IMPLEMENTATION OF MOBILE CROWD NETWORK FOR EQUALIZED DISTRIBUTION OF WORK PROCESSING

Abstract

In this project we implement distributed framework for performing a simple computation task through word file with the help of client and server. A central server will take a large file and distribute the file to different devices. Mobile computing devices will then be allowed to connect to the server through android app. Connecting, the devices will assign tasks to complete, and upon completing the tasks, the results will be sent back to the server to be mapped. Each device will get certain line from file to calculate the frequency of each word in the line. When the device will send back the output processing time taken by the device and no of line will be calculated. Once the server will understand which device is working fast more work will be assign to that system automatically. Again the process will be repeat and the final word count and highest number of times words repeated will be calculated in the end. The inherent problems of mobile computing such as resource scarcity, security and low connectivity pose problems for most applications. And the dynamic nature of mobile computing makes interacting and coordinating work difficult. We help by pooling together the processing power of mobile devices within a crowd to form mobile crowd. We explore this concept of ‘work stealing’ crowd computing in a distributed processing on an opportunistic network and focus on the optimized processing of work.