SkarpChapter 10 of 20
Azure Storage Services: Types, Redundancy, and Tiers
Open up Azure Storage to see how blobs, files, disks, and queues differ, and how redundancy and access tiers affect durability, performance, and cost.
Big Picture: Why Azure Storage Matters
Where Your Data Lives
Azure Storage is where your data actually lives. It provides durable, highly available, and scalable storage services for many kinds of workloads.
What AZ-900 Cares About
For AZ-900, you must recognize storage service types, understand redundancy options, choose access tiers, and identify key migration tools at a conceptual level.
Toolbox Analogy
Think of Azure Storage as a toolbox: different storage types are tools, redundancy is how many copies and where, and access tiers are how often you plan to use the data.
Stable Core Concepts
As of 2026, the core Azure Storage concepts you learn here map directly to the exam: you need trade-off understanding, not SKU-level memorization.
Storage Accounts and Core Services Overview
What Is a Storage Account?
A storage account is the top-level container for your Azure Storage data. It defines which services you can use, plus redundancy, performance, security, and networking settings.
Azure Blob Storage
Blob Storage holds large amounts of unstructured data such as images, videos, backups, and logs. It is accessed via URLs, SDKs, or REST APIs over HTTP/HTTPS.
Azure Files
Azure Files provides fully managed file shares accessible over SMB or NFS. It is ideal for lifting traditional on-premises file shares into Azure without rewriting apps.
Azure Disk Storage
Azure Disk Storage offers block-level disks for VMs. These disks act like physical drives for OS and data, with high performance and low latency for compute workloads.
Queues and Tables
Azure Queue Storage handles small messages for decoupled apps. Table Storage is a scalable NoSQL key/attribute store for semi-structured data at a high level.
Modern Account Type
General-purpose v2 storage accounts are the modern default, supporting all key services, tiers, and redundancy options relevant to AZ-900.
Matching Storage Services to Real-World Scenarios
Blob Storage Example
A mobile app storing user photos and videos at scale needs cheap object storage with URL access. Azure Blob Storage is the right match.
Azure Files Example
An app expects a classic network path like \\server\share. Azure Files provides SMB or NFS file shares so you can lift and shift without code changes.
Disk Storage Example
A SQL Server VM needs predictable IOPS and low latency. Azure Disk Storage supplies OS and data disks tuned for VM performance.
Queue Storage Example
A web front end queues orders and background workers process them. Azure Queue Storage buffers small messages for asynchronous processing.
Table Storage Example
Millions of simple telemetry records with few query needs fit Azure Table Storage, a low-cost NoSQL key/attribute store.
Exam Pattern
On AZ-900, map the scenario to the pattern: objects, file shares, VM disks, messages, or simple NoSQL data.
Redundancy Concepts: Why Multiple Copies Matter
What Is Redundancy?
Redundancy is about how many copies of your data exist and where they are stored. It impacts durability, availability, and cost.
Automatic Copies
Azure automatically maintains multiple copies based on the redundancy choice. You configure this at the storage account level.
Three Core Families
For AZ-900, focus on LRS (one datacenter), ZRS (multiple zones in one region), and GRS (secondary region as well as primary).
Exam Wording
Look for wording like “protect against datacenter failure” or “protect against regional outage” to choose the right redundancy level.
LRS, ZRS, and GRS (and Friends): Comparing Redundancy Options
LRS: Local Protection
LRS keeps three copies of your data in a single datacenter. It protects against hardware failure, but not against losing the whole datacenter or region.
ZRS: Zone-Level Resilience
ZRS stores synchronous copies across three Availability Zones in one region. It survives a datacenter failure while staying available for read and write.
GRS and RA-GRS
GRS replicates data from a primary to a secondary paired region asynchronously. RA-GRS adds read access to the secondary for global read scenarios.
GZRS and RA-GZRS
GZRS combines ZRS in the primary with geo-replication to an LRS secondary. RA-GZRS lets you read from the secondary as well.
Choosing an Option
Think: LRS for cheapest local durability, ZRS for zone failure, and GRS or GZRS (with RA variants) for full regional outage protection.
Access Tiers: Hot, Cool, and Archive
Why Access Tiers?
Access tiers let you balance storage cost against access cost and latency based on how frequently you read blob data.
Hot Tier
Hot is for frequently accessed data: highest per-GB storage cost, but lowest access cost and latency. Use it for active app data.
