Centralized cloud servers offer significant advantages in terms of space reduction, minimized equipment wiring, and ease of installation and construction, particularly in new projects with highly integrated device space and performance, as well as in smart upgrade projects for existing rolling stock on older lines.
Edge cloud servers utilize high-performance processing modules to form resource pools for CPU and AI computing power. They support flexible resource expansion and node upgrades as needed, and are capable of hosting in-car systems such as passenger information systems, video analytics systems, health management systems, along with operational control, intelligent analysis, and other functions for third-party systems. The on-board edge cloud platform consists of Centralized Cloud Servers and compartment switches.
| Structure | 3U19-inch | ||
|---|---|---|---|
| Power module | DC 110V (77V~137.5V),power:150W | ||
| Data processing node | Optional | ||
| Specification | Standard | Premium | XinChuang |
| CPU specification | 8-core Arm® Cortex®-A78AE v8.2 64位 CPU 2MB L2 + 4MB L3 | 8-core 64-bit big.LITTLE architecture 4*Cortex-A76 + 4*Cortex-A56 | |
| GPU specification | 1024-core NVIDIA Ampere architecture GPU with 32 Tensor Cores | 1792-core NVIDIA Ampere architecture GPU with 56 Tensor Cores | ARM Mali-G610 MC4 embedded high-performance 2D image acceleration module |
| Memory specification | 128-bit LPDDR5 | 256-bit LPDDR5 | 64-bit LPDDR5 |
| Computing performanc | 100 TOPS | 200 TOPS | NPU 6TOPS |
| Memory capacity | 16GB | 32GB | 8GB |
| eMMC capacity | Support external NVMe | 64GB eMMC 5.1 | 32GB eMMC 5.1 |
| Storage capacity | 2T | ||
| Operating system | Ubuntu 22.04 LTS | UbuntuKylin 22.04 | |
| Network | 100M/1000M Adaptive | ||
| Power consumption | 25W | 40W | 5.4W |
© 2026 Suzhou Huaqi Intelligent Technology Co., Ltd.