Data Warehousing

Cloudera Data Platform Runtime includes Apache Hive, Apache Iceberg, and Apache Impala for storing and accessing data in the Hive metastore database. Hive addresses enterprise data warehouse demands for transactional data in the ORC file format. Iceberg is a table format for huge analytics that you can query using familiar SQL statements. Impala performs high-performance, low-latency SQL queries on data in Parquet and other formats.

Describes how Hive metastore (HMS) detects client types and and stores compatible tables, authorizes access to tables from Spark, and stores metadata of multiple services, such as Hive and Impala.

Describes how to launch Hive, execute commands, and issue queries from Beeline.

Covers how to use Hive 3 to query flat and transactional data using SQL statements.

Includes information about mature ACID v2 operations on transactions, compaction of files that accumulate during, ingestion, and query vectorization.

Describes how to set up Hive to generate statistics and control the number of concurrent connections to Hive.

Discusses how to choose an authorization model based on your use case.

Covers accessing Hive tables from Spark through the Spark Direct Reader or Hive Warehouse Connector, using JdbcStorageHandler to access an external data source, and connecting to Business Intelligence (BI) tools.

Explains low-latency analytical processing, caching, and tuning options.

You build on your SQL experience to analyze big data tables in Iceberg format on your file system or object store.

Presents the task topics for configuring client access to Impala, and starting and stopping Impala.

Describes a set of security features Impala provides to protect your critical and sensitive data.

Describes how to customize your environment after installing Impala.

Describes how to tune Impala queries and other SQL operations.

Presents the task topics for managing resources and metadata in Impala.

Describes how to monitor Impala service to run smoothly and avoid conflicts with other components running on the same cluster.