FAST-HEP overview#
Workflows#
FAST-HEP workflows start from an author.yaml, then compile to normalized author and execution-plan files.
flowchart TD
subgraph Compile["Compilation and planning"]
Author["author.yaml"]
Profiles["profiles and registries"]
Normalised["normalised workflow"]
Dependency["dependency inference"]
Plan["execution plan"]
Author --> Normalised
Profiles --> Normalised
Normalised --> Dependency
Dependency --> Plan
end
subgraph Execute["Runtime execution"]
Runtime["runtime execution"]
Outputs["artifacts and outputs"]
Runtime --> Outputs
end
Plan --> Runtime
Implementation packages contribute functionality through registry profiles. The workshop examples show how flow, carpenter, curator, and render profiles combine in an author file.
Packages#
The FAST-HEP ecosystem is split into focused packages.
Each package owns a small part of the workflow stack and exposes functionality through public APIs, profiles, registries, or command-line tools.
flowchart LR
Flow["<b>fasthep-flow</b><br/>workflow compilation<br/>and orchestration"]
Curator["<b>fasthep-curator</b><br/>dataset inspection<br/>metadata and provenance"]
Carpenter["<b>fasthep-carpenter</b><br/>analysis transforms<br/>histograms and cutflows"]
Render["<b>fasthep-render</b><br/>plots, reports<br/>and rendering"]
CLI["<b>fasthep-cli</b><br/>command-line interface"]
Workshop["<b>fasthep-workshop</b><br/>tutorials, examples<br/>and training material"]
Flow --> Carpenter
Curator --> Carpenter
Carpenter --> Render
CLI --> Flow
Workshop --> Flow
Package |
Purpose |
Documentation |
Git Repo |
|---|---|---|---|
|
Workflow compilation, planning, runtime orchestration |
||
|
Analysis transforms, histogramming, awkward/ROOT processing |
||
|
Dataset inspection, provenance, diagnostics, metadata |
||
|
Plotting, reports, rendering pipelines |
||
|
Unified command-line interface |
||
|
Shared utilities and lightweight tooling |
||
|
Tutorials, examples, and training material |
— |
|
|
Meta package and compatibility bundle |
— |
|
|
Integration workspace and ecosystem validation |
— |