pgEdge has announced v25 of its distributed Postgres platform, adding several features to make it easier to manage and scale distributed PostgreSQL clusters.
The company’s Spock extension, which offers advanced logical replication capabilities, was updated to version 5.0, adding several new features.
In Spock 5.0, new nodes can be added to an existing Spock cluster without causing downtime on the origin or existing subscriber nodes. It creates a temporary replication slot and subscription, which allows the new node to clone the origin’s state in parallel and then shut down after it has been synced.
According to pgEdge, the benefits beyond not requiring downtime are safer scaling of production clusters, minimal manual intervention needed, and compatibility with the standard Spock CLI.
Another major new feature in Spock 5 is Apply-Replay for exception handling. When a replicated transaction runs into an exception, Spock will buffer it in memory instead of terminating the apply worker. Then, if an exception occurs, the apply worker enters exception-handling mode and replays the buffered transaction from memory.
Previously, when an exception occurred, the apply worker restarted, re-requested the transaction from the origin, and continued on. This earlier way of exception handling had a few downsides, such as the worker needing to restart, the transaction needing to be re-fetched over the network, consumption of XID resources during resources, and introduction of replication lag and resource overhead.
The company says that Apply-Replay significantly improves stability, and is particularly helpful in high-throughput environments or in intermittent network conditions.
pgEdge Platform v25 also adds support for a broader range of scenarios for automatic conflict resolution. For example, scenarios with two Inserts using the same Primary Key will result in the second insert being converted into an all-column update.
Other new features include a new interactive installation process, pgBackrest for easier configuration of backups, and enhancements to Active Consistency Engine (ACE), pdEdge’s tool for validating and repairing data consistency.
