Tag Archives: Geo-redundant storage

#Azure : Storage replication


Microsoft Azure storage offers numerous type of availability and durability of the data, within the datacenter, across datacenters, within the same region, or across regions. Based on your needs, you can select the right replication methodology. For example, if you would like to save your data from catastrophic failure in a single region then choose replication option that supports replication across regions.

Please keep in mind that you can configure replication options when you are creating a storage account and each region doesn’t support all replication options. Microsoft Azure offers four different replication options.

LRS: Locally redundant storage

ZRS: Zone-redundant storage

GRS: Geo-redundant storage

RA-GRS: Read-access geo-redundant storage

Table below provides you a quick overview about differences between all four replication options.

strategy LRS ZRS GRS RA-GRS
Data is replicated across multiple datacenters. No Yes Yes Yes
Data can be read from a secondary location as well as the primary location. No No No Yes
Designed to provideĀ _durability of objects over a given year. at least 99.999999999% (11 9’s) at least 99.9999999999% (12 9’s) at least 99.99999999999999% (16 9’s) at least 99.99999999999999% (16 9’s)

Courtesy: Microsoft

Let me explain you all four replication options in detail.

Locally redundant storage: maintains three copies of your data. It replicates your data within a scale unit, which is hosted in a datacenter in the region in which you create your storage account. A scale unit is nothing but a set of multiple racks, which hosts storage nodes. To maintain high availability, these replicas reside in separate fault domains and update domain. A fault domain is nothing but a group of nodes, which belongs to a single point of failure. While an update domain is a group of nodes that can be upgraded at the same time. LRS is cost effective solution but doesn’t safeguard your data from datacenter level failure.

Zone-redundant storage: maintains three copies of your data. ZRS is little confusing as of now because of its two versions. ZRS is in preview, which falls under general purpose v2 storage accounts, it replicates data synchronously across multiple availability zones with in a region and very useful for highly available applications. While the existing or old ZRS capability is now referred to as ZRS classic, which falls under general purpose v1 storage accounts. ZRS classic replicates data asynchronously three times across two to three facilities within same region or in some cases across two regions. ZRS classic are planned to depreciate by March 2021 and once new ZRS generally available in a region then ZRS classic can’t be created.

Geo-redundant storage: maintains six copies of your data. It replicates three copies in one region and another three copies in another region. In primary region, it replicates your data within a scale unit, which is hosted in a datacenter in the region in which you create your storage account. A scale unit is nothing but a set of multiple racks, which hosts storage nodes. To maintain high availability, these replicas reside in separate fault domains and update domain just like LRS. While in secondary region also it does the same thing but between the region data replication take place in asynchronous mode. Your data doesn’t become available in case of primary region failure, until Microsoft initiates the failover. Primary and secondary region association is pre-defined based on the locations and can’t be changes manually. Once you create a storage account, you just need to specify your primary azure region. Here is the list of primary and their respective secondary regions.

Primary Secondary
North Central US South Central US
South Central US North Central US
East US West US
West US East US
US East 2 Central US
Central US US East 2
North Europe West Europe
West Europe North Europe
South East Asia East Asia
East Asia South East Asia
East China North China
North China East China
Japan East Japan West
Japan West Japan East
Brazil South South Central US
Australia East Australia Southeast
Australia Southeast Australia East
India South India Central
India Central India South
India West India South
US Gov Iowa US Gov Virginia
US Gov Virginia US Gov Texas
US Gov Texas US Gov Arizona
US Gov Arizona US Gov Texas
Canada Central Canada East
Canada East Canada Central
UK West UK South
UK South UK West
Germany Central Germany Northeast
Germany Northeast Germany Central
West US 2 West Central US
West Central US West US 2

Courtesy: Microsoft

Read-access geo-redundant storage: maintains six copies of your data. It replicates three copies in one region and another three copies in another region. In primary region, it replicates your data within a scale unit, which is hosted in a datacenter in the region in which you create your storage account. A scale unit is nothing but a set of multiple racks, which hosts storage nodes. To maintain high availability, these replicas reside in separate fault domains and update domain just like LRS. While in secondary region also it does the same thing but between the region data replication take place in asynchronous mode. In case of RA-GRS, your data is available in read mode always, even in case of primary region failure. However, you can’t get write access on your data from secondary region until Microsoft initiates the failover. Primary and secondary region association is pre-defined based on the locations and can’t be changes manually like GRS. Once you create a storage account, you just need to specify your primary azure region.

Advertisements