JSONB support for semi-structured data

You can use JSON and JSONB to provide semi-structured data. This is most useful for data that contains user-defined fields, such as field names that are defined by individual users and vary from user to user. We recommend using this in a semi-structured way, for example:

CREATE TABLE metrics (
  time TIMESTAMPTZ,
  user_id INT,
  device_id INT,
  data JSONB
);

Copy

When you are defining a schema using JSON, ensure that common fields, such as time, user_id, and device_id, are pulled outside of the JSONB structure and stored as columns. This is because field accesses are more efficient on table columns than inside of JSONB structures. Storage is also more efficient.

You should also use the JSONB data type, that is, JSON stored in a binary format, rather than JSON data type. JSONB data types are more efficient in both storage overhead and lookup performance.

Note

Use JSONB for user-defined data rather than sparse data. This works best for most data sets. For sparse data, use NULLable fields and, if possible, run on top of a compressed file system like ZFS. This will work better than a JSONB data type, unless the data is extremely sparse, for example, more than 95% of fields for a row are empty.

When you index JSONB data across all fields, it is usually best to use a GIN (generalized inverted) index. In most cases, you can use the default GIN operator, like this:

CREATE INDEX idxgin ON metrics USING GIN (data);

Copy

For more information about GIN indexes, see the PostgreSQL documentation.

This index only optimizes queries where the WHERE clause uses the ?, ?&, ?|, or @> operator. For more information about these operators, see the PostgreSQL documentation.

JSONB columns sometimes have common fields containing values that are useful to index individually. Indexes like this can be useful for ordering operations on field values, multicolumn indexes, and indexes on specialized types, such as a postGIS geography type. Another advantage of indexes on individual field values is that they are often smaller than GIN indexes on the entire JSONB field. To create an index like this, it is usually best to use a partial index on an expression accessing the field. For example:

CREATE INDEX idxcpu
  ON metrics(((data->>'cpu')::double precision))
  WHERE data ? 'cpu';

Copy

In this example, the expression being indexed is the cpu field inside the data JSONB object, cast to a double. The cast reduces the size of the index by storing the much smaller double, instead of a string. The WHERE clause ensures that the only rows included in the index are those that contain a cpu field, because the data ? 'cpu' returns true. This also serves to reduce the size of the index by not including rows without a cpu field. Note that in order for a query to use the index, it must have data ? 'cpu' in the WHERE clause.

This expression can also be used with a multi-column index, for example, by adding time DESC as a leading column. Note, however, that to enable index-only scans, you need data as a column, not the full expression ((data->>'cpu')::double precision).