Task Distribution on Cloud Using Smart phones for Image Processing

Android devices have become a popular and “anywhere, everywhere” computational resource for a wide range of requirements. Due to its ‘mobile’ nature, it allows people to carry high computational power in their hands where the computational power is comparable against that of a desktop or laptop. This computational power i.e CPU, Storage Memory and RAM of them is almost same like a desktop computer or Laptop in the recent years. However, Android devices were not being used for executing any computation intensive tasks till 2015 extensively. A recent study shows that users keep their Android devices idle for 8 hours on an average in the night time at their home. During this period, the device would be idle for most of the times except that it needs some CPU cycles and memory for download and/or other user requested applications running in the background. This processing power can be utilize for task scheduling. Android devices follows greedy algorithm for scheduling the tasks and faces a unique set of technical challenges due to the heterogeneity in CPU clock speed, variability in network bandwidth, and lower availability than servers. This paper uses a new scheduling algorithm (Linear Programming Algorithm) and we addressed many of these challenges to develop a distributed computing infrastructure using smart phones for task of Image Processing.