Azure Storage Redundancy

The data stored in the Azure storage accounts are always replicated to different data centers to make sure the high availability. Replication copies the data and stores in other regions/data centers based on the redundancy option. And this also protects the data even in the times of hardware failure.

It is very important to choose the redundancy option while creating a storage account. The redundancy options available in Azure are as follows:

  • Locally Redundant Storage (LRS)
    • Replicates data within the region where the primary storage account is created.
    • Three copies of data will be maintained and three replicas reside in the separate fault domains and update domains.
    • This is the lowest cost option in the offerings of redundancy.
    • It provided highest bandwidth of storage replication.
    • In disaster of the data center, it is difficult to recover the data as all replicas hosted on the same data center.
  • Zone Redundant Storage (ZRS)
    • Replicates data across data centers within one or two regions.
    • Three copies of data will be maintained. It provides high availability than LRS.
    • It is only available for block blobs in general purpose storage accounts.
    • Replication involves delay and replica may not available until Microsoft initiates failover to secondary.
    • ZRS cannot be converted to any other replication options once it is created.
    • And these accounts do not have metrics and logging capability.
  • Geo-Redundant Storage (GRS)
    • Replicates data across regions which is hundreds of miles away.
    • Six copies will be maintained, where three store locally and three in secondary region. And also replication to region involves delay.
    • These six copies of data will be maintained and managed across different fault domains and update domains.
    • The replica is not available unless Microsoft initiates failover to the secondary.
    • This provides high availability during in the disaster of primary region.
  • Read-Access Geo-Redundant Storage (RA-GRS)
    • Replicates data across regions which is hundreds of miles away.
    • Six copies will be maintained, where three store locally and three in secondary region. And also replication to region involves delay.
    • It provides the read-only access to the secondary region and the access keys will remain same for both primary and secondary endpoints.
    • Using this, the user has to put more focus towards endpoints. Application also has to design to manage which endpoint to interact.
    • This provides high availability during in the disaster of primary region.
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s