The PostgreSQL International Improvement Group right this moment introduced the discharge of PostgreSQL 17, the most recent model of the open supply database.
In response to the group’s announcement, PostgreSQL 17 has improved efficiency and scalability whereas adapting to new information entry and storage patterns required by cloud native computing and the rise of AI.
Among the many key new options is enhanced help for JSON, which was one of many causes customers started adopting the database. On this launch, the implementation of the SQL/JSON normal is usually full, in keeping with Tom Kincaid, SVP of Database Server Improvement at EDB, a serious contributor to the challenge. “I believe one of many issues individuals ask for probably the most is the implementation of JSON_TABLE, which lets you take a JSON doc and make a view of it as a relational desk… it actually speaks to the extensibility of Postgres, but in addition the continued evolution in direction of making it simpler to undertake.”
PostgreSQL 17 now helps SQL/JSON constructors (JSON, JSON_SCALAR, JSON_SERIALIZE) and question capabilities (JSON_EXISTS, JSON_QUERY, JSON_VALUE), in keeping with the group’s announcement, giving builders different methods of interfacing with their JSON information. This launch provides extra jsonpath expressions, with an emphasis of changing JSON information to a local PostgreSQL information kind, together with numeric, boolean, string, and date/time sorts, the group introduced.
Additional, in keeping with the group’s announcement, the brand new model brings extra options to MERGE, which is used for conditional updates. New capabilities embody a RETURNING clause and the flexibility to replace views., in addition to bulk loading and information exporting, together with as much as a 2x efficiency enchancment when exporting massive rows utilizing the COPY command. “COPY efficiency additionally has enhancements when the supply and vacation spot encodings match, and features a new possibility, ON_ERROR, that permits an import to proceed even when there’s an insert error,” the announcement stated.
One other key function Kincaid identified is the addition of incremental backup. “Postgres databases could be 10s of terabytes, in some circumstances, 100 terabytes or extra, and having the ability to do a full backup on that would take a number of days, simply relying in your {hardware} and your structure and instruments,” he stated. “However now with incremental backup, you’ll be able to mix incremental backups right into a full backup, separate from the precise database server, so that you don’t must put some in depth load in your database server to do a full backup.”
PostgreSQL 17 brings system-wide efficiency good points and a extra sturdy developer expertise. Listed below are the launch notes for a whole checklist of recent and adjusted options.