Three universities are coming together in a project known as the Aristotle Cloud Federation to build a federated cloud system designed to help researchers who need flexible workflows and analysis tools for large data sets.
The name comes from Aristotle’s quote:
“The whole is greater than the sum of its parts”,
Reflecting the hope that multiple institutions and collaborations will become part of the project. The Aristotle project recently received a grant for the $5 million, five year project from the National Science Foundation to build the system.
Research and Sciences Based Support
Lead by Cornell University, University at Buffalo, and University of California-Santa Barbara, the cloud will be shared by seven science teams with over forty global collaborators. The three lead universities have each built a private cloud based on the software Eucalyptus, an open source project with a commercial product from Hewlett Packard Enterprise.
Initials users of the Aristotle Cloud Federation will include those involved in:
- Astronomy *Earth and atmospheric sciences
- Civil engineering
- Food sciences
These areas were chosen to best demonstrate the value of sharing resources and data among institutions while minimizing the wait time to perform large scale calculations.
Because the Eucalyptus software is compatible with Amazon Web Services, workloads may use those services also (if needed).
Project leaders are David Lifka, Director of the Cornell University Center for Advanced Computing (CAC) will lead the project with colleagues Tom Furlani, Director of the UB Center for Computational Research, and Rich Wolski, Professor of Computer Science at UCSB.
“The goal of the Aristotle Cloud Federation is to develop a federated cloud model that encourages and rewards institutions for sharing large-scale data analysis resources…”
Supercomputers at locations can be used for work on enormous data sets, but the wait time to use them can be up to a week. The Aristotle Cloud Federation will enable the resources to be efficiently allocated while optimizing workloads by predicting where the item will run best.
The objective, project leaders say, is to maximize “time to science” – the time it takes a researcher to actually obtain their scientific results.
Documentation, components, and best practices developed by the Aristotle team through the grant will be made available to the national community to further innovate in research and other sharing situations.