About JADE

JADE is a UK Tier-2 resource, funded by EPSRC, owned by University of Oxford and hosted at the Hartree Centre. The hardware was supplied and integrated by ATOS Bull.

A consortium of UK universities and The Alan Turing Institute, led by the University of Oxford, has been awarded £5 million by the Engineering and Physical Sciences Research Council (EPSRC) to continue the world leading research enabled by the Joint Academic Data science Endeavour (JADE) service. This forms part of a continued investment by EPSRC in the UK’s regional Tier 2 high-performance computing facilities, which aim to bridge the gap between institutional and national resources.

JADE is unique amongst the Tier 2 centres in being designed for the needs of machine learning and related data science applications. There has been huge growth in machine learning in the last 5 years, and JADE was the first national facility to support this rapid development. JADE will accelerate the research of world-leading machine learning groups in universities across the UK and national centres such as The Alan Turing Institute and Hartree Centre.

The system design exploits the capabilities of NVIDIA's MAXQ Deep Learning System which has eight of its Tesla V100 GPUs tightly coupled by its high-speed NVlink interconnect. NVIDIA has clearly established itself as the leader in massively-parallel computing for deep neural networks, and the DGX MAXQ runs optimized versions of many standard machine learning software packages such as Caffe, TensorFlow, Theano and Torch.

This system design is also ideal for a large number of molecular dynamics applications, as such JADE will also provide a powerful resource for molecular dynamics researchers within the UK HECBioSim community.

Img Img Img
Img

SYSTEM SPECIFICATION

  • 63 NVIDIA MAXQ Deep Learning Systems comprising:
    1. 8 of NVIDIA's Tesla V100 GPUs
    2. NVIDIA's high-speed NVlink interconnect
    3. 4 TB of SSD for machine learning datasets
  • over 1PB of DDN AI400 storage including 70TB of NVMe storage
  • Mellanox EDR networking
  • optimized versions of major machine learning software packages such as Caffe, TensorFlow, Theano and Torch
  • system integration/delivery by Atos, hosting by STFC Hartree
  • system management by Atos / STFC Hartree

           

QMUL logo      

           

We are always looking for new ideas and feedback.

Any questions or comments, please report it via GitHub issue tracker.

GitHub