The new V3 server family supports 6th generation Intel
To support a wide range of AI application scenarios, it supports a variety of platforms such as CPU, GPU, and FPGA, and provides full-stack openness across components, firmware, nodes, and racks. It supports scale-up expansion as well as scale-out expansion to meet the expansion requirements of AI models and supports a wide range of AI scenarios.
Optimized data storage that supports massive data processing tasks required for large-scale language models is also an advantage, providing a capacity of up to 128TB on a single drive, a single server capacity of up to 6PB, and 2.7 times increased storage in a single rack.
In addition, the Catus V3 server supports liquid cooling and maximizes efficiency through two-phase cold plates, negative pressure technology, special coating on the motherboard PCB, and an automatic power-off function in the event of water leakage.
In addition, an open firmware management platform and a completely modular design were applied, and a comprehensive error diagnosis system was established through AI-based built-in intelligence.
Source: www.itworld.co.kr
