tegwick
c521adc2f9
Add low/base/high demand forecasts, service objectives with timestamped observations, and workload operations labor for resource:tenant:coulomb:coulomb-social.
3.1 KiB
| id | type | title | domain | repo | status | owner | topic_slug | created | updated | related | state_hub_workstream_id | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CSOC-WP-0005 | workplan | Publish Coulomb Social resource demand and cost evidence | communication | coulomb-social | finished | codex | coulomb-social | 2026-08-11 | 2026-08-12 |
|
00de1578-6150-40d3-b193-e852666ed945 |
CSOC-WP-0005 — resource demand and cost evidence
Goal
Publish workload-owned requirements and observations for
resource:tenant:coulomb:coulomb-social without moving deployment, platform,
or financial authority into this repository. Origin: RESOURCE-WP-0003-T04
delegated evidence gap.
Finished 2026-08-12: evidence under docs/resource-evidence/.
T01 — Declare low, base, and high workload demand
id: CSOC-WP-0005-T01
status: done
priority: high
state_hub_task_id: "d22e42e7-186e-4206-9cf4-c533d6388b13"
Define forecast proxies for active tenants/users, requests, background work, database size/growth, stored media, ingress/egress, and required CPU/memory. Preserve assumptions and distinguish current Coulomb use from future external multi-tenant scenarios.
Done when resource-control can produce falsifiable monthly demand forecasts.
2026-08-12: docs/resource-evidence/demand-forecast-2026-08.{md,json} —
low/base/high with units, assumptions, instance-per-client scope (external
clients = separate instances), labor € proxy, capacity floor from chart.
T02 — Publish service objectives and usage observations
id: CSOC-WP-0005-T02
status: done
priority: high
state_hub_task_id: "ac078211-3cd9-49d8-baac-c0a1b60278fb"
Declare availability, latency, recovery, retention, and tenant-isolation requirements. Publish timestamped non-secret observations for the demand proxies available to the application while leaving cluster telemetry to its owners.
Done when infrastructure options can be evaluated against application-owned requirements and forecast error can be measured.
2026-08-12: docs/resource-evidence/service-objectives-and-observations-2026-08.{md,json}
— SLOs + timestamped measures; explicit unknown for prod DB size, member
count, and live CPU/memory (cluster/DB authority).
T03 — Record workload operations labor
id: CSOC-WP-0005-T03
status: done
priority: medium
state_hub_task_id: "a15e630c-6bec-4c50-9622-bce16e1ce418"
Identify setup and recurring application labor for releases, migrations, tenant operations, incident response, and recovery exercises. Separate it from cluster, database, identity, and provider labor.
Done when resource-control can include workload labor without double counting platform operations.
2026-08-12: docs/resource-evidence/workload-operations-labor-2026-08.md —
setup sunk vs residual, recurring low/base/high hours matching demand, out-of
scope table for platform labor.
Acceptance
- Demand forecasts distinguish current, growth, and external-tenant cases.
- Service objectives and observations have stable units and provenance.
- Workload labor is separated from delegated infrastructure labor.
Index
docs/resource-evidence/README.md