Unified Communication X (UCX) Community

Unified Communication X (UCX) Community

Tuesday, May 31, 2022 9:00 AM to 10:00 AM · 1 hr. (Europe/Berlin)
Hall E - 2nd Floor
Exascale SystemsHPC Workflows

Information

n order to exploit the capabilities of new HPC systems and to meet their demands in scalability, communication software needs to scale on millions of cores and support applications with adequate functionality to express their parallelism. UCX is a collaboration between industry, national labs and academia that consolidates multiple technologies that provides a unified open source framework. The UCX project is managed by the UCF consortium (http://www.ucfconsortium.org/) and includes members from LANL, ORNL, ANL, Ohio State University, AMD, ARM, IBM, NVIDIA and more. The session will serves as the UCX community meeting, and will introduce the latest development and specification to HPC developers and the broader user community.

High-level programming models for communication (e.g., MPI, SHMEM) can be built on top of middleware, such as Portals, GASNet, UCCS, and ARMCI or use lower-level network-specific interfaces, often provided by the vendor. While the former offer high-level communication abstractions and portability across different systems, the latter offer proximity to the hardware and minimize overheads related to multiple software layers. An effort to combine the advantages of both is UCX, a communication framework for high-performance computing systems.

UCX has already been integrated with upstream of Open MPI project and OpenSHMEM, being used with MPICH and more. UCX is now being deployed in many supercomputers worldwide and US DOE systems. The session will enable a dialog on the future plans for UCX and review the operations of the UCX consortium. It will include performance evaluation based on testing results conducted on several supercomputing platforms.
Contributors:

  • Dhabaleswar K. Panda (The Ohio State University)
  • Gilad Shainer (NVIDIA)
  • Yanfei Guo (Argonne National Laboratory)
  • Steve Poole (Los Alamos National Laboratory)
  • Jeff Kuehn (Los Alamos National Laboratory)
  • Pavel Shamis (Arm)
  • Perry Schmidt (IBM)
  • Jeff Kuehn (AMD)
Format
On-site