Evaluation Testbeds

Through Argonne’s Joint Laboratory for System Evaluation (JLSE), the ALCF provides access to leading-edge hardware and software for research activities aimed at evaluating future extreme-scale computing systems, technologies, and capabilities.

Joint Laboratory for System Evaluation (JLSE) is a collaboration between the CELS computing divisions including Leadership Computing Facility (LCF), Mathematics and Computer Science (MCS), Computational Science (CPS) and Data Science and Learning (DSL) with the aim of evaluating future high-performance computing platforms.

These activities will be conducted with the goal of addressing Argonne’s needs in a variety of areas. Examples include:

  • Improving science productivity on current and future ALCF platforms.
  • Investigating alternative approaches to current and future deployments (both hardware and software) within ALCF.
  • Maintaining a range of hardware and software environments for testing research ideas.
  • Helping to drive standards in standard forums on benchmarks, programming models, programming languages, memory technology, etc.

Select JLSE Testbeds

  • Arcticus, DevEP, Iris: Intel discrete and integrated GPU testbeds for ECP and ESP projects to develop, optimize, and scale applications and software for Aurora
  • Aurora Software Development Kit: Frequently updated version of the publicly available Intel oneAPI toolkit for Aurora development
  • Arm Ecosystem: Apollo 80 Fujitsu A64FX Arm system, NVIDIA Ampere Arm and A100 test kits, and an HPE Comanche with Marvell ARM64 CPU platform provide an ecosystem for porting applications and measuring performance on next-generation systems
  • Atos Quantum Learning Machine: Platform for testing and developing quantum algorithms and applications
  • Intel Xeon Clusters: Cascade Lake, Skylake, and Cooper Lake Xeon clusters enable a variety of research activities, including testing AI and learning applications
  • NVIDIA and AMD GPUs: Clusters of NVIDIA V100, A100, and A40 GPUs, and AMD MI50 and MI100 GPUs for preparing applications for heterogenous computing architectures
  • Presque: Intel DAOS nodes for testing the Aurora storage system

For a complete listing, see the JLSE website.