GPU-Cluster Requirements.md 1.74 KB
Newer Older
1
# Current Situation
2

3
## Hardware Stack
4 5 6 7 8 9 10 11
Currently planned node-configuration:
- 2x Intel Xeon 6230 CPU
- 192 GB RAM
- 8 GPUs

We are currently looking into two possible GPU configurations:
- 80% SP / 20% DP GPUs (72x RTX 2080Ti / 16x V100)
- 60% SP / 40% DP GPUs (48x RTX 2080Ti / 24x V100)
12

13
## Software Stack
14 15 16 17
Currently planned:
- Scientific Linux 7 (to ensure compatibility with the existing CPU-Cluster)
- In the long future we are looking to migrate to CentOS since the support for SL7 got discontinued.

18
# Additional Information
19 20 21
Difference between SingePrecision and DoublePrecision:  
[simple explanation](https://www.thecrazyprogrammer.com/2018/04/single-precision-vs-double-precision.html)  
Desktop GPUs like the RTX2080 are SingePrecision cards, where professional ones like the V100 are DoublePrecision ones.
22

23
# Requirements
24

25 26 27
Section | Project Description | Hardware Requirements | Software Requirements | Comments
--- | --- | --- | --- | ---
copy | paste | this | line | !
28
2.4 | Deep learning for fast magnitude estimation of earthquakes | Single precision GPU with > 10 GB GPU memory, at least 200 GB main memory, 8 CPU cores | up to date nvidia driver, the rest works fine with conda (cuda, tensorflow, pytorch) | I fear that 2 CPUs for 8 GPUs might be to less. The machine I'm currently running on has two Xeon Gold 5122 and four RTX 2080 Ti and is CPU bound.
29 30 31 32 33

Section | Project Description | Hardware Requirements | Software Requirements | Comments
--- | --- | --- | --- | ---
copy | paste | this | line | !
2.8 | Deep learning with solar images | Single precision GPU with > 20 GB GPU memory (RTX Titan not 2080Ti) | Nvidia driver. Module loads for standard python software, Virtualenv. Horovod required for multi node computations | I suggest 4 GPUs per node not 8.