■ This is an article posted on June 2020, 8, so the content of the information may be out of date.
A customer inquired about a Linux machine capable of parallel computation with a GPU for real-time analysis of satellite data.
We proposed a rack mount configuration that supports up to 100 GPUs and a CUDA Toolkit installed within the budget of 3 million yen.
Regarding CUDA, we assume only basic installation, but if you wish to make settings other than installation, please contact us.
【Main Specifications】
| CPU | Xeon Processor W-2295 (3.00GHz 18 core) |
| memory | 128GB (32GB x 4) |
| storage | 1TB M.2 SSD |
| Video | NVIDIA Geforce RTX2080Ti 11GB x 2 |
| network | on board (Gigabit LAN x 2) |
| Housing + power supply | 4U rack mount case (width 483 x height 177 x depth 505mm) + 1500W |
| OS | 8 CentOS |
| Other | CUDA Toolkit installation "Safety +" 3-year warranty |
| ■ Click here for details and inquiries about this PC case Example of real-time satellite data analysis machine |
