1. Introduction to Dataverse Storage
Microsoft Dataverse (formerly Common Data Service) is the cloud-based data storage layer for Dynamics 365 and Power Platform. It provides a structured, secure, and scalable database for business applications.
Key Features of Dataverse Storage:
✔ Relational database (tables, relationships, business logic)
✔ Security & compliance (role-based access, encryption)
✔ Scalability (handles large datasets with performance optimization)
✔ Integration-ready (APIs, Power Automate, Azure Synapse)
2. Dataverse Storage Components
A. Table (Entity) Storage
- Standard Tables: Pre-built (e.g., Account, Contact, Opportunity)
- Custom Tables: User-defined (e.g., “Project Task,” “Inventory Item”)
- Storage Consumption:
- Each record consumes ~2 KB (varies by columns)
- File/Image data stored separately (see “File & Image Storage”)
B. File & Image Storage
- File Columns: Attach documents (PDF, Excel, etc.)
- Stored in Azure Blob Storage (not in SQL)
- Default 4 TB per environment (separate from DB storage)
- Image Columns: Profile pictures, product images
- Optimized for thumbnails & fast loading
C. Log & Audit Data
- Audit Logs: Track record changes (configurable retention)
- Plug-in Tracing: Debug logs for custom code
- Consumes storage but can be archived.
D. Managed & Unmanaged Solutions
- Solutions package customizations (forms, workflows, etc.)
- Storage impact: Minimal (~MBs per solution)
3. Dataverse Capacity & Limits
A. Storage Allocation
| Storage Type | Included (Per Environment) | Additional Purchase |
|---|---|---|
| Database | 1 GB (default) | Available in increments |
| File | 4 GB | Expandable up to 4 TB |
| Log | 500 MB | Configurable retention |
B. Key Storage Limits
✔ Max Database Size: Up to 4 TB (Enterprise plans)
✔ Max File Storage: 4 TB per environment
✔ Record Limit: Millions per table (performance degrades after ~1M)
✔ Column Limit: ~1,000 columns per table
4. How Dataverse Storage is Calculated
A. Database Storage (SQL-based)
- Each record ≈ 2 KB (varies by columns)
- Example Calculation:
- 100,000 Account records × 2 KB = 200 MB
- 1,000,000 Contact records × 2 KB = 2 GB
B. File Storage (Azure Blob)
- Actual file size (1 MB PDF = 1 MB storage)
- Example:
- 10,000 invoices (avg. 500 KB each) = 5 GB
C. Log Storage
- Audit logs ≈ 1 KB per change
- Example:
- 100,000 audit entries = 100 MB
5. Managing & Optimizing Dataverse Storage
A. Monitoring Storage Usage
- Power Platform Admin Center → Analytics → Capacity
- Key Metrics:
- Database used vs. allocated
- File storage consumption
- Log storage trends
B. Reducing Database Storage
✔ Archive old records (e.g., closed cases, historical data)
✔ Use alternate storage (Azure SQL, Data Lake) for analytics
✔ Delete unused tables (custom entities)
C. Optimizing File Storage
✔ Set retention policies (auto-delete old files)
✔ Use SharePoint for large document libraries
✔ Compress files before upload
D. Managing Log Storage
✔ Adjust audit log retention (default = 90 days)
✔ Export & purge old logs (Power Automate)
6. Advanced Storage Strategies
A. Tiered Storage (Hot/Cold Data)
- Hot data: Frequently accessed (keep in Dataverse)
- Cold data: Archived (move to Azure Blob or Data Lake)
B. Data Partitioning (Large Tables)
- Vertical partitioning: Split columns into related tables
- Horizontal partitioning: Archive old records
C. External Data Sources
- Virtual Tables: Query external DBs (SQL, SharePoint)
- Azure Synapse Link: Replicate to data warehouse
7. Troubleshooting Storage Issues
| Issue | Solution |
|---|---|
| “Storage Full” errors | Purchase more capacity or archive data |
| Slow performance | Optimize queries, index columns |
| File upload failures | Check file storage limits |
| Audit logs missing | Adjust retention policy |
Key Takeaways:
✔ Dataverse storage is split between Database (SQL), Files (Blob), and Logs
✔ Monitor usage in Power Platform Admin Center
✔ Optimize by archiving, compressing, and partitioning
Best Practices:
- Regularly audit storage (avoid surprises)
- Use external storage (SharePoint, Azure) for large files
- Purge unused data (old logs, test records)
- Plan for growth (scale storage as needed)