Pegasus (workflow management)
| Company type | Open-source software |
|---|---|
| Industry | Scientific workflow management |
| Founded | 2001 |
| Founder | Information Sciences Institute, University of Southern California; University of Wisconsin-Madison |
Area served | Scientific research communities |
| Products | Workflow management system |
Pegasus represents an open-source system for workflow management created through a collaboration between the Information Sciences Institute at the University of Southern California and the University of Wisconsin-Madison. It empowers researchers to construct and run scientific workflows across a variety of computational infrastructures, including national cyberinfrastructure, cloud platforms, and high-performance computing clusters. Pegasus models workflows as directed acyclic graphs (DAGs) and provides APIs compatible with environments such as Jupyter Notebooks, Python, R, and Java. When workflows are executed, it transforms these abstract representations into runnable workflows controlled by HTCondor.
This platform caters to numerous scientific disciplines like helioseismology, earthquake engineering, bioinformatics, astronomy, and gravitational-wave physics. A significant achievement of Pegasus was its instrumental role in the LIGO Scientific Collaboration’s inaugural direct observation of gravitational waves, highlighting its strength in managing sophisticated and data-heavy scientific processes.
Implemented in Java, Python, and C, Pegasus runs on Linux and macOS operating systems and is available under the Apache License 2.0. Its architecture is designed to support scalable workflow orchestration critical for progressing computational research in various scientific domains.
Area of applications
Primary fields where Pegasus is utilized include:
- Gravitational-Wave Physics
- Earthquake Science
- Bioinformatics
- Volcanic Mass Flow Workflows
- Diffusion Image Processing and Analysis
- Spallation Neutron Source (SNS)
History
The inception of Pegasus dates back to 2001.