XTT657V2-L AI server
  • XTT657V2-L AI server
  • XTT657V2-L AI server
  • XTT657V2-L AI server
  • XTT657V2-L AI server

XTT657V2-L AI server

Brand CHINA

Product origin CHINA

Delivery time Within 15 days

Supply capacity 9999

The XTT657V2-L server is a flagship AI liquid-cooled server developed based on Phytium’s latest S5000C-64C server processing chip and fully domestically produced hardware components. It is designed for applications such as large model training, inference, HPC, and rendering. Featuring two Phytium S5000C 64-core processors built on the latest FTC862 core architecture, the server delivers at least twice the computational performance compared to the previous S2500 generation. It supports 8 to 10 full-height, full-length GPU cards, making it well-suited for large-scale AI operations in computing centers, industry markets, and internet sectors.

XTT657V2-L AI server

Product Features

Domestic Platform with Independent Control

  • Fully Domestic Components: Our server uses entirely domestic Feiteng ARM processors and related components, ensuring comprehensive independent control and compliance with national information security requirements.

  • Domestic GPU Support: Supports fully domestic GPU cards and firmware, compatible with multiple BMC solutions for Xinchuang and industry markets, and ARMv8.2 architecture, significantly enhancing performance.


Green Energy Saving and Efficient Consumption

  • Liquid Cooling Technology: Equipped with the Feiteng S5000C-64 (2.5G) liquid cooling processor, providing a significant performance boost compared to air-cooled processors, while improving system energy efficiency by over 20%.

  • Intelligent Fan Control: Adjusts fan speed based on the temperature of key components, saving energy while reducing noise and enhancing overall system efficiency.

  • Efficient Cooling: Supports 40°C liquid supply, with low flow and low resistance design for efficient cooling, reducing data center energy consumption and carbon emissions, meeting green data center standards.


Extreme AI Performance

  • Powerful Computing: Supports 8-10 full-height full-length GPU cards (domestic PCIe GPU training cards), providing exceptional AI computing power to meet large model training and inference requirements.

  • High-Efficiency Processors: Equipped with 2 Feiteng S5000C 64-core processors and 16 DDR5 memory slots, with a maximum memory frequency of 4000MHz, delivering outstanding computing performance and responsiveness.


Flexible Configuration and Customization

  • Modular Design: Hardware modular decoupling design allows flexible configuration of computing, storage, and network, meeting various business needs and supporting personalized customization.

  • Diverse Storage Options: Supports flexible configuration of 3.5” or 2.5” hard disk modules, with a maximum of 24 U.2 NVMe slots to accommodate different data storage requirements.

  • Rich Expansion: Supports up to 15 standard PCIe Gen5 expansion slots and OCP NIC 3.0 network expansion, providing multiple network configuration options to meet future upgrade needs.


Stable and Reliable with Intelligent Management

  • Redundant Design: Key system components feature redundant and hot-swappable design, supporting tool-free assembly and disassembly, enhancing fault maintenance efficiency and system availability.

  • Intelligent Management: Integrated smart management chip supports IPMI2.0, Redfish, SNMP, and other management protocols, with features including remote KVM, virtual media, component status monitoring, and anomaly alarms for comprehensive remote system-level intelligent management.

  • GPU AI server

Technical Specifications

  • CPU: 2 Feiteng S5000C-64C CPUs, 64 FTC862 cores, ARMv8.2 compatible, 2.1GHz frequency, TDP 330W, or high-performance version 2.5GHz, TDP 370W

  • GPU Support: Supports 8/10 full-height full-length dual-width (DW-FHFL) or single-width (SW-FHFL) GPU cards

  • Memory: 16 DDR5 DIMM slots, supports up to 4000MHz, maximum 1.0TB

  • Storage Controller: Onboard 6Gb/s SATA controller, supports 12Gb/s SAS HBA and RAID controller, supports RAID0/1/10 and RAID0/1/5/6/50/60

  • PCIe Expansion: Standard 12 PCIe Gen5 slots, 1 PCIe Gen4 slot, OCP 3.0 X8 slot

  • Local Storage: Supports up to 24 U.2 NVMe SSDs, 8/16/24 2.5” hard disks, 8/12 3.5” hard disks, and 2 M.2 SSDs

  • Network: 1GbE RJ45 BMC IPMI management port, optional 4×1GbE/2×10GbE/2×25GbE OCP3.0 network cards or PCIe standard network cards, and 100/200G NIC/IB PCIe cards

  • I/O Ports: Front: 2 USB 3.0, 1 VGA port; Rear: 1 serial port, 2 USB 3.0 ports, 1 VGA port, 1 RJ45 management port

  • Fans: 8 hot-swappable high-speed fans, N+1 redundancy

  • Power Supply: Supports 2-4 CRPS power modules, 1+1/3+1/2+2 redundancy, with power options of 1600W/2000W/2400W/2700W

  • Management Features: Supports IPMI, SOL, KVM Over IP, virtual media, power consumption control, and leak detection

  • Security Features: Liquid cooling leak monitoring component, lockable chassis cover, chassis intrusion detection, key-lock front panel, TCM

  • Dimensions: 4U rackmount, 3.5” hard disk chassis dimensions: 448x175x822mm, 2.5” hard disk chassis dimensions: 448x175x772mm

  • Weight: Net weight approximately 55KG, including accessories and packaging approximately 80KG

  • Temperature: Standard operating temperature 5ºC - 35ºC, extended operating temperature 5ºC - 40ºC, transport and storage temperature -40ºC - 65ºC

  • Humidity: Operating humidity 30% - 80%, storage humidity 5% - 93%

  • Operating System Support: Kylin OS, OpenEuler OS, UOS, and other open-source OS (please consult sales for specific support)


Why Choose Us

  • Comprehensive Domestic Control: Our servers feature fully domestic Feiteng ARM processors and components, ensuring system control and compliance with national security standards.

  • Leading Green Energy Saving Technology: Advanced liquid cooling technology and intelligent fan control optimize performance while enhancing energy efficiency, reducing energy consumption and noise, contributing to a greener data center.

  • Exceptional AI Performance: Equipped with high-performance Feiteng S5000C processors and GPU support, our servers excel in handling large-scale AI training and inference tasks.

  • Flexible Configuration Options: Modular design and a wide range of configuration choices allow for tailored solutions to meet various business needs and future upgrades.

  • Stable and Reliable Management: With redundant designs and intelligent management features, our servers ensure high stability and efficiency, reducing downtime and offering comprehensive technical support.


Get the latest price? We'll respond as soon as possible(within 12 hours)

Privacy policy

close left right