An evaluation for optimal (big) data processing with Hadoop

Xiao jing Zhao

The motivation of this work is to find out the best practice of optimal (big) data processing with SQL-like queries on Hadoop. We emphase on kpi as the key performance indicator.The test platform, a cluster of 15 working nodes (HP ProLiant DL380p Gen8 LFF server), is provided by one of the biggest insurance company worldwide. 

An evaluation for optimal (big) data processing with Hadoop