A Percona proof of concept tests a tiering layer in front of a saturated MongoDB replica set, without changing the application, and measures gains of up to 5.2x in operations per second.
The sys.schema_unused_indexes view only sees the server it's queried on. A Percona tool cross-checks results across the primary and replicas before suggesting any removal DDL.
A technical article on the Awide Labs blog details how Polar Proxy sends reads from an open PostgreSQL transaction to a replica without losing snapshot, locks, or read committed consistency.
A real-world test on PostgreSQL 18.6 shows what DROP TABLE actually costs in locks, WAL, and replicas, based on a case GitLab brought to PGConf.EU in Valencia. The right answer almost always involves partitioning before deleting.
Percona Operator for MySQL 1.3.0 now tunes MySQL configuration based on pod size, recovers lost transactions during asynchronous cluster failures, and encrypts the binlogs used for point-in-time recovery.
Developer Igor Suhorukov embedded vectorized execution into PostgreSQL and, in tests he ran himself, beat ClickHouse when reading Parquet files and came close on the MergeTree engine. For anyone deciding where to run analytics, the result is attention-grabbing, but it's not yet a reason to switch databases.
A post published by Munaf Khatri on the UnitAutogen blog (October 29, 2026) details, clause by clause, where what Supabase signs in the BAA ends and where the work of whoever writes the RLS policies, audit triggers, and schema functions begins.
When the restore command fails, PostgreSQL assumes by definition that the WAL archive ended right there, even when the real cause was network instability. Christophe Pettus shows how this exit code confusion can end a recovery silently, with no error in the log.
Devrim Gündüz, maintainer of the official PostgreSQL RPM repository, removed most of the EPEL dependencies in RHEL 8, 9, and 10. PostGIS, pgBackRest, patroni, and dozens of Python tools now install using only PGDG and RHEL's own repositories.
Released in PostgreSQL 16, in 2023, the parameter creates a second connection reserve, separate from the superuser's, so that monitoring, backup, and DBA can still get in when the application consumes the whole pool.
A backup job that finishes without error proves far less than it seems. The only thing that actually proves recovery is restoring, bringing up the cluster, and validating the data at a point in the past.
A pull request merged on October 24, 2026 teaches the operator to create, rotate, and expire PostgreSQL role passwords, with the database itself enforcing the deadline via VALID UNTIL.
PolarFS, the file system behind PolarDB, started writing a Disk-Paxos protocol to the shared block itself to ensure only one node writes at a time. The change, described in a post published on October 24, 2026, closes a flaw that only blocked two processes on the same host, not on different hosts.
Jeremy Schneider's investigation shows how the prefetch feature, enabled by default in recent PostgreSQL versions, can turn sequential writes into random reads and lock up clusters with synchronous replication.
Ahsan Hadi, from pgEdge, tested migrating PostgreSQL 17 to 18 on a multi-master cluster with Spock and documented the two traps that no status check reveals.
