The mission of the Extreme-scale Algorithms and Solver Resilience (EASIR) project is to close the performance gap between the peak capabilities of HPC hardware and the performance realized by high performance computing applications. To carry out this mission, the EASIR project team develops architecture-aware algorithms and libraries, and the supporting runtime capabilities, to achieve scalable performance and resilience on heterogeneous architectures.
The project team includes personnel from ORNL, Sandia National Laboratories, University of Illinois, University of California Berkeley, and the University of Tennessee (ICL). ICL’s efforts focus on providing components and services in a vertically integrated software stack, from low-level runtime process and thread scheduling to multicore aware library interfaces, multicore dense linear algebra, scalable iterative methods, and advanced parallel algorithms that break traditional parallelism bottlenecks.
In Collaboration With
- Oak Ridge National Laboratory
- Sandia National Laboratories
- University of California Berkeley