Skip to content

Azure Storage Fundamentals

Azure Storage Account Basics

What is an Azure Storage Account?

A storage account provides a unique namespace for your Azure Storage data that's accessible worldwide via HTTP or HTTPS.

Storage Account Types

Type Description Use Case Redundancy Options
General Purpose v2 Latest storage account type Recommended for most scenarios All options
General Purpose v1 Legacy type Legacy applications only LRS, GRS, RA-GRS
BlockBlobStorage Premium performance for blobs High transaction rates, low latency LRS, ZRS
FileStorage Premium performance for files High-performance file shares LRS, ZRS
BlobStorage Legacy blob-only Legacy, use GPv2 instead LRS, GRS, RA-GRS

Performance Tiers

Tier Description IOPS Throughput Use Case
Standard Magnetic drives Lower Lower General purpose, cost-effective
Premium SSD drives Higher Higher I/O intensive applications

Access Tiers (Blob Storage)

Tier Description Storage Cost Access Cost Use Case
Hot Frequently accessed Higher Lower Active data
Cool Infrequently accessed Lower Higher Short-term backup, infrequent access
Archive Rarely accessed Lowest Highest Long-term backup, compliance

Blob Storage Deep Dive

What is Blob Storage?

Object storage solution for the cloud, optimized for storing massive amounts of unstructured data.

Blob Types

Type Description Use Case Max Size
Block Blobs Optimized for uploading large amounts of data Text/binary files, media 190.7 TB
Append Blobs Optimized for append operations Log files, audit trails 195 GB
Page Blobs Optimized for random read/write VM disks, databases 8 TB

Container & Blob Hierarchy

Storage Account
├── Container 1
│   ├── Blob 1
│   ├── Blob 2
│   └── Virtual Directory/
│       └── Blob 3
└── Container 2
    └── Blob 4

Lifecycle Management

Automatically transition blobs between access tiers based on rules: - Age-based: Move data based on days since creation/modification - Prefix-based: Apply rules to specific blob paths - Blob type-based: Different rules for different blob types

Example Policy: 

If blob age > 30 days → Move to Cool
If blob age > 90 days → Move to Archive  
If blob age > 2555 days → Delete

Blob Security Features

  • Shared Access Signatures (SAS): Time-limited access tokens
  • Azure AD authentication: Role-based access control
  • Storage Service Encryption: Automatic encryption at rest
  • Client-side encryption: Encrypt before uploading
  • Immutable storage: Write-once, read-many (WORM) compliance

File Storage (Azure Files)

What is Azure Files?

Fully managed file shares in the cloud accessible via Server Message Block (SMB) and Network File System (NFS) protocols.

Key Characteristics

  • SMB/NFS protocols: Standard file sharing protocols
  • Cross-platform: Windows, Linux, macOS
  • Cloud or on-premises: Accessible from anywhere
  • Shared access: Multiple VMs can mount same share
  • Azure File Sync: Sync with on-premises file servers

Performance Tiers

Tier Max IOPS Max Throughput Use Case
Standard 1,000 60 MB/s General file sharing
Premium 100,000 5 GB/s High-performance applications

Azure File Sync

  • Purpose: Centralize file shares in Azure while keeping local performance
  • Components: Storage Sync Service, Sync Group, Server Endpoint, Cloud Endpoint
  • Benefits: Cloud backup, disaster recovery, distributed access
  • Cloud Tiering: Keep frequently accessed files local, others in cloud

Use Cases

  • Lift and shift: Replace on-premises file servers
  • Shared application data: Configuration files, shared content
  • Development environments: Shared code repositories
  • Container storage: Persistent storage for containers

Queue Storage

What is Queue Storage?

Message queue service for reliable messaging between application components.

Key Features

  • Asynchronous messaging: Decouple application components
  • At-least-once delivery: Messages delivered reliably
  • Visibility timeout: Hide messages during processing
  • Time-to-live: Automatic message expiration
  • Base64 encoding: Text-based message format

Message Lifecycle

  1. Producer puts message in queue
  2. Consumer gets message (becomes invisible to others)
  3. Consumer processes message
  4. Consumer deletes message from queue

Use Cases

  • Work distribution: Distribute tasks across workers
  • Load leveling: Buffer between fast producers and slower consumers
  • Reliable messaging: Ensure messages aren't lost
  • Background processing: Trigger asynchronous operations

Table Storage

What is Table Storage?

NoSQL key-value store for rapid development using massive semi-structured datasets.

Data Model

  • Table: Collection of entities
  • Entity: Row in table (max 1MB)
  • Property: Column in entity (max 252 properties)
  • Partition Key + Row Key: Unique identifier for entity

Key Characteristics

  • Schema-less: No fixed schema required
  • Scalable: Handles billions of entities
  • Fast: Single-digit millisecond latency
  • Cost-effective: Lower cost than relational databases
  • ACID transactions: Within single partition

Partition Strategy

  • Good partition key: Distributes load evenly
  • Bad partition key: Creates hot partitions
  • Examples:
  • Good: UserID, Date ranges
  • Bad: Single value, Sequential numbers

Disk Storage

Managed Disks Overview

Azure-managed virtual hard disks (VHDs) for virtual machines.

