Today we are excited to announce the public preview of the new Geo-Replication feature for Azure Service Bus in the premium tier. This feature ensures that the metadata and data of a namespace are continuously replicated from a primary region to a secondary region. Moreover, this feature allows promoting a secondary region at any time. The Geo-Replication feature is the latest option to insulate Azure Service Bus applications against outages and disasters. Other options are Geo-Disaster Recovery and Availability Zones.

Differentiation

There are currently two features that provide Geo-Disaster Recovery in Azure Service Bus for the Premium tier. First, there is Geo-Disaster Recovery (Metadata DR) that just provides replication of metadata. Second, Geo-Replication, which is now in public preview, provides replication of both metadata and data. Neither Geo-Disaster Recovery feature should be confused with Availability Zones. Regardless of if it is Metadata DR or Geo replication, both geo-graphic recovery features provide resilience between Azure regions such as East US and West US.

Availability Zones are available on all Service Bus tiers, and support provides resilience within a specific geographic region, such as East US. For a detailed discussion of disaster recovery in Microsoft Azure, see this article.

Concepts

The Geo-Replication feature implements metadata and data replication in a primary-secondary replication model. It works with a single namespace, and at a given time there’s only one primary region, which is serving both producers and consumers. There is a single hostname used to connect to the namespace, which always points to the current primary region. After promoting a secondary region, the hostname points to the new primary region, and the old primary region is demoted to secondary region. After the new secondary has been re-initialized, it is possible to promote this region again to primary at any moment.

Replication modes

There are two replication modes, synchronous and asynchronous. It’s important to know the differences between the two modes.

Asynchronous replication

Using asynchronous replication, all requests are committed on the primary, after which an acknowledgment is sent to the client. Replication to the secondary regions happens asynchronously. Users can configure the maximum acceptable amount of lag time, the offset between the latest action on the primary and the secondary regions. If the lag for an active secondary grows beyond user configuration, the primary will throttle incoming requests.

Synchronous replication

Using synchronous replication, all requests are replicated to the secondary, which must commit and confirm the operation before committing on the primary. As such, your application publishes at the rate it takes to publish, replicate, acknowledge, and commit. Moreover, it also means that your application is tied to the availability of both regions. If the secondary region goes down, messages aren’t acknowledged and committed, and the primary will throttle incoming requests.

Promotion

The customer is in control of promoting a secondary region, providing full ownership and visibility for outage resolution. When choosing Planned promotion, the service waits to catch up the replication lag before initiating the promotion. On the other hand, when choosing Forced promotion, the service immediately initiates the promotion.

Pricing

The Premium tier for Service Bus is priced per Messaging Unit. With the Geo-Replication feature, secondary regions run on the same number of MUs as the primary region, and the pricing is calculated over the total number of MUs. Additionally, there is a charge for based on the published bandwidth times the number of secondary regions. During the early public preview, this charge is waived.

More information on this feature can be found in the documentation. 

​Microsoft Tech Community – Latest Blogs –Read More