specification: API Commons Rate Limits
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/RateLimits
provider: Oracle GoldenGate
providerId: oracle-goldengate
created: '2026-05-04'
# Provenance stamped 2026-08-11: this artifact was written by the API Evangelist
# bulk sweep dated 2026-05-04, not harvested from the provider. See roadmap#35.
method: generated
modified: '2026-05-05'
reconciled: false
tags:
  - CDC
  - Data Integration
  - Oracle Cloud
  - Rate Limiting
  - Real-Time Replication
description: >-
  Oracle GoldenGate's management REST APIs (Service Manager, Administration
  Server, Distribution Server, Receiver Server, Performance Metrics Server)
  do not publish per-second or per-minute rate-limit numbers. Throughput is
  bound by the OCPU and memory of the GoldenGate deployment, by the
  configured replication trail throughput, and by the underlying source and
  target database I/O. The OCI managed service additionally inherits the
  general OCI control-plane throttling behaviour for resource lifecycle
  operations.
sources:
  - https://docs.oracle.com/en/middleware/goldengate/core/
  - https://docs.oracle.com/en/cloud/paas/goldengate-service/
  - https://docs.oracle.com/en-us/iaas/Content/General/Concepts/servicelimits.htm
notes: >-
  No public per-tier RPS/RPM numbers are published. Limits are effectively
  the capacity of the deployed OCI compute / GoldenGate edition plus any
  inherited OCI control-plane throttling.
responseCodes:
  unauthorized: 401
  forbidden: 403
  notFound: 404
  throttled: 429
  serviceUnavailable: 503
limits:
  - name: Management REST API Throughput
    scope: deployment
    metric: varies
    limit: bound by GoldenGate deployment OCPU
    notes: >-
      Throughput is governed by the OCPU count of the GoldenGate deployment
      (managed-service edition or on-premises host) and is not exposed as a
      separate per-tenancy rate-limit table.
  - name: OCI Control-Plane Operations
    scope: tenancy
    metric: requests_per_second
    limit: see OCI service-limits documentation
    notes: >-
      Lifecycle operations against OCI GoldenGate (create, update, scale,
      delete deployment) inherit the OCI control-plane throttling envelope
      shared by all OCI services in the tenancy.
  - name: Replication Trail Throughput
    scope: deployment
    metric: events_per_second
    limit: bound by deployment OCPU and source/target I/O
    notes: >-
      Sustained replication throughput depends on the GoldenGate deployment
      sizing, trail file storage, and source/target database I/O capacity.
policies:
  - name: Capacity-Bound Throttling
    description: >-
      Throughput limits are determined by deployment OCPU, trail storage IO,
      and source/target database capacity rather than a published
      per-tenancy rate-limit table.
  - name: Backoff Strategy
    description: >-
      Clients should implement exponential backoff with jitter on 429 / 5xx
      responses; honor any Retry-After header when present.
  - name: Health and Metrics
    description: >-
      Use the Performance Metrics Server REST API (or OCI Monitoring for the
      managed service) for visibility into lag, throughput, and CPU
      saturation rather than polling the management API at high frequency.
  - name: Capacity Adjustment
    description: >-
      Scale the deployment OCPU count or migrate to a larger compute shape
      to lift effective throughput; raise OCI control-plane limits via a
      service request if needed.
maintainers:
  - FN: Kin Lane
    email: kin@apievangelist.com