Disk Types

Type Performance Use Case
Ultra Disk Highest IOPS and throughput Mission-critical workloads
Premium SSD High performance Production workloads
Standard SSD Moderate performance Web servers, dev/test
Standard HDD Basic performance Backup, non-critical

Disk Performance Comparison

Type Max IOPS Max Throughput Max Size
Ultra 160,000 2,000 MB/s 65,536 GB
Premium SSD 20,000 900 MB/s 32,767 GB
Standard SSD 6,000 750 MB/s 32,767 GB
Standard HDD 2,000 500 MB/s 32,767 GB

Managed Disk Features

  • Snapshots: Point-in-time backups
  • Images: Template for creating VMs
  • Shared disks: Multi-VM access for clusters
  • Encryption: Azure Disk Encryption, SSE
  • Availability zones: Zone-redundant storage

Redundancy Options

Local Redundancy Options

Type Copies Protection Cost
LRS 3 copies in single datacenter Hardware failures Lowest
ZRS 3 copies across 3 zones Zone failures Low

Geo-Redundancy Options

Type Copies Protection Read Access
GRS 6 copies (3 local + 3 remote) Regional disasters No
RA-GRS 6 copies (3 local + 3 remote) Regional disasters Yes
GZRS 6 copies (ZRS local + LRS remote) Zone + regional disasters No
RA-GZRS 6 copies (ZRS local + LRS remote) Zone + regional disasters Yes

Choosing Redundancy

Need read access to secondary region? → RA-GRS or RA-GZRS
Need zone redundancy? → ZRS, GZRS, or RA-GZRS
Cost-sensitive? → LRS
High availability required? → GRS or higher

Storage Security

Access Control Methods

Method Description Use Case
Storage Account Keys Full access to account Administrative access
Shared Access Signatures Limited access with expiration Temporary access
Azure AD Role-based access control User/application access
Anonymous access Public read access Public websites

Encryption Options

  • Encryption at rest: Automatic with Microsoft-managed keys
  • Customer-managed keys: Use your own encryption keys
  • Client-side encryption: Encrypt before sending to Azure
  • Encryption in transit: HTTPS/TLS for all connections

Network Security

  • Firewall rules: Restrict access by IP address
  • Virtual network service endpoints: Access from specific VNets
  • Private endpoints: Private IP access from VNet
  • Secure transfer: Require HTTPS for all operations

Azure Storage Services Integration

Content Delivery Network (CDN)

  • Purpose: Cache blob content globally for low latency
  • Providers: Microsoft, Verizon, Akamai
  • Benefits: Reduced latency, bandwidth savings, global scale
  • Use cases: Static websites, download distribution, streaming

Azure Import/Export

  • Purpose: Transfer large amounts of data using physical drives
  • Import: Ship drives to Azure datacenter
  • Export: Azure ships drives with your data
  • Use cases: Initial cloud migration, backup offloading, disaster recovery

AzCopy

  • Purpose: Command-line utility for copying data to/from storage
  • Features: Cross-platform, resume capability, synchronization
  • Performance: Optimized for high throughput
  • Use cases: Data migration, backup, synchronization

Storage Decision Framework

Choosing Storage Type

Unstructured data/files? → Blob Storage
Shared file system needed? → Azure Files
Message queuing? → Queue Storage
NoSQL key-value data? → Table Storage
Virtual machine disks? → Managed Disks

Blob Access Tier Selection

Accessed frequently? → Hot tier
Accessed monthly? → Cool tier
Long-term archive? → Archive tier
Unknown pattern? → Start with Hot, use lifecycle management

Performance Tier Selection

High IOPS/throughput needed? → Premium
Cost-sensitive? → Standard
Ultra-high performance? → Ultra Disk (for VMs)

Practical Examples

Static Website Hosting

  • Website files: Blob Storage ($web container)
  • Custom domain: Configure custom domain and SSL
  • CDN: Azure CDN for global distribution
  • Backup: Geo-redundant storage for disaster recovery

Data Lake Architecture

  • Raw data: Blob Storage (hierarchical namespace)
  • Processing: Azure Data Factory, Databricks
  • Analytics: Azure Synapse Analytics
  • Archive: Lifecycle policies to Archive tier

Enterprise File Sharing

  • Primary storage: Azure Files premium tier
  • Hybrid sync: Azure File Sync for on-premises
  • Backup: Azure Backup for file shares
  • Security: Azure AD authentication, encryption

Application Backend

  • User uploads: Blob Storage with SAS tokens
  • Session state: Table Storage or Redis Cache
  • Background jobs: Queue Storage for task distribution
  • Logs: Blob Storage append blobs

Best Practices

Performance Optimization

  • Choose appropriate performance tier for workload
  • Use multiple storage accounts to scale beyond limits
  • Implement proper partitioning strategy for Table Storage
  • Use CDN for globally distributed content
  • Monitor performance metrics and adjust accordingly

Security Best Practices

  • Use Azure AD authentication when possible
  • Implement least privilege access with RBAC
  • Use SAS tokens with minimal required permissions
  • Enable firewall rules and VNet integration
  • Regularly rotate storage account keys
  • Enable encryption for sensitive data

Cost Optimization

  • Use appropriate access tiers for blob storage
  • Implement lifecycle management policies
  • Use Azure Storage reserved capacity for predictable workloads
  • Monitor and clean up unused snapshots and old data
  • Choose appropriate redundancy level for requirements

Common Pitfalls

Blob Storage Issues

  • Not using lifecycle policies for cost optimization
  • Incorrect access tier selection
  • Poor container/blob naming strategy
  • Not implementing proper security controls
  • Ignoring storage analytics and monitoring

File Storage Issues

  • Wrong performance tier for workload requirements
  • Inadequate network bandwidth for file sync
  • Not configuring proper backup strategies
  • Authentication issues with domain-joined scenarios
  • Performance bottlenecks with large files

General Storage Issues

  • Not planning for storage account limits
  • Inadequate redundancy for business requirements
  • Poor key management practices
  • Not monitoring storage costs and usage
  • Incorrect region selection affecting performance