Cool Tier
Cool suits infrequently accessed data kept online, like short-term backups. Storage is cheaper than hot, but access costs more.
Archive Tier
Archive is for rarely accessed data that can wait hours to retrieve. Storage is very cheap, but rehydration and access are expensive.
Exam Patterns
Map patterns: frequent access → hot, occasional access → cool, compliance or long-term rarely used → archive.
Choosing Tiers and Redundancy for Sample Workloads
Photo-Sharing App
For a busy photo app: Blob Storage, hot tier, and ZRS or GZRS/RA-GZRS for strong availability and durability within and across zones.
Long-Term Reports
Monthly reports: store as blobs, move from hot to cool, then archive. Choose LRS or GRS based on how critical regional resilience is.
Lift-and-Shift File Server
For SMB-based apps: Azure Files with LRS or ZRS. Exam focus is on recognizing Azure Files for traditional file share workloads.
Compliance Archive
Raw sensor data kept 10 years but rarely read: Blob Storage, archive tier, and likely GRS or GZRS for regional outage protection.
Order Queue
Order-processing messages: Queue Storage with LRS or ZRS. Match messages → queues, not blobs or files.
Data Migration and Transfer: Getting Data Into Azure
Why Migration Tools?
Beyond choosing storage types, you must move data into Azure. AZ-900 wants you to recognize the main tools and when to use them.
Storage Explorer and AzCopy
Storage Explorer is a GUI for managing storage. AzCopy is a fast command-line tool for scripted or large-scale data transfers.
Azure Data Box
Data Box is a physical device for moving tens of terabytes or more when network bandwidth is limited or too slow.
Azure Migrate and Online Writes
Azure Migrate supports larger lift-and-shift projects. Apps can also write directly to Blob Storage via SDKs or REST APIs.
Exam Clues
Petabytes and slow internet → Data Box; automated large copy → AzCopy; manual GUI management → Storage Explorer.
Quick Check: Storage Types and Redundancy
Test your understanding of Azure storage services and redundancy options.
A company runs a web app that stores user-uploaded images and must stay available even if one datacenter in the region fails. Which combination is the best fit?
- Azure Files with LRS
- Azure Blob Storage with ZRS
- Azure Disk Storage with LRS
- Azure Blob Storage with Archive tier and LRS
Show Answer
Answer: B) Azure Blob Storage with ZRS
User-uploaded images are unstructured object data, so Blob Storage is appropriate. The requirement to survive a datacenter failure within the region points to ZRS, which stores synchronous copies across Availability Zones. LRS alone would not protect against datacenter loss, and the Archive tier is for rarely accessed data, not active app content.
Quick Check: Access Tiers and Migration Tools
Reinforce your understanding of tiers and data transfer options.
You need to move 80 TB of historical log files from an on-premises datacenter to Azure for long-term retention. The internet connection is slow and heavily used during business hours. Which option is most appropriate?
- Use Azure Storage Explorer to upload files over the internet to the hot tier
- Use AzCopy over the existing internet link to copy data to the cool tier
- Order an Azure Data Box, copy the data to it, and ingest into Blob Storage archive tier
- Attach Azure Disk Storage directly to the on-premises servers and copy the logs
Show Answer
Answer: C) Order an Azure Data Box, copy the data to it, and ingest into Blob Storage archive tier
For tens of terabytes with limited bandwidth, Azure Data Box is designed to physically move data to Azure. Because the logs are historical and for long-term retention, the archive tier of Blob Storage is the best cost fit. Storage Explorer and AzCopy over a slow link would be time-consuming; Azure Disks cannot be attached directly on-premises.
Key Azure Storage Concepts Review
Use these flashcards to reinforce core terms and distinctions that appear frequently in AZ-900 scenarios.
- Azure Blob Storage
- An Azure storage service for massive amounts of unstructured data such as text and binary objects (images, videos, backups, logs), accessed via HTTP/HTTPS using URLs, SDKs, or REST APIs.
- Azure Files
- A fully managed file share service in Azure, accessible over SMB or NFS, ideal for lift-and-shift scenarios where applications expect traditional network file shares.
- Azure Disk Storage
- Block-level storage designed to be attached to Azure virtual machines as OS or data disks, providing high performance and low latency for VM workloads.
- Azure Queue Storage
- A simple messaging service for storing and retrieving small messages, enabling decoupled and asynchronous communication between application components.
