Infrastructure or equipment
The HPC facilities of the KU Leuven consist of a Tier-1 and a Tier-2 cluster.
Tier-1 thin node supercomputer BrENIAC
This system has been in use since October 2016. On this cluster one can run highly parallel, large scale computations that rely critically on efficient communication. The hardware consists of:
- 580 computing nodes, each with two 14-core Intel Xeon processors (Broadwell, E5-2680v4) and 128 GiB RAM (435 nodes) or 256 GiB (145 nodes)
- EDR InfiniBand interconnect, High bandwidth (11.75 GB/s per direction, per link)
- Storage system (capacity of 634 TB, Peak bandwidth of 20 GB/s) BrENIAC thus contains 16.240 compute cores and has a peak performance of more than 600 TFlops and 634 TB storage.
The KU Leuven/UHasselt Tier-2 cluster (ThinKing and Cerebro)
The Tier2 infrastructure of the KU Leuven/UHasselt consists of different sections:
Compute nodes:
Thin node section with a total memory capacity of this section is 30 TB, the total peak performance is about 232 Tflops in double precision arithmetic.
SMP section (also known as Cerebro): a SGI UV2000 system with 64 sockets with a 10-core "Ivy Bridge" Xeon E5-4650 CPU (2.4 GHz, 25 MB level 3 cache), spread over 32 blades and connected through a SGI-proprietary NUMAlink6 interconnect.
Accelerator section:
- 5 nodes with two 10-core "Haswell" Xeon E5-2650v3 2.3GHz CPUs, 64 GB of RAM and 2 GPUs Tesla K40.
- The central GPU and Xeon Phi system is also integrated in the cluster.
- 8 nodes have two nVidia K20x cards each installed. Each K20x has 14 SMX processors (Kepler family; total of 2688 CUDA cores) that run at 732MHz and 6 GB of GDDR5 memory with a peak memory bandwidth of 250 GB/s (384-bit interface @ 2.6 GHz). The peak floating point performance per card is 1.31 Tflops in double and 3.95 Tflops in single precision.
- 20 nodes have four Nvidia Tesla P100 SXM2 cards each installed (3584 CUDA cores @1328 MHz, 5.3 DP Tflops/GPU).
Central storage available to all nodes:
- A NetApp NAS system with 30 TB of storage, used for the home-and permanent data directories. All data is mirrored almost instantaneously.
- A 284 TB GPFS parallel file system from DDN, mostly used for temporary disk space.
- A 600 TB archive storage optimised for capacity and aimed at long-term storage of very infrequently accessed data.