The Azure platform is a smart cloud service that Microsoft provides for modern data storage scenarios. With 5 types of storage, 4 levels of redundancy, and 3 types of storage accounts, Microsoft Azure Cloud offers multiple types of scalable, high-availability storage.
In this know-it-all guide, we explain the different Azure storage services, storage types, and accounts.
Table of Contents
Organizations have varying requirements for their cloud-hosted data. With a virtually limitless capacity, Azure provides a pay-as-you-go model, requiring payment only for what has been used.
Azure offers different database options such as Azure Cosmos DB, Azure SQL Database, and Azure Table Storage. Storing and sending messages can be done with Azure Queues and Event Hubs, while Azure Files and Azure Blobs are used to store loose files.
Different Types of Data Storage in Azure
The data storage services offered by the Azure storage platform include:
The Azure File storage supports the needs of the Azure Virtual Machine (VM) environment. Files stored can be accessed from different Virtual Machines, like a network share.
Applications that depend on regular file sharing can be moved to this cloud storage. With Server Message Block (SMB) protocols SMB 2.1, SMB 3.0, or SMB 3.1.1, the stored files can be accessed simultaneously from multiple locations. Similar files can be shared with Read and Write permissions by multiple VMs.
Azure file shares can be mounted from the cloud or on-premises deployment of Linux, Windows, and macOS. A URL pointing to the file and containing a Shared Access Signature (SAS) token gives a user access to the files from anywhere.
Blob means Binary Large Object and implies unstructured data like images, videos, backup files, music files, etc. Also known as object storage, text or binary data like documents, media files, or application installers can be stored in blob storage.
Microsoft Azure offers 3 different ways of blob storage:
These are composed of small blocks, each with a unique block ID, and store text and binary data.
While the blobs can vary in size, the maximum limit is 4.77 TB per blob. They’re optimized for data streaming and parallel uploads. Blobs can be inserted, deleted, and replaced but not modified without completely rewriting.
Currently, larger block blobs with 190.7 TiB storage capacity are available. The maximum number of blocks per blob is 50,000, and each block can have a capacity of 4,000 MiB.
When changes are made to block blobs, data gets appended at the end of the blob. Data can only be appended, not updated or deleted.
Append blobs are beneficial for logging data from virtual machines or multiple input/output operations like in databases.
Considering disk limitations, page blobs consist of 512-byte pages, with the maximum blob size being 8 TB. These are used for fast write and read operations.
It functions as disks for Azure VMs and stocks virtual hard drive files. Any disk run on Microsoft Azure’s VM uses page blobs. These are the basis for the Azure VM environment.
The Azure queue stores a large number of messages which can be accessed with authenticated HTTP or HTTPS requests from anywhere in the world.
A queue can store as many messages as the total storage capacity of the account permits. A single queue message can be up to 64 KB.
Azure queue storage is used to carry messages from the Azure web role to the Azure worker role, and to create a backlog of work done and processes, asynchronously.
Storing large amounts of structured data can be done using Azure tables. This is a cheap and highly scalable way to store NoSQL data, although it gets expensive if files are accessed frequently.
Since the data doesn’t have a fixed structure, this is a schema-less design. It is easier to store datasets that don’t require any foreign keys or joins.
With Azure tables, petabytes of data can be stored. It is particularly useful for flexible datasets like user data for web apps, device information, address books, etc.
An attached virtual hard disk can persistently store and access data. Azure disk storage is based on Page blobs and allows disks to be created for virtual machines. However, the disk can be accessed only from one virtual machine.
The 2 different disk speeds are:
Standard – Uses HDDs that are slow and cheap.
Premium – Uses SSDs that are fast and expensive.
At the time of this writing, Microsoft announced the preview of Premium SSD v2, the next generation of Microsoft Azure Premium SSD Disk Storage. This new disk offering provides the most advanced block storage solution designed for a broad range of input/output (IO)-intensive enterprise production workloads that require sub-millisecond disk latencies as well as high input/output operations per second (IOPS) and throughput—at a low cost.
The 2 types of Azure disk storage include:
Managed – Azure creates and manages the disks for the users, who need only to select the disk size and desired type (standard or premium).
