JADE II (Tier 2 GPU cluster)

Warning

JADE II is retired. The operator’s final withdrawal timetable specified:

  • 1 September 2024: registration of new groups and accounts ended.

  • 15 November 2024: batch and interactive compute access was withdrawn, extended from the originally announced date of 1 November.

  • 6 January 2025: all service access was withdrawn and physical decommissioning began.

This page retains the specifications of the former service.

The JADE II cluster was a 2020 renewal of the Joint Academic Data Science Endeavour (JADE) and served as a leading GPU facility in the UK supporting world-leading research in machine learning.

The cluster comprised 63 NVIDIA DGX MAX-Q servers, each containing 8 NVIDIA Tesla V100 GPUs linked by NVIDIA NVLink interconnects.

Members of the University of Sheffield were able to access this resource for free (for use with deep learning research).

JADE II Specification

Hardware

  • 63 Nodes of DGX MAX-Q, each with:

    • 8x Nvidia V100 32GB GPUs

    • 512GB RAM

  • 70TB DDN AI400 shared storage (NVMe) for read intensive/streaming applications

  • 1PB Lustre shared storage (spinning disk)

  • EDR infiniband interconnect

Software

  • Redhat Enterprise Linux 8