Expanse Documentation
Expanse is an HPC workflow orchestration platform that simplifies running computational workloads across multiple clusters. Expanse Core provides the complete infrastructure for workflow execution, data transfer, and governance, giving you everything needed to run production workloads while we fine-tune failure prediction models on your specific data.
What's Included in Expanse Core
- CLI tool for workflow execution and management
- Zero-copy data transfer between clusters using Apache Arrow IPC
- Node registry for managing compute nodes and workflows
- Cross-language support (Python, C, Fortran) with automatic runtime bootstrapping
- Software licence management across clusters
- Governance and audit trails for job provenance
Coming in Expanse Pro
Expanse Pro Preview
ML-based failure prediction, fine-tuned on your workload data. As we collect telemetry from your jobs during Expanse Core, we'll train models specific to your infrastructure and workload patterns.
Quick Links
- Getting Started - Install and run your first workflow
- Nodes - Define computational steps
- Workflows - Chain nodes into pipelines
- Data Transfer - Cross-language, cross-cluster data flow
- CLI Reference - Full command documentation