Documentation Index
Fetch the complete documentation index at: https://mintlify.com/QucoonAI/mcsp_docs/llms.txt
Use this file to discover all available pages before exploring further.
These figures are planning targets, not contractual commitments. All architectural decisions in this documentation are validated against the Year 1 figures. Year 3 projections are used to verify that the architecture does not require fundamental rework at projected growth.
Traffic Scale
| Metric | Year 1 Estimate | Year 3 Projection |
|---|---|---|
| Registered users | 5 million | 20 million |
| Monthly active users | 2 million | 8 million |
| Peak concurrent streams | 150,000 | 600,000 |
| Content uploads per day | 15,000 items | 60,000 items |
| Total content items | 3 million | 20 million |
| Peak API RPS | 50,000 RPS | 200,000 RPS |
| Peak CDN egress | ~1.5 Tbps | ~6 Tbps |
| Active creator accounts | 50,000 | 200,000 |
| Monthly subscription transactions | 200,000 | 1 million |
Bandwidth Profile
| Scenario | Estimate |
|---|---|
| Average video stream bitrate (blended 720p/1080p) | ~3.5 Mbps |
| Average audio stream bitrate (blended 128/256 kbps) | ~0.2 Mbps |
| 150,000 concurrent streams (~10% audio) | ~480 Gbps sustained |
| Peak factor (1.5×) | ~720 Gbps |
| CDN over-provisioning buffer (2×) | ~1.5 Tbps capacity required |
| Nigeria origin bandwidth (residency content) | ~10 Gbps |
| Upload ingress | ~5 Gbps (15,000 video + 5,000 audio uploads/day) |
Storage Growth Model — Year 1
| Component | Estimate |
|---|---|
| Raw video (pre-transcode staging, 30-day retention) | ~290 TB |
| Raw audio (pre-transcode staging, 30-day retention) | ~10 TB |
| Transcoded video — hot tier (top 10% of catalogue, all resolutions) | ~1.9 PB |
| Transcoded audio — hot tier (top 10% of audio catalogue, all bitrate variants) | ~50 TB |
| Transcoded video — cold tier (remaining video catalogue) | ~4.8 PB |
| Transcoded audio — cold tier (remaining audio catalogue) | ~150 TB |
| Nigeria Residency storage (~5% of uploads) | ~200 TB |
| Cover art, thumbnails, and metadata | ~20 TB |
| ML training data (user events, feature store) | ~100 TB |
| Logs, metrics, audit data | ~50 TB |
| Total (Year 1) | ~7.6 PB |