Established by the ALCF and the Mathematics and Computer Science (MCS) Division, the JLSE centralizes Argonne’s research activities aimed at evaluating future extreme-scale computing systems, technologies, and capabilities.
JLSE users leverage existing infrastructure and next-generation hardware and software to explore low-level experimental computer and computational science, including operating systems, messaging, compilers, benchmarking, power measurements, I/O, and new file systems.
By providing access to leading-edge computing resources and fostering collaborative research, the JLSE enables researchers to address Argonne and DOE needs in a variety of areas, which include:
- Improving science productivity on future hardware and software platforms
- Providing an avenue for Argonne researchers to work collaboratively with HPC vendors on prototype technologies for petascale and beyond
- Investigating alternative approaches to current and future system deployments
- Maintaining a range of hardware and software environments for testing research ideas
- Helping to drive standards on benchmarks, programming models, programming languages, memory technologies, etc.
Select JLSE Systems
- Intel Xeon Phi Knights Landing Cluster
- IBM Power System S822LC
- Atos Quantum Learning Machine
- Intel Xeon Platinum Skylake Cluster
- HPE Comanche Prototype ARM64 Cluster
- Kubernetes Cluster with Rancher
- NVIDIA DGX–1
- IBM Elastic Storage Server GL6
For a complete listing, see the JLSE website.