8000 docs(architecture): document jobset and workunit state mapping and sync by AntonTyb · Pull Request #1411 · google/xmc · GitHub
[go: up one dir, main page]
Skip to content

docs(architecture): document jobset and workunit state mapping and sync #1411

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
AntonTyb wants to merge 3 commits into
base: main
Choose a base branch
from
+809 −2
Open

docs(architecture): document jobset and workunit state mapping and sync #1411

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
6e34de9
docs(architecture): document jobset and workunit state mapping and sync
AntonTyb Oct 26, 2026
f7dccca
Update jobset_and_work_unit_states.md
AntonTyb Oct 26, 2026
53c84bb
docs(architecture): address PR #1411 review comments on state mapping…
AntonTyb Oct 26, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,7 @@ XManager on Cloud (XMC) is a high-performance backend application that exposes g
Explore the folders and guides below to set up, build, or deploy XMC:

* ### [Architecture Overview](docs/architecture/overview.md)
Start here. The four planes (control, execution, telemetry, components), how the twelve services interact, request routing at the edge, data stores, and cross-cutting concerns. Continues into the [Component Reference](docs/architecture/components.md), the [Execution Model](docs/architecture/execution_model.md), and [Use Cases](docs/architecture/use_cases.md) for ML researchers, platform engineers, and new contributors.
Start here. The four planes (control, execution, telemetry, components), how the twelve services interact, request routing at the edge, data stores, and cross-cutting concerns. Continues into the [Component Reference](docs/architecture/components.md), the [Execution Model](docs/architecture/execution_model.md), [JobSet & WorkUnit States](docs/architecture/jobset_and_work_unit_states.md), and [Use Cases](docs/architecture/use_cases.md) for ML researchers, platform engineers, and new contributors.

* ### [Main Flows](docs/flows/README.md)
Sequence-level walkthroughs of the principal paths: [experiment lifecycle](docs/flows/experiment_lifecycle.md), [work unit lifecycle](docs/flows/work_unit_lifecycle.md) (the asynchronous actuation path), [reconciliation](docs/flows/reconciliation.md), and [metrics & observability](docs/flows/metrics_and_observability.md).
Expand Down
5 changes: 4 additions & 1 deletion docs/architecture/execution_model.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -112,7 +112,10 @@ stateDiagram-v2
```

The diagram shows the transitions the platform drives; the authoritative enum and its comments are
in [`status.proto`](../../api/protos/experiment/status.proto).
in [`status.proto`](../../api/protos/experiment/status.proto). For the exact `JobSet`-to-`WorkUnitState`
evaluation priority rules, storage transition guards, dual-loop synchronization mechanics, and
end-to-end event sequences, see
[JobSet, WorkUnitExecution, and WorkUnit States](jobset_and_work_unit_states.md).

Two derived signals matter operationally:

Expand Down
Loading
Loading
0