In this chapter, we saw how data is stored and accessed using a Hadoop SQL interface called Hive. We studied various partitioning and indexing strategies in Hive. The working examples helped us to understand JSON data access and schema evolution using Avro in Hive. In the second section of the chapter, we studied a NoSQL data store called HBase and its difference with respect to RDBMS. The row design of the HBase table is very crucial to balancing reads and writes to avoid region hotspots. One has to keep in mind the HBase table design best practices discussed in this chapter. The working example shows the easier paths of data ingestions into an HBase table and its integration with Hive.
In the next chapter, we will take a look at tools and techniques for designing real-time data analytics.
