In November 2018 we announced the imminent arrival at the University of the largest IBM POWER9 AI cluster in the UK. We are now inviting pilot users on to the system. If you are interested in using this powerful AI resource then please contact us by creating an "Other BEAR Request" ticket in the Advanced Research Computing Service section of the IT Service Desk, as described here. We are particularly looking for people who use TensorFlow, PyTorch, or other GPU-accelerated software to contact us. We will help you use the service.
As of April 2019, the BEAR AI service consists of three parts:
1. Three POWER9 HPC nodes in BlueBEAR, each with four NVIDIA V100 GPUs and 1TB of RAM. Any researcher in the University may apply to access these nodes to run GPU-accelerated software.
2. Four POWER9 HPC nodes in the special CaStLeS part of BlueBEAR, each with four NVIDIA V100 GPUs and 1TB of RAM. Access to these resources is available for Life Sciences researchers. For further information, including how to apply for access see the CaStLeS overview.
3. ARC are in the process of commissioning four further POWER9 nodes for CaStLeS to run Watson Machine Learning Accelerator. Access to these resources will be available for Life Sciences researchers. We will announce a call for pilot users for this service in due course.
Each of our BEAR AI systems has:
- Dual IBM POWER9 CPUs with 18 cores each, which currently present themselves as 144 cores using simultaneous multithreading (SMT4).
- Four NVIDIA Tesla V100 Tensor Core GPUs
- 1 TB system memory
- High speed NVIDIA NVLink interconnect fully meshed between the GPUs and also into the system memory
- 100G InfiniBand interconnect to other nodes and storage systems