dhaka 3 Applications -Visual Text Translation - Image Processing - Face Recognition - Sensing and Monitoring - Reality Augmentation - Real time Multimedia Dept. of Computer Science and Engineering, University of Dhaka 4 Constraints - CPU Power - Memory Size - Storage Space - Limited Bandwidth - Battery Lifetime Problems - Delay - Energy - QoS Dept. of Computer Science and Engineering, University of Dhaka 5 • Migrate computationally intensive software modules to higher end devices • Reduce energy consumption or response time Dept. of Computer Science and Engineering, University of Dhaka 6 Cloud Execution -Components are offloaded to remote server -Remote server does the Processing Constraints - Higher Latency - Network Disruption Dept. of Computer Science and Engineering, University of Dhaka 7 Cloudlet Execution -The device offloads the task to a nearby cloudlet. Constraints - Resource contension - Limited resource Dept. of Computer Science and Engineering, University of Dhaka 8 MDC Execution -Offloading to mobile devices. nearby Motivation - Device idle resource Dept. of Computer Science and Engineering, University of Dhaka 9 • Apportion of code to be offloaded • Selection of devices • Maintenance of connectivity among mobile devices • To ensure device availability Dept. of Computer Science and Engineering, University of Dhaka 10 • An algorithm for assigning application modules to available mobile devices • A mobile application as a emulation testbed • Comparative analysis to show the effectiveness of the proposed algorithm Dept. of Computer Science and Engineering, University of Dhaka 11 Dept. of Computer Science and Engineering, University of Dhaka 12 • A powerful job structure to speedup computing and conserve energy. • Provides an algorithm to disseminate tasks among mobile devices. Shi, Cong, et al. "Serendipity: enabling remote computing among intermittently connected mobile devices." Proceedings of the thirteenth ACM international symposium on Mobile Ad Hoc Networking and Computing. ACM, 2012. Dept. of Computer Science and Engineering, University of Dhaka 13 • An emulation testbed to quantify the potential gain of offloading tasks. • An MDC experimental platform for the assessment of MDCbased solutions. Abderrahmen Mtibaa, Khaled A Harras, and Afnan Fahim. Towards computational offloading in mobile device clouds. Cloud Computing Technology and Science (CloudCom), 2013 IEEE 5th International Conference on, volume 1, pages 331–338. IEEE. Dept. of Computer Science and Engineering, University of Dhaka 14 • An architecture for computational offloading including cloud, cloudlet and mobile device • Minimizes response time and energy while maximizing network lifetime. Abderrahmen Mtibaa, Khaled A Harras, Karim Habak, Mostafa Ammar, and Ellen W Zegura. Towards mobile opportunistic computing. In 2015 IEEE 8th International Conference on Cloud Computing, pages 1111–1114. IEEE, 2015. Dept. of Computer Science and Engineering, University of Dhaka 15 No consideration of dependency between the modules in parallel execution Dept. of Computer Science and Engineering, University of Dhaka 16 • Mobile device as a cloud service provider (VM) • Flexible amount of shareable resources • A three tier architecture Dept. of Computer Science and Engineering, University of Dhaka 17 List of Notations M Processing modules Ctime Level execution time L Total dependency levels Donor device energy Associativity time of the donor B Data rate between donor and cloudlet etime Execution time ttime Communication time N Total available devices Dept. of Computer Science and Engineering, University of Dhaka 18 Energy Constraint Bandwidth Constraint Associativity Constraint Dept. of Computer Science and Engineering, University of Dhaka 19 Device Parameters Application Parameters Dept. of Computer Science and Engineering, University of Dhaka 20 Dept. of Computer Science and Engineering, University of Dhaka 21 Impact on dependency levels Dept. of Computer Science and Engineering, University of Dhaka 22 Impact on number of available devices Dept. of Computer Science and Engineering, University of Dhaka 23 • An offloading decision making algorithm for compute intensive mobile applications • Algorithm outperforms the state-of-the-art works in terms of execution time. Dept. of Computer Science and Engineering, University of Dhaka 24 This work is supported by a grant for the ”Research Fellowship (2015-2016)” funded by the Information and Communication Technology Division, Ministry of Posts, Telecommunications and Technology, Government of Bangladesh. Dept. of Computer Science and Engineering, University of Dhaka Information 25 [1] M. Meeker. Internet trends 2015-code conference. Glokalde, 1(3), 2015. [2] Mobile App Usage Statistics Overview. http://www.statista.com/topics/1002/mobile-app-usage/, Access Date: 19/05/2016. [3] Karthik Kumar, Jibang Liu, Yung-Hsiang Lu, and Bharat Bhargava. A survey of computation offloading for mobile systems. Mobile Networks and Applications, 18(1):129–140, 2013. [4] Mike Jia, Jiannong Cao, and Lei Yang. Heuristic offloading of concurrent tasks for computation-intensive applications in mobile cloud computing. In Computer Communications Workshops (INFOCOM WKSHPS), 2014 IEEE Conference on, pages 352–357. IEEE, 2014. [5] Zixue Cheng, Peng Li, Junbo Wang, and Song Guo. Justin-time code offloading for wearable computing. IEEE Transactions on Emerging Topics in Computing, 3(1):74–83, 2015. [6] Mahbub E Khoda, Md Abdur Razzaque, Ahmad Almogren, Mohammad Mehedi Hassan, Atif Alamri, and Abdulhameed Alelaiwi. Efficient computation offloading decision in mobile cloud computing over 5g network. Mobile Networks and Applications, pages 1–16, 2016. [7] Mahadev Satyanarayanan, Paramvir Bahl, Ram´ on Caceres, and Nigel Davies. The case for vm-based cloudlets in mobile computing. IEEE pervasive Computing, 8(4):14–23, 2009. Dept. of Computer Science and Engineering, University of Dhaka 26 [8] Abderrahmen Mtibaa, Khaled A Harras, and Afnan Fahim. Towards computational offloading in mobile device clouds. In Cloud Computing Technology and Science (CloudCom), 2013 IEEE 5th International Conference on, volume 1, pages 331–338. IEEE, 2013. [9] Cong Shi, Vasileios Lakafosis, Mostafa H Ammar, and Ellen W Zegura. Serendipity: enabling remote computing among intermittently connected mobile devices. In Proceedings of the thirteenth ACM international symposium on Mobile Ad Hoc Networking and Computing, pages 145–154. ACM, 2012. [10] Abderrahmen Mtibaa, Khaled A Harras, Karim Habak, Mostafa Ammar, and Ellen W Zegura. Towards mobile opportunistic computing. In 2015 IEEE 8th International Conference on Cloud Computing, pages 1111–1114. IEEE, 2015. [11] Luiz Andr´ e Barroso and Urs Holzle. The case for energyproportional computing. Computer, 40(12):33–37, 2007. Dept. of Computer Science and Engineering, University of Dhaka 27