Unmanaged – The user must manage the disk storage and corresponding storage account.
For additional safety measures, Microsoft Azure has 4 data replication options in the cloud storage:
Locally redundant storage (LRS) – Maintains 3 copies of a file in a data center but in separate fault domains and upgrade domains. This is available for all 5 types of storage and is the basic and cheapest way to replicate data.
Zone redundant storage (ZRS) – Maintains 3 copies of a file, just like in LRS; however, the data is asynchronously replicated across data centers in 1-2 regions. This is available only for block blobs. Even when the primary data center is unavailable, the data stored is durable.
Geo-redundant storage (GRS) – A total of 6 copies of the data are maintained in 2 separate regions, hundreds of miles apart. This way, the data is durable even if the primary region has a complete regional outage or disaster.
Read-access geo-redundant storage (RA-GRS) – This storage provides read-only access to the data stored in the secondary location by GRS. It helps maximize the storage account’s availability.
Types of Azure Storage Accounts
Creating an Azure account provides access to any type of Azure storage. With an account, data transfers can be done to or from services in the account.
The 3 main types of Azure storage accounts are:
General Purpose Storage
A general purpose v2 account operates with all types of Azure storage, except disk storage. This account provides space to access blobs, files, queues, and tables in one unified account.
It can be used as a NoSQL data store, to store object data, and set up file shares in the cloud.
Creating disks inside Azure storage requires a Microsoft Azure VM first.
A blob storage account specializes in storing block blobs and append blobs. This account can also be used to select an access tier based on the frequency of data access.
The three access tiers are:
Hot – Used for data that is frequently accessed. It grants the lowest latency possible, hence is more expensive.
Cold – Used for data that is less frequently accessed. It provides higher latency and hence comes with a lower price tag.
Archive – Used for data that is rarely accessed. It provides flexible latency requirements, on the order of hours. The Archive access tier has the lowest storage cost, but higher data retrieval costs and latency compared to the Hot and Cool tiers.
Here is a detailed documentation of Microsoft Azure Storage.
Azure Files offers four different tiers of storage called premium, transaction optimized, hot, and cool to allow you to tailor your shares to the performance and price requirements of your workload:
Premium – Premium file shares enable highly I/O intensive workloads, reduced latency, and consistently high performance. These are good for databases, development environments, and website hosting. It is offered on high-performance solid-state drive-based storage.
Transaction optimized – Used for transaction-heavy workloads that don’t require the premium file share latency. It’s suitable for applications requiring file or backend storage and is offered on standard storage hardware backed by hard disk drives (HDDs).
Hot – Hot file shares are useful for most general-purpose workloads, including for lifting and shifting on-premises file shares to Azure, especially with Azure File Sync. Hot file shares are offered on standard storage hardware backed by HDDs.
Cool – A cost-efficient, optimized storage for storing online archives, this is offered on standard storage hardware backed by HDDs.
Learn more about how to automatically move files between Azure file share tiers and optimize storage costs.
Azure storage provides several benefits:
Security – A key model helps authenticate a user with Azure Active Directory, making it difficult for intruders to access information. The use of Shared Access Signature (SAS) restricts data access and the integration with Microsoft Defender for storage, making it secure.
Durability and high availability – To keep data safe and secure in the event of hardware failures or natural disasters, the stored data is replicated and stored in separate geographical locations. It is easy to access the replicated data.
Accessibility – Data can be accessed from anywhere using HTTP or HTTPS. The Azure storage explorer and portal provide a user-friendly way to work with data.
Scalability – Based on requirements, the data storage is massively scalable.
Managed – Azure handles the hardware maintenance, updations, and critical issues.
In this article, we explained the different Azure storage services, storage types, and accounts.
Azure Storage is evolving, and new features are being added constantly. No matter what solution you are building for the cloud, you will find a compelling use for Azure Storage.
Read into how to monitor Azure Storage Account Activity Log with Microsoft Sentinel to look for threat visibility and response, proactive hunting, and alert detection.
To learn more about Azure Storage including Azure Blobs and Azure File Shares, check my recently published online course: Azure Storage Essential Training.
Thank you for reading my blog.
If you have any questions or feedback, please leave a comment.