Parallel model creation in cloud platform

摘要

A system, method, and computer-readable medium to receive at least one data table relating to a data set; receive a plurality of dimensions; combine, by a first parallelization process, the plurality of dimensions into a plurality of different clusters, each cluster being a grouping of different dimensions; transmit each of the plurality of different clusters to a cloud platform; divide, by a second parallelization process, the at least one data table into a plurality of different chunks of data; and transmit each of the plurality of different chunks of data to the cloud platform, in parallel to and independent of the transmission of the plurality of different clusters the cloud platform, the cloud platform to generate a data model based on a combination of the transmitted plurality of different clusters and the transmitted plurality of different chunks of data.