Distributed Architecture for Grid Square Statistics

This chapter introduces distributed architecture for storing, searching, and computing Grid Square statistics. Since the data of Grid Square statistics is enormous, we need to consider efficient computation methods under limited computer resources. The distributed architecture gives us many solutions for handling Grid Square statistics. This chapter explains the fundamental principle of parallel computation with MapReduce algorithms or divide-and-conquer methods. We mention Amdahl’s law and Gustafson’s law. Moreover, we find an implementation of Web API a useful method to exchange data among elements of a distributed system. Finally, we propose a parallelization and distributed architecture for Grid Square statistics.