mesh-job-orchestrator

What problem does it solve?

Creates and manages distributed compute jobs by decomposing them into work units using MapReduce, Pipeline, ScatterGather, or LayerSplit strategies, enabling scalable execution across multiple workers.

Core Features & Use Cases

• Decomposes large jobs into manageable work units for parallel processing across a compute mesh.
• Tracks progress, retries failed units, and aggregates results to provide end-to-end visibility.
• Use cases include large-scale data processing, parallel workflows, and heterogeneous environments requiring dynamic scheduling.

Quick Start

Create a new distributed compute job with a chosen strategy and payloads to start orchestration and monitor progress.