- Azure Table Storage
- A NoSQL key/attribute store for semi-structured data, offering scalable and low-cost storage for structured but schema-flexible datasets.
- Locally redundant storage (LRS)
- A redundancy option that keeps three synchronous copies of your data within a single datacenter in one region, protecting against hardware failures but not datacenter or regional loss.
- Zone-redundant storage (ZRS)
- A redundancy option that stores synchronous copies of your data across multiple Availability Zones in a single region, protecting against datacenter-level failures within that region.
- Geo-redundant storage (GRS) / Read-access GRS (RA-GRS)
- Redundancy options that replicate data asynchronously from a primary region to a secondary paired region, protecting against regional outages; RA-GRS additionally allows read access to the secondary.
- Hot access tier
- A Blob Storage tier optimized for frequently accessed data, with higher storage cost per GB but lower access and transaction costs and low latency.
- Cool access tier
- A Blob Storage tier for infrequently accessed data that still needs online access, with lower storage cost than hot but higher access and transaction costs.
- Archive access tier
- A Blob Storage tier for rarely accessed data that can tolerate hours of retrieval latency, offering very low storage cost but high rehydration and access costs.
- AzCopy
- A command-line utility optimized for high-performance data transfer to and from Azure Storage, suitable for scripted or large-scale migrations.
- Azure Data Box
- A physical appliance provided by Microsoft to transfer large amounts of data (tens of terabytes or more) to Azure when network bandwidth is limited or too slow.
- Azure Storage Explorer
- A graphical desktop tool for managing and transferring data in Azure Storage accounts, useful for interactive, ad-hoc operations.
Key Terms
- AzCopy
- A command-line utility optimized for high-performance data transfer to and from Azure Storage, suitable for scripted or large-scale migrations.
- Azure Files
- A fully managed file share service in Azure, accessible over SMB or NFS, ideal for lift-and-shift scenarios where applications expect traditional network file shares.
- Azure Data Box
- A physical appliance provided by Microsoft to transfer large amounts of data to Azure when network bandwidth is limited or too slow.
- Hot access tier
- A Blob Storage tier optimized for frequently accessed data, with higher storage cost per GB but lower access and transaction costs and low latency.
- Cool access tier
- A Blob Storage tier for infrequently accessed data that still needs online access, with lower storage cost than hot but higher access and transaction costs.
- Azure Blob Storage
- An Azure storage service for massive amounts of unstructured data such as text and binary objects (images, videos, backups, logs), accessed via HTTP/HTTPS using URLs, SDKs, or REST APIs.
- Azure Disk Storage
- Block-level storage designed to be attached to Azure virtual machines as OS or data disks, providing high performance and low latency for VM workloads.
- Archive access tier
- A Blob Storage tier for rarely accessed data that can tolerate hours of retrieval latency, offering very low storage cost but high rehydration and access costs.
- Azure Queue Storage
- A simple messaging service for storing and retrieving small messages, enabling decoupled and asynchronous communication between application components.
- Azure Table Storage
- A NoSQL key/attribute store for semi-structured data, offering scalable and low-cost storage for structured but schema-flexible datasets.
- Azure Storage account
- A top-level Azure resource that groups storage services such as blobs, files, queues, and tables, and defines settings like redundancy, performance, security, and networking.
- Azure Storage Explorer
- A graphical desktop tool for managing and transferring data in Azure Storage accounts, useful for interactive, ad-hoc operations.
- Geo-redundant storage (GRS)
- A redundancy option that replicates data asynchronously from a primary region to a secondary paired region, protecting against regional outages; data in each region is stored with LRS.
- Zone-redundant storage (ZRS)
- A redundancy option that stores synchronous copies of your data across multiple Availability Zones in a single region, protecting against datacenter-level failures within that region.
- Locally redundant storage (LRS)
- A redundancy option that keeps three synchronous copies of your data within a single datacenter in one region, protecting against hardware failures but not datacenter or regional loss.
- Geo-zone-redundant storage (GZRS)
- A redundancy option that combines ZRS in the primary region with asynchronous replication to an LRS secondary region, providing both zone-level and regional resilience.
- Read-access geo-redundant storage (RA-GRS)
- A variant of GRS that additionally allows read access to the secondary region, useful for read-heavy global workloads and disaster recovery scenarios.
- Read-access geo-zone-redundant storage (RA-GZRS)
- A variant of GZRS that also allows read access to the secondary region, combining zone redundancy, geo-replication, and secondary read access.