A Hybrid System of Hadoop and DBMS for Earthquake Precursor Application

Compared with traditional data warehouse applications, big data analytics are huge and complex, and requires massive performance and scalability. In this paper, we explore the feasibility and versatility of building a hybrid system that not only retains the analytical DBMS, but also could handle the demands of rapidly exploding data applications. We propose a hybrid system prototype which takes DBMS as the underlying storage and execution units, and Hadoop as an index layer and a cache. Experiments show that our system meets the demand, and will be appropriate for analogous big data analysis applications.