Storage and Recording Capacity
Use this page to plan recording storage growth, retention policy, and backend-specific lifecycle controls.
For baseline deployment prerequisites and session-recording compatibility checks, start with SRA Requirements.
For recording retrieval and operational review, use RDP Recordings, Web Access Session Recording, and Session Management.
Capacity Model by Session Type
Recorded size depends on codec settings, session duration, and user interaction density.
Use these planning ranges as initial estimates:
| Session type | Typical artifact | Planning range | Lower-band scenario | Upper-band scenario |
|---|---|---|---|---|
| RDP video recording | Encoded video (.enc or .enc.gzip) | Approximately 150 MB to 1.5 GB per hour | Mostly static admin tasks, low screen-change rate, shorter sessions | High-motion UI activity, frequent full-screen refreshes, longer continuous sessions |
| Web access recording | Browser session video artifact | Approximately 100 MB to 1.2 GB per hour | Form-based workflows and low navigation churn | Media-heavy pages, frequent page reloads, high interaction density |
| SSH and terminal sessions | Text command transcript and metadata | Usually much smaller than video artifacts; plan by retention count and audit needs | Short command-only sessions with minimal output | Long sessions with verbose command output and frequent file/content dumps |
Treat these ranges as sizing inputs, not hard limits. Validate with representative workloads before final capacity commitments.
How to Estimate Where You Fall in the Range
Use this quick method to choose a realistic planning band:
- Sample 3 to 5 representative sessions per access type (RDP, web, SSH).
- Capture duration and resulting artifact size for each sample.
- Convert each sample to hourly size (
artifact_size / session_minutes * 60). - Use the 75th percentile of your sample set as the baseline planning value.
- Add a 20% to 30% buffer for peak periods and retention-policy overlap.
For initial planning before full measurements are available:
- Use the lower half of the range for low-interaction operational access.
- Use the upper half of the range for high-activity troubleshooting, GUI-heavy work, or long-lived sessions.
Recording Storage Backends
RDP recording storage supports:
- Amazon S3
- S3-compatible object storage
- Azure Blob Storage
- Local filesystem on the SRA host
Use RDP Recordings for backend configuration flags and examples.
Retention and Cleanup Strategy
Plan retention at two levels:
- Platform-level retention policy (bucket/container lifecycle rules).
- Runtime-level operational cleanup policy (for local storage and non-lifecycle paths).
Recommended approach:
- Keep raw storage retention short for high-volume environments.
- Archive only required compliance windows.
- Separate hot review window from long-term archival policy.
Backend-Specific Lifecycle Guidance
Amazon S3
Use S3 lifecycle rules at bucket or prefix level to expire, transition, or archive recording objects automatically.
Azure Blob Storage
Use Azure lifecycle management and tiering policies to move older recordings to lower-cost tiers.
GCS and Other Archive Targets
If recordings are copied to Google Cloud Storage or another archive platform in your post-processing pipeline, apply equivalent object lifecycle rules there as well.
S3-Compatible Storage
For S3-compatible systems, apply vendor-specific object lifecycle policy at bucket level and validate API compatibility for lifecycle enforcement.
Local Storage Limits and Production Risk
Local storage is useful for testing and short-lived environments, but is risky for long-term production retention.
Primary risks:
- Host disk exhaustion can interrupt recording workflows.
- Storage scaling is tied to host lifecycle operations.
- Disaster recovery is harder without external object storage replication.
For production environments, prefer object storage backends and reserve local storage for temporary buffering or non-critical scenarios.