PostgreSQL Deployment & Operations

Production guidance for PostgreSQL hosting, company database provisioning, high availability, backup/recovery, connectivity and routine IMS database operations.

Deployment role

PostgreSQL is the operational database engine for IMS company databases and the global IMS platform data. Production database hosting should be treated as a dedicated infrastructure responsibility, separate from user workstations and—with larger workloads—separate from heavy background/integration workers.

Company database provisioning

IMS Global creates the databaseCreate a new company database from IMS Global → Companies → Create DB. IMS Global creates and initializes the database after the administrator confirms the company/database profile and supplies the requested PostgreSQL administrator credentials. Do not create or clone the company database manually.

Production architecture

For environments where IMS availability is business-critical, use PostgreSQL streaming replication with a tested failover design and one stable client endpoint. The exact HA tooling and topology are infrastructure choices; IMS clients should not depend on a hard-coded physical primary node.

Storage and capacity

Favor low-latency enterprise storage, adequate memory for the active database/reporting workload, and enough CPU headroom for concurrent POS, reporting and background operations. Capacity should be based on measured workload rather than a fixed Store-count formula.

Connections and sessions

Monitor PostgreSQL sessions, idle-session behavior and application-name grouping. Connection pooling or additional proxies are infrastructure options, not mandatory IMS requirements; introduce them only after validating compatibility with the deployed client/service behavior.

Backup and recovery

Maintain tested backups of company databases and the global IMS database together with the PostgreSQL roles/configuration required for recovery. Retention and disaster-recovery design should match the business recovery objective, and restore testing is part of the backup process.

Maintenance

Keep PostgreSQL versions, OS patching, storage health, replication state, backup status and database statistics under routine review. Validate failover/recovery changes outside peak retail activity before relying on them in production.