Batch Runner

Overview

Batch Runner is an Apify actor that manages multiple actor runs from a single input. It handles execution sequencing, resource constraints, and can optionally merge datasets from the completed runs.

Key Features

• Sequential Batch Execution: Runs a list of actors or existing runs one after another.
• Resource Management: Starts new runs only when platform memory limits allow, preventing overconsumption errors.
• Dataset Merging: Optionally combines the default datasets from all successful runs into a single output dataset.
• Flexible Failure Handling: Configurable to either fail the entire batch on any run failure or continue with remaining tasks.

How to Use

Configure the actor with an input object specifying the tasks to run and execution options. The actor processes each item in the batch array in order, waiting for resources and completion as needed.

Input/Output

Input Schema
The input is a JSON object with the following properties:

• batch (Array, Required): A list of tasks. Each task can be:
  • An object with a runId to wait for an existing run.
  • An object defining a new actor run (actorId, input, timeout, build, memoryMb).
• mergeDatasets (Boolean, Optional, Default: false): If true, merges items from the default dataset of each successful run into the Batch Runner's own default dataset.
• failOnAnyRunFails (Boolean, Optional, Default: false): If true, the entire batch fails immediately when any individual run fails. If false, it continues with the remaining tasks.

Example Input:

{
  "batch": [
    { "runId": "UuwtsB8GJ4JpsqTdh" },
    {
      "actorId": "useful-tools/wait-and-finish",
      "input": { "minWaitTime": 30 },
      "memoryMb": 256
    }
  ],
  "mergeDatasets": true,
  "failOnAnyRunFails": false
}

Output
The actor's run output contains a summary. If mergeDatasets is enabled, the merged results are available in the actor's default dataset.