Microsoft Azure Data Storage

Introduction

Microsoft Azure Data Storage is one of the cloud platform service that provides benefits like highly available, robust, secure and scalable storage in wide range of storage types for objects. It can be accessed from anywhere through http or https via REST API. Microsoft Azure itself provides multiple ways to interact with it like Portal, CLI, Powershell and Client Libraries for building developer application using .Net, JAVA or Python, etc. 

Data Storage Types

Block Storage is in the form of disks and being used as a part of traditional usage.

Object Storage is in the form of files i.e. all types of file and being used as a part of modern usage. For e.g., BLOB [Binary Large Objects]

Data Storage Services

1. Azure Blobs - Allows unstructured data like files, images, audio, video to be stored and accessed at a massive scale.

2. Azure Files - For managed file shares in cloud or on premises deployment like shared access of the file from different VM's.

3. Azure Queues - For storing & retrieving message queues upto 64 KB.

4. Azure Tables - For NoSQL schemaless storage of structured data. It is now a part of Azure COSMOS DB.

5. Azure Disks - For block storage of Azure VM's.

Step by step

In order to access data storage, you need to create Storage Account. It gives ability to create different types of storage. Most of Azure services SLA based and first priority is availability. Pricing structure is Pay-as-you-go.

1. Go to Azure Portal, Search Storage Account and click Create.

2. Under Project Details section Select Subscription, Resource group [Existing if available or create new] you want to create storage under.

3. Under Instance Details section give the storage account name with required and specified validation and Select Region to be place in.

4. Select Performance type based on requirement. It has two types:

a. Standard - This type of category will provide magnetic HDD based disks. It is limited in terms of disk performance because it can provide only a 500 IOPS (Input/output operations per second) and upto 60 MB per second of bandwidth per disk.

b. Premium - This type of category will support high speed SSD based disks. It offers high performance for applications hosted in Azure, because they use solid-state drives in the backend. It provides highest IOPS and bandwidth depending on the disk types like 7500 IOPS/disk and 250 MB/second respectively. Supports low latency (A delay between request and response).

5. Select Redundancy type based on requirement. It has 4 types:

Predominantly, redundancy is nothing but the replication. Moreover, to understand is that this is not as similar as backup. It can be implemented in Azure by two ways:

i. Replication in Primary Region.

a. LRS (Locally Redundant Storage): Three copies of your data in single physical data center in one availability zone. For e.g. If your one of the data copy gets crashed then you should be able to sustain with another available copy immediately.

b. ZRS (Zone Redundant Storage): Three copies of your data one in each all 3 availability zones in one region. For e.g. If your one of the availability zone gets crashed then you should be able to sustain with another available AZ immediately.

ii. Replication in Primary and Secondary Region.

a. GRS (Geo Redundant Storage): Three copies of your data in single availability zone of primary region and three copies of data in one availability zone of secondary region. For e.g. If your primary AZ gets crashed then you should be able to sustain with another available secondary AZ immediately.

b: GZRS (Geo Zone Redundant Storage): Three copies of your data in single availability zone of primary region and three copies of data in one each in all three availability zones of secondary region.

Note*: In terms of Azure, Primary and secondary regions are region pair. Every region in Azure has its own paired region that you can not change. Usually, paired regions are in the same country except few. For e.g. West India - South India, etc.

In GRS & GZRS, the secondary copy of the data is not available for use until the FAILOVER from primary to secondary.

6. Checkbox "Make read access to data available in the event of regional unavailability".

Replication aka Redundancy gets changed on the selection of checkbox check/uncheck. This checkbox gets appeared when the Redundancy selection is GRS and GZRS and disappeared when selection is LRS or ZRS. If the checkbox is checked then the secondary data will be available as Read only.

a. If the Redundancy is selected as GRS and checkbox is checked then Redundancy will be considered as RAGRS (Read Access only Geo Redundant Storage).

b. If the Redundancy is selected as GZRS and checkbox is checked then Redundancy will be considered as RAGZRS (Read Access only Geo Zone Redundant Storage).

For e.g.,

7. Keep rest of setting as is and click on Review + Create button then Create button. Storage Account gets created.