liveBook · Manning

Workflow object

As explained in Designing Deep Learning Systems: A guide for software engineers

An Argo Workflows resource representing one workflow instance together with its definition and execution state.

A workflow object is a live Argo Workflows resource representing one workflow instance. It combines the workflow definition with its execution state and runs natively on Kubernetes. section 9.3.2

A workflow object organizes execution through:

  • An entrypoint template, which identifies the first template to execute. section 9.3.2
  • Templates, which define executable instructions and resemble functions. section 9.3.2
  • Steps that can pass parameters through artifacts such as files. section 9.3.2

How Designing Deep Learning Systems: A guide for software engineers covers it

How deep it goes
Defines object state, entrypoints, templates, steps, and artifact passing, then demonstrates run inspection through Argo and Kubernetes commands.
Examples
Uses a four-step pipeline for checking data, transforming it, saving it to a database, and notifying a data-science team.
What it assumes
Assumes familiarity with Kubernetes while explaining workflow and template responsibilities directly.
What you can do after
Operate workflows by submitting manifests and inspecting runs with Argo CLI or Kubernetes CLI commands.
How it connects
Connects workflow objects with templates, artifacts, Kubernetes pods, and Kubernetes custom resources.
Cover of Designing Deep Learning Systems: A guide for software engineers
liveBookmanning.com
Read Designing Deep Learning Systems: A guide for software engineers on liveBook Get the book at manning.com

This explainer was generated by Manning's knowledge graph from the text of Designing Deep Learning Systems: A guide for software engineers. The section links point to the parts of the book it draws on.