The University of Michigan and IT major IBM announced that they will jointly develop and deliver “data-centric” supercomputing systems.
The collaboration was announced in San Jose at the second annual OpenPOWER Summit 2016.
With this partnership, the university intends to increase the pace of scientific discovery in various fields such as aircraft and rocket engine design, cardiovascular disease treatment, materials physics, climate modelling and cosmology.
IBM said the system would enable computing applications for physics to interact, in real time, with big data to improve scientists’ ability to make quantitative predictions.
In connection with the mission, U-M researchers have designed a computing resource called ConFlux with IBM’s guidance.
This enables high performance computing clusters to communicate directly and at interactive speeds with data-intensive operations. ConFlux is funded by a grant from the National Science Foundation.
The project is hosted at U-M with an ecosystem of hardware and software. It is expected benefit large-scale data-driven modeling of complex physical problems, such as the performance of an aircraft engine, which consists of trillions of molecular interactions.
“There is a pressing need for data-driven predictive modeling to help re-envision traditional computing models in our pursuit to bring forth groundbreaking research,” said Karthik Duraisamy, assistant professor in the U-M Department of Aerospace Engineering and director of U-M’s Center for Data-driven Computational Physics.
IBM said ConFlux meshes well with its recent focus on data-centric computing systems.
U-M is also studying cardiovascular disease for the National Institutes of Health.
A combination of noninvasive imaging such as results from MRI and CT scans with a physical model of blood flow, U-M hopes to help doctors estimate artery stiffness within an hour of a scan.
Such a facility would serve as an early predictor of diseases like hypertension..
Furthermore, the university has also planned studies to better understand climate science.