Wiwynn is showcasing its latest AI computing solutions at GTC 2024. These include new AI server racks powered by the NVIDIA GB200 NVL72 system. Wiwynn also unveiled the UMS100, an advanced rack-level liquid-cooling management system designed to meet the increasing demands for high computing power and efficient cooling mechanisms in the burgeoning era of Generative AI (GenAI).
Wiwynn is showcasing its latest AI computing solutions at GTC 2024. These include new AI server racks powered by the NVIDIA GB200 NVL72 system. Wiwynn also unveiled the UMS100, an advanced rack-level liquid-cooling management system designed to meet the increasing demands for high computing power and efficient cooling mechanisms in the burgeoning era of Generative AI (GenAI).
As a key partner of NVIDIA, Wiwynn has achieved readiness for the NVIDIA GB200 NVL72 system. This system features the cutting-edge NVIDIA GB200 Grace Blackwell Superchip, integrated with the latest networking technologies such as the NVIDIA Quantum-X800 InfiniBand and NVIDIA Spectrum-X800 Ethernet. Advancements in technology are set to significantly improve the abilities of generative AI, allowing for models with trillion-parameter scales.
Wiwynn’s strategy involves crafting optimized, rack-level, liquid-cooled AI solutions that leverage their strengths in high-speed data transmission, energy efficiency, system integration, and advanced cooling technologies. Its goal is to satisfy emerging customer requirements for performance, scalability, and diversity within data center ecosystems.
In response to the increasing importance of liquid cooling in AI and HPC applications, Wiwynn will also showcase the UMS100, its universal liquid-cooling management system. This innovative system offers a suite of functionalities, including real-time monitoring, cooling energy optimization, rapid leak detection, and containment capabilities. It is also designed to integrate smoothly with existing data center management systems via the Redfish interface. It supports industry-standard protocols for compatibility with various cooling distribution units (CDUs) and sidecars.
Wiwynn UMS100 Universal Cooling Management System
| Feature | Description |
| Type | Liquid Cooling Rack Manager |
| Controller | OCP DC-SCM-based Redundant Controller (1+1 Redundancy) |
| Leak Detection | Rapid Leakage Detection and Containment |
| API & Interface | Redfish API & GUI |
| