Application use v5

Developing an application with PGD is mostly the same as working with any PostgreSQL database. What's different, though, is that you need to be aware of how your application interacts with replication. You need to know how PGD behaves with applications, the SQL that is and isn't replicated, how different nodes are handled, and other important information.

  • Application behavior looks at how PGD replication appears to an application, such as:
    • The commands that are replicated
    • The commands that run locally
    • When row-level locks are acquired
    • How and where triggers fire
    • Large objects
    • Toast
  • DML and DDL replication shows the differences between the two classes of SQL statements and how PGD handles replicating them. It also looks at the commands PGD doesn't replicate at all.

  • Nodes with differences examines how PGD works with configurations where there are differing table structures and schemas on replicated nodes. Also covered is how to compare between such nodes with LiveCompare and how differences in PostgreSQL versions running on nodes can be handled.

  • Application rules offers some general rules for applications to avoid data anomalies.

  • Timing considerations shows how the asynchronous/synchronous replication might affect an application's view of data and notes functions to mitigate stale reads.

  • Table access methods (TAMs) notes the TAMs available with PGD and how to enable them.

  • Feature compatibility shows which server features work with which commit scopes and which commit scopes can be daisychained together.