High availability and disaster recovery

Availability zones

An availability zone is a logically and physically isolated location within an IBM Cloud region where your data is processed and hosted.

• An availability zone has independent power, cooling, and network infrastructures that are isolated from other zones to strengthen fault tolerance by avoiding single points of failure between zones.
• An availability zone offers high bandwidth and low inter-zone latency within a region.

A region (location) is a geographically and physically separate group of one or more availability zones with independent electrical and network infrastructures isolated from other regions.

• Regions are designed to remove shared single points of failure with other regions and guarantee low inter-zone latency within the region.
• Each region has 3 different data centers (DC) for redundancy.

The following table lists the high-availability (HA) status for the regions (locations) where the IBM Cloud Monitoring service is available:

Where

• A geography is a geographic area or larger political body that contains one or more regions.

• A region is a defined geographic territory.

  A region could be a specific postal code area, a town, a city, a state, a group of states, or even a group of countries.

  A region contains multiple availability zones to meet local access, low latency, and security requirements for the region.

• N/A means feature that is not applicable to that geography.

• MZR means multi-zone region. Learn more.

Availability of a monitoring instance

When you provision a monitoring instance, you select the MZR (location) where the instance is created. The region determines where the monitoring data is processed and the data is hosted.

A multizone region (MZR) consist of 3 or more availability zones that are independent from each other to ensure that single failure events affect only a single zone.

By default, each monitoring instance consist of 3 zones, one primary zone and two secondary zones:

• Each zone is located in a different data center in the region.
• The data in your primary zone is automatically replicated to the secondary zones with low latency. You don't need to do anything to enable the replication.
• When the primary zone fails, a secondary zone is elected as the primary to prevent your service instance from being affected.
• If 2 zones fail at the same time, the service is down.

The MZR architecture offers automatic failover between 2 zones, and high availability for a monitoring instance withing a region.

Disaster recovery (DR) of the monitoring service in a region

Disaster recovery is about surviving a catastrophic failure or loss of availability in a single location.

IBM Cloud Monitoring follows IBM Cloud requirements for planning and recovering from disaster events.

If a regional disaster occurs, consider the following information:

• The estimated recovery time for rebuilding the regional site and restoring the service at another location is 24 hours.
• You will have to update the endpoints of applications and monitoring agents to point to the ingestion endpoint in the new location.
• You will have to restore the service instance's metadata, that is, dashboards and alerts definitions, from your backups.

Historical data may be lost during a disaster. If you require historical metrics for auditing purposes, backup the metrics regularly by querying the metrics from the service and storing them at a remote backup site. For more information, see Extracting metrics from a Monitoring instance by using the API.