Global Green Computing Network for AI Infrastructure

Global Green Computing Network for AI Infrastructure

Saudi Arabia 100,000-GPU National Green Intelligent Computing Center

Saudi Arabia green intelligent computing center

Core Financial Outlook

The initial investment for hardware infrastructure and data center construction totals US$5 billion. The project is expected to generate US$1.6 billion in comprehensive revenue in the first year, and achieve US$2.11 billion in net profit by the third year of steady-state operations, delivering sustainable and long-term investment returns.

Project Scale

A 100,000-GPU high-performance computing cluster is planned for the initial phase, with a three-phase expansion roadmap to continuously scale computing capacity and enhance global AI computing supply.

Core Infrastructure Advantages

Powered by low-cost, sustainable clean energy in the Middle East, the project ensures reliable power supply through a fully liquid-cooled data center architecture and achieves industry-leading ultra-low PUE energy efficiency standards.

Key Application Scenarios

Designed to support trillion-parameter foundation model pre-training, multimodal AIGC content generation, cutting-edge supercomputing research, and secure private AI deployment for governments and enterprises.

Global Computing Nodes:
Medium- to Long-Term Development Plan

North America New York Computing Power Hub

North America New York Computing Power Hub

A compliant operational gateway in the United States, serving as the centralized hub for North American AI enterprises and cloud providers. It enables unified cross-border computing power orchestration, scheduling, and distribution.

European Compliant Computing Node

European Compliant Computing Node

Designed to comply with EU GDPR data governance requirements, providing localized data-isolated computing services for multinational enterprises and research institutions across Europe.

Asia Regional Computing Sub-Center

Asia Regional Computing Sub-Center

Serving AI R&D companies, enterprises, and government clients across the Asia-Pacific region, establishing a regional computing power scheduling network while meeting local data sovereignty and regulatory compliance requirements.

Four Foundational Core Technology
Support Systems

High-Speed Interconnect Architecture for Ten-Thousand-Scale GPU Clusters

High-Speed Interconnect Architecture for Ten-Thousand-Scale GPU Clusters

Overcoming network latency and communication bottlenecks in large-scale distributed training, enabling ultra-high throughput and stable cluster operations.

Distributed Intelligent Computing Power Scheduling Algorithms

Distributed Intelligent Computing Power Scheduling Algorithms

Dynamically allocating global computing workloads, balancing node utilization, and significantly reducing idle capacity losses.

Green Energy Intelligent Thermal Management System

Green Energy Intelligent Thermal Management System

Leveraging liquid cooling and dynamic computing-power-to-energy matching technologies to continuously reduce data center energy consumption, supporting carbon asset development and the realization of green premiums.

Digital Full-Lifecycle Audit System for Computing Power Assets

Digital Full-Lifecycle Audit System for Computing Power Assets

Providing real-time data recording and verification throughout the hardware lifecycle, enabling fully traceable and auditable asset operations and revenue management.