An efficient hierarchical data placement algorithm for massive spatial data storage systems

This paper presents a novel hierarchy data placement algorithm based on data geographic scope to distribute several petabytes of spatial data among tens or hundreds of thousands of devices. Specially, it uses Max-Min algorithm to classify the devices into some classes for different devices configuration. Then we propose a hashing algorithm based on SIEVE [1] and Linear Hashing [2] to assign data between classes and within a class with the geographic scope of data.