You can use SQL to query data in maps, Kafka topics, or a variety of file systems. Results can be sent directly to the client or inserted into maps or Kafka topics.
|For a hands-on introduction to SQL, see Get Started with SQL Over Maps.|
To use the SQL service, the Jet engine must be enabled on the cluster. For information about how to enable the Jet engine, see Securing Jobs.
You can connect to the SQL service of a Hazelcast member using one of the following options:
|You cannot run SQL queries on lite members. This limitation will be removed in future releases.|
The easiest way to run SQL queries on a cluster is to connect to the SQL shell.
To connect to the SQL shell, run the
hz-cli sql command of your member’s built-in CLI.
<cluster-name> placeholder with the name of your cluster and the
<address> placeholder with your member’s IP address.
docker run --network hazelcast-network -it --rm hazelcast/hazelcast:5.3.0-SNAPSHOT hz-cli --targets <cluster-name>@<address> sql
Before you can query data in SQL, you need to create a mapping to the data source. Mappings store essential metadata about the source’s data model, data access patterns, and serialization formats so that the SQL service can connect to the data source and query it.
Currently, you can create mappings for the following data sources:
Mapping names and field names are case-sensitive.
For example, you can access an
employee but not as
Keywords or built-in function names are case-insensitive.
Hazelcast supports a subset of standard ANSI SQL statements as well as some custom ones for creating jobs, mappings, and streaming queries.
For a list of supported statements, see SQL Statements.
SQL has the following limitations. We plan to remove these limitations in future releases.
The only supported Hazelcast data structure is map.
You cannot query other data structures such as replicated maps.
Although you can create bitmap indexes in SQL, SQL queries do not leverage those indexes to improve query performance.
You cannot join results from two streaming sources.
See Join Tables.
You cannot run SQL queries on lite members. This limitation will be removed in future releases.
You cannot persist SQL metadata on disk. This means that when a cluster restarts, it does not retain any SQL mappings or views that you have created.
When a SQL statement is submitted for execution, the SQL service parses and optimizes it. The result is an execution plan in the form of a job that is cached and reused by subsequent executions of the same statement.
For details about how Hazelcast models and executes jobs, see How Hazelcast Models and Executes Jobs.