Learn how to decide the number of Storage Accounts you need in Azure

Deciding how many Azure storage accounts you need depends on several factors, including performance, scalability, data management, security, and compliance requirements.

Below is a detailed guide to help you determine the appropriate number of storage accounts.

Key Considerations

1. Performance Requirements

Throughput and IOPS Limits

• Each storage account has performance limits. For example:

• Standard Storage: 20,000 IOPS for block blobs.

• Premium Storage: Higher performance for low-latency applications.

• If your workload exceeds these limits, you may need multiple storage accounts to distribute the load.

Data Ingress/Egress

• Standard accounts support up to 50 Gbps for ingress and 50 Gbps for egress (region-dependent). If your workloads exceed these, create additional accounts.

2. Scalability

Data Capacity

• A single storage account has a capacity limit (currently 5 PB for standard accounts, and 35 TB for premium accounts).

• If your total storage requirement exceeds this, divide the data across multiple accounts.

Logical Partitioning

• Use multiple storage accounts to organize data for better manageability (e.g., by department, project, or region).

3. Geographic Distribution

Latency Optimization

• Create storage accounts in regions closer to users or applications to reduce latency.

Data Sovereignty

• For compliance with regulations (e.g., GDPR), you may need separate accounts in specific regions to store sensitive data.

4. Security and Access Control

Access Control Policies

• If different teams or applications need isolated access to data, use separate storage accounts to simplify access management.

Shared Access

• To minimize risks, avoid sharing a single storage account across many unrelated services unless required.

5. Workload Types

Different types of workloads may benefit from different configurations:

Blob Storage

For unstructured data like images, videos, and logs.

File Storage

For shared files or lifting and shifting applications.

Queue/Table Storage

For messaging and NoSQL workloads.

Use separate accounts to isolate and optimize specific workloads.

6. Cost Management

• Use separate storage accounts to allocate costs to specific teams, departments, or projects.

• Leverage Azure's tagging feature to organize costs, but separate accounts can provide clearer billing segmentation.

7. Disaster Recovery and Redundancy

Geo-Redundancy

• For critical data, consider creating accounts with Geo-Redundant Storage (GRS) or Read-Access Geo-Redundant Storage (RA-GRS).

• If different recovery strategies are required for various data sets, use separate accounts.

Backup and Recovery

• Use separate accounts for backups to avoid interference with production workloads.

8. Compliance and Legal Requirements

• Certain industries require data to be stored in separate logical units for compliance reasons.

• If specific types of data (e.g., personal, financial, or health data) require additional controls, create dedicated accounts for them.

Best Practices for Deciding

1. Start Small and Scale Gradually

Begin with fewer accounts and monitor performance and scalability. Add more accounts as needed.

2. Partition Logically

Divide storage accounts by application, department, geography, or workload type.

3. Monitor Usage

Use Azure Monitor to track storage account metrics (e.g., throughput, capacity) and adjust the number of accounts accordingly.

4. Leverage Naming Conventions

Use meaningful names to indicate purpose, region, or application (e.g., app1-storage-eastus).

Example Scenarios

Scenario 1: Small Business

• Data: Less than 1 TB.

• Throughput: Low.

• Access: Single application/team.

• Recommendation: One storage account is sufficient.

Scenario 2: Enterprise Application

• Data: Hundreds of TBs.

• Throughput: High with multiple workloads (e.g., logs, backups, analytics).

• Regions: Distributed users.

• Recommendation: Use multiple accounts—one per workload or region.

Scenario 3: Compliance-Driven Organization

• Data: Sensitive and non-sensitive data.

• Regulation: Data must be stored in specific regions.

• Recommendation: Separate storage accounts for sensitive data in compliant regions.

Summary

By evaluating these factors, you can determine the optimal number of storage accounts to meet your technical, business, and compliance needs effectively.