Azure Storage Replication Explained

Azure Storage is a managed data storage service in Microsoft Azure cloud which is highly redundant and protected from any kind of failure as it provides different level of data replication and redundancy. In this blog I would briefly explain about different  replication mechanism for Azure Storage along with graphical representation.

Azure Storage Replication Mechanism:

1. Locally Redundant Storage (LRS):
LRS synchronously replicates data three times within a single physical datacenter in a region, provides protection against server rack or storage cluster failure but can’t sustain Datacenter level (Availability Zone level also) failure. Provides at least 99.999999999% (11 nines) durability of the data.

2. Zone Redundant Storage (ZRS):
With ZRS, data is replicated synchronously across three Availability Zones (AZs) in an Azure region which means even if one of the AZ completely fails we can still continue to read and write the data without any interruption or data loss as each Availability Zone is an independent physical location within an Azure region, however this can’t sustain if the complete Azure region is impacted due to any unexpected failure. ZRS provides at least 99.9999999999% (12 9’s) durability of the data.

3. Geo Redundant Storage (GRS):
With GRS, first data is replicated synchronously three times within a single physical datacenter in the primary Azure region using LRS mechanism. It then replicates your data asynchronously to a single physical location in the secondary Azure region. After the data replicated to the secondary location, it’s also replicated within that location using LRS.GRS protects against region level disasters and provides at least 99.99999999999999% (16 9’s) durability of the data.

4. Geo Zone Redundant Storage (GZRS):
With GZRS, data is replicated across three Azure availability zones in the primary region and is also replicated to a secondary geographic region for protection from region level disasters. GZRS provides at least 99.99999999999999% (16 9’s) durability of the data. The key difference between GRS and GZRS is how data is replicated in the primary region. Within the secondary location, data is always replicated synchronously three times using LRS.

The main difference between GRS and GZRS replication is basically how the data is replicated in the primary region, data in secondary Azure region is always replicated three times using LRS.

Note: Your application or client can’t read or write in secondary Azure region with GRS or GZRS replication unless there is a failover to the secondary region. If you would like read access to secondary Azure region then you configure your storage account to use read-access geo-redundant storage (RA-GRS) or read-access geo-zone-redundant storage (RA-GZRS).

If primary Azure region where Storage Account resides goes unavailable due to any unplanned or planned events we can manually perform the failover to secondary region, once failover is completed, the secondary region becomes the primary region and we can again read and write data.

You can refer following Microsoft documentation to understand how exactly Azure Storage account failover works.
Disaster recovery and account failover:

Few important points to remember:
1. Storage (General Purpose v1) supports only LRS, GRS and RA-GRS replication type.
2. Premium Performance tier storage has only LRS replication type, BlockBlobStorage, FileStorage storage type under Premium performance tier has LRS and ZRS both.
3. Azure Managed Disks are only replicated with three times within single Availability Zone using Locally Redundant Storage (LRS) mechanism.

I hope you find this helpful to understand the Azure Storage replication, I will continue to explore and write more about Azure Storage and other stuffs soon. Please stay tuned.

Leave a Reply

Your email address will not be published.