Our lab mainly utilizes massively parallel MD simulation package OpenMM using graphical processing units (GPUs). Our lab creates a high-performance supercomputing ecosystem equipped with >100 GPUs (mostly GeForce 30 series + 10 RTX 4090 !), which enables a single simulation across multiple GPUs, or alternatively run a different simulation on each one at the same time.
GPU clusters (12 Nodes, 150 GPUs, 5,000 Tflops)
4th generation GPU cluster (RTX 4090 x 10 GPUs, 900 Tflops)
Server 15
- Integrated GPU : 10 GPUs (NVIDIA GeForce RTX 4090)
- Node : Intel(R) Xeon(R) Gold 6242R CPU (3.10GHz, 40 Cores) * 2 / 250 GB RAM
3rd generation GPU cluster (10 Nodes, 104 GPUs, 3626 Tflops)
Server 1
- Integrated GPU : 8 GPUs (NVIDIA GeForce RTX 3090)
- Node : Intel(R) Xeon(R) Gold 6230 CPU (2.10GHz, 40 Cores) * 2 / 250 GB RAM
Server 3
- Integrated GPU : 10 GPUs (NVIDIA GeForce RTX 3080 Ti)
- Node : Intel(R) Xeon(R) CPU E5-2680 v4 (2.40GHz, 28 Cores) * 2 / 250 GB RAM
Server 5
- Integrated GPU : 8 GPUs (NVIDIA GeForce RTX 3080 Ti)
- Node : Intel(R) Xeon(R) Gold 6242R CPU (3.10GHz, 40 Cores) * 2 / 250 GB RAM
Server 6
- Integrated GPU : 12 GPUs (NVIDIA GeForce RTX 3080 Ti)
- Node : Intel(R) Xeon(R) CPU E5-2699 v4 (2.20GHz, 44 Cores) * 2 / 250 GB RAM
Server 11
- Integrated GPU : 12 GPUs (NVIDIA GeForce RTX 3090)
- Node : Intel(R) Xeon(R) Gold 6242R CPU (3.10GHz, 40 Cores) * 2 / 250 GB RAM
Server 12
- Integrated GPU : 12 GPUs (NVIDIA GeFOrce RTX 3090)
- Node : Intel(R) Xeon(R) Gold 6242R CPU (3.10GHz, 40 Cores) * 2 / 250 GB RAM
Server 13
- Integrated GPU : 8 GPUs (NVIDIA GeForce RTX 3090)
- Node : Intel(R) Xeon(R) Gold 6230R CPU (2.10GHz, 52 Cores) * 2 / 125 GB RAM
Server 14
- Integrated GPU : 12 GPUs (NVIDIA GeForce RTX 3080 Ti)
- Node : Intel(R) Xeon(R) Gold 6242R CPU (3.10GHz, 40 Cores) * 2 / 250 GB RAM
Server 16
- Integrated GPU : 10 GPUs (NVIDIA GeForce RTX 3090)
- Node : Intel(R) Xeon(R) Gold 6242R CPU (3.10GHz, 40 Cores) * 2 / 250 GB RAM
Server 17
- Integrated GPU : 12 GPUs (NVIDIA GeForce RTX 3080 Ti)
- Node : Intel(R) Xeon(R) Gold 6248R CPU (3.00GHz, 48 Cores) * 2 / 250 GB RAM
2nd generation GPU cluster (1 Nods, 35 GPUs, 471 Tflops)
Master
- Integrated GPU : 3 GPUs (NVIDIA GeForce RTX 2080 Ti)
- Node : Intel(R) Core(TM) i9-10900X CPU (3.70GHz, 20 Cores) / 100 GB RAM
Left
- 32 GPUs (NVIDIA GeForce RTX 2080 Ti)
Non-GPU clusters
Data Storage
nas1 : 490 TB
nas2: 490 TB
nasall : 80 TB