High availability and disaster recovery
Find out how this service provides disaster recovery in the event the service goes down.
| Service | High Availability | Disaster Recovery | Where can you host this service? |
|---|---|---|---|
| Genesys Info Mart | N = 1 (singleton) | Limited active spare | Primary or secondary unit |
See High Availability information for all services: High availability and disaster recovery
Writer's note: The rest of the page is new content, basically best-guess. Borys, please advise what else we need to say on this page. Possible GIM Classic content to cannibalize: Standby and Disaster Recovery in the GIM Dep Guide, Architecture page in the GIM Business Continuity Guide.
All the Genesys Info Mart services are stateful, and you cannot deploy mirrored services in parallel.
- For high availability (HA), Genesys Info Mart relies on inherent Kubernetes design and functionality. Observe Kubernetes best practices to ensure HA of Genesys Info Mart services.
- For disaster recovery (DR), you can deploy a complete production replica in another region, to operate in parallel and independently populate a separate Info Mart database. In the event of a disaster at the first site, you can switch to using the Info Mart database at the second site. Both Info Mart databases reflect activity of the entire contact center and contain data that is nearly identical. However, because the two sets of data are processed independently, there are differences in database-generated numbers.
DR architecture[ | edit source]
The following diagram extends Architecture diagram — Data flows to illustrate the DR architecture for Genesys Info Mart.
Writer's note: The diagram extrapolates from Borys's original architecture diagram to extend the above-referenced data flows diagram. If this is conceptually correct, I will ask a member of the IX diagrams team to modify and pretty up the data flows diagram accordingly. I will also ask them (a) to remove the connection numbers from the other connections and (b) to make the arrowheads from the GSPs into all the Kafkas bigger. If the official diagram is not available by the time the rest of this page is approved, can we publish it with this interim diagram in the meantime?
