Cloud Technology for Digital Media Technology
Start at Session 0
New to the course? Session 0 — Start here carries the accounts, the tools, the student kit, the safe-data policy and the readiness checklist in one page. Do it before the first session.
A ten-session course for a digital-media PhD cohort. Ten four-hour sessions on a DigitalOcean, Cloudflare and OpenRouter stack, a lab ladder, a research question each week, and a capstone that reuses the same campus media unit. English is the course language; every lesson page also carries optional Thai and Simplified Chinese summaries in collapsible blocks.
What must be ready before each session
The ladder below is what each session assumes you already have. Every account, the domain, the SSH keys and the GPU runtimes are in Session 0; do a session's row before that session starts.
| Session | What must be ready | Set it up in Session 0 |
|---|---|---|
| 01 Cloud foundations | A DigitalOcean account with two-factor authentication and a backup method, the instructor added as a Resource Viewer and a $10 spend alert; an SSH key pair and a scoped API token. | Accounts, SSH keys |
| 02 Performance and latency | A root domain registered and an Active Cloudflare zone for it; a DigitalOcean scoped token with droplet:update. | Domain, Accounts |
| 03 Storage, RAID and ZFS | The DigitalOcean account, project and token from Session 1; the lab creates its own Block Storage volumes and tears them down. | Accounts |
| 04 Networking and delivery | An Active Cloudflare zone for your root domain and a Cloudflare account with R2 enabled. | Domain, Accounts |
| 05 Containers and the Nextcloud NAS | The DigitalOcean account, project and SSH key from Session 1; an Active Cloudflare zone with a Zone:DNS:Edit token for the Nextcloud hostname. | Accounts |
| 06 Economics and vendor lock | A Linode/Akamai account with a payment method, two-factor authentication and a per-service token, plus the DigitalOcean, Cloudflare and OpenRouter accounts. | Accounts |
| 07 Security and identity | The Session 1 SSH key and DigitalOcean token, and a Cloudflare API token with the scopes the lab names. | SSH keys, Accounts |
| 08 Sensitive data and managed AI | The OpenRouter account and API key created in Session 6, and the course synthetic-media pack; never substitute real personal media. | Accounts, Safe data |
| 09 LLM inference and fine-tuning | A phone-verified Kaggle account, or an Apple-silicon Mac with the kit's S9 dependency group; internet access for Kaggle's install cell. | Accounts, Tools, The kit |
| 10 Image ML on a GPU + capstone | A phone-verified Kaggle account for the T4 run and the faculty RTX card (or the RunPod fallback); the kit's S10 dependency group or the light install. | Accounts, Tools |
Every document the site publishes opens as a rendered view: each session's lab.md (the numbered lab brief), refs.md (the reference list) and PLO.md (that session's scoring rows), the master PLO.md, the starter-pack README.md and COMMANDS.md files, and the dataset README.md files. Each rendered page carries a download link for the raw Markdown working file — the copy you edit, run and submit; notebook.ipynb (the runnable notebook) is a download only. The rendered pages are for reading and the raw files are what you work in; the lesson pages carry the teaching.
The ten sessions
| # | Session | Lab | PLO | Points |
|---|---|---|---|---|
| 01 | Cloud foundations | Account, first VM, firewall and spend alert | PLO1 / PLO3 | 7 |
| 02 | Performance and latency | Latency, sizing and the edge | PLO1 / PLO2 | 7 |
| 03 | Storage, RAID and ZFS | Volumes, RAID 5/6/10 and ZFS | PLO2 / PLO1 | 7 |
| 04 | Networking and delivery | Cloudflare CDN, R2 and the Worker sample | PLO2 / PLO1 | 7 |
| 05 | Containers and the Nextcloud NAS | Docker Compose and a Nextcloud NAS | PLO2 / PLO1 | 7 |
| 06 | Economics and vendor lock | TCO, the move, and the portability scorecard | PLO1 / PLO3 | 7 |
| 07 | Security and identity | SSH keys, DO/Cloudflare tokens and signed R2 access | PLO3 / PLO2 | 7 |
| 08 | Sensitive data and managed AI | Managed inference, ZDR and PII redaction | PLO3 / PLO1 | 7 |
| 09 | LLM inference and fine-tuning | QLoRA fine-tune on the caption catalogue | PLO2 / PLO3 | 7 |
| 10 | Image ML on a GPU + capstone | Three-configuration training comparison; capstone brief | PLO1 / PLO2 / PLO3 | 7 |
PLO weighting
| PLO | Points | Where it is evidenced |
|---|---|---|
| PLO1 — evaluate architectures | 30 | Weekly research answers and the capstone proposal |
| PLO2 — build and operate | 40 | Weekly lab evidence and the capstone implementation |
| PLO3 — analyse responsibility and risk | 30 | Weekly research and lab evidence, the risk register, and the capstone report |
Weekly activities 70 (research 40 + lab 30) + capstone 30 (proposal 5, implementation 15, report/presentation 10) = 100 points. The master reconciliation is PLO.md (rendered view; its page links the raw Markdown working file).
Capstone
Design and prototype the next phase of the campus media unit's catalogue: add AI tagging to the 5,000-asset library, with a governance rule, inside a fixed monthly budget. The stack is the course's: private R2 for the catalogue and tags, public R2 for the thumbnails, the Nextcloud/RAID NAS as origin and backup, the S4 Worker as the catalogue API and ingest gate, and OpenRouter as the governed inference path. All 5,000 assets are tagged in checkpointed batches of 100 with at least 99% schema-valid records; the budget is the $80 spine with the $5 inference line and the $10 backfill cap, and cloud resources run in a bounded 72-hour window. Proposal 5 points, implementation 15, report and presentation 10. The starter pack and its command list are in samples/capstone/; both open as rendered views and link their raw Markdown working files. The rubric and the ten-item evidence bundle are in the assessment scheme.