Impala uses HDFS as its file storage system. Impala parses and processes structured data, while HDFS provides reliable underlying storage. Impala provides fast data access without moving data in HDFS.
Impala uses Hive metadata, Open Database Connectivity (ODBC) driver, and SQL syntax. Unlike Hive, which is over MapReduce, Impala implements a distributed architecture based on daemon and handles all query executions on the same node. Therefore, Impala is faster than Hive by reducing the latency caused by MapReduce.
Kudu can be closely integrated with Impala to replace the combination of Impala, HDFS, and Parquet. You can insert, query, update, and delete data in Kudu tablets using Impala's SQL syntax. In addition, you can use JDBC or ODBC to connect to Kudu for data operations, using Impala as the broker.
The default Impala tables use data files stored in HDFS, which is ideal for batch loading and query of full table scanning. However, HBase provides convenient and efficient query of OLTP-style organization data.