Comparisons of PostgreSQL high availability tooling, anchored on PostgreSQL itself, which DeployOpen deploys on VMs and Kubernetes.
PostgreSQL high availability on VMs or Kubernetes with an external configuration store against a Kubernetes-native operator. Covers where your databases run, failover design, backup tooling and the skills each needs to operate.
Three ways to run highly available PostgreSQL: Patroni on virtual machines with a configuration store, the CloudNativePG operator on Kubernetes, or a managed service such as Amazon RDS for PostgreSQL. Covers failover design, readable standbys, what control you give up on a managed service, and who operates each layer.
PostgreSQL against Oracle's MySQL Community Server for new applications: licensing, release and support models (MySQL LTS and Innovation tracks), high availability options, features reserved for MySQL Enterprise Edition and migration effort.
PostgreSQL against GPLv2 MariaDB Server: governance, yearly LTS and quarterly rolling releases, replication and Galera clustering, vector search, compatibility with MySQL applications and the effort to run each in production.
The three open-source relational databases compared for self-hosted applications: licensing and governance, release support windows, high availability, paid enterprise features and the application compatibility questions that decide migrations.
Two PostgreSQL failover managers on VMs: Patroni's leader election through etcd, Consul, ZooKeeper or Kubernetes against EDB's repmgr toolset, on split-brain protection, supported versions, licensing, switchover and day-two operations.
Lightweight PgBouncer pooling (session, transaction and statement modes at about 2 kB per connection) against Pgpool-II's pooling, read load balancing, failover and watchdog: what each does, transaction pooling caveats and which belongs in front of an HA cluster.
The three main PostgreSQL operators for Kubernetes compared: CloudNativePG's failover through the Kubernetes API, Zalando's Patroni and Spilo based operator, and Crunchy PGO with pgBackRest, on licensing, image terms, backups and supported versions.
Choosing a replication method: physical streaming replication for HA standbys against logical replication for table subsets, cross-version and cross-platform copies, with what logical replication does not carry (DDL, sequences, large objects).
Tell us what you're working on, whether it's a deployment, an audit, a security test or a cyber range. You'll speak with an engineer who can help you scope it.