Machine equipped with Geforce RTX4090 x4 | Information dissemination media for research and development TEGAKARI

A customer who develops technology related to augmented reality consulted us about a machine equipped with four Geforce RTX4090s for Deep Learning learning.
The CPU specs are not very necessary, and we would like to introduce it with a budget of about 250 million yen.

Based on the information you provided, we considered the following configuration.

【Main Specifications】

CPU	Xeon Silver 4314 (2.40GHz 16 cores)
memory	128GB
storage	1TB SSD S-ATA x2
video	NVIDIA Geforce RTX4090 x4
network	on board (1GbE x1 10GBase-T x1)
Housing + power supply	Cube case + 1600W x2
OS	Ubuntu 22.04

When using four RTX4090s, depending on the CPU selected, the maximum power consumption in the specifications is expected to exceed 4W, so it is necessary to secure a power supply.In this example, two power supply units are installed and two power supply outputs are used.

This configuration can be used with a 100V power supply, but in that case there will be no room for the power supply unit capacity, so we recommend using a 200V power supply.Also, when changing the CPU to a model with higher power consumption, it is necessary to use 200V power supply depending on the power consumption after the change.

Note that the two power supplies in this example are not intended for redundancy, but to ensure the amount of power supplied.If one power supply fails, the system will not function, so please avoid using it in critical applications.

The configuration of this case study is based on the conditions given by the customer.
Please feel free to contact us even if you are considering different conditions from what is posted.

■ 注意点
When the RTX4090 x4 is fully operated, the fan noise becomes quite loud, and depending on the installation environment, it may not be suitable for use under high load conditions.
We also accept proposals for introducing machines that emphasize quietness, so please feel free to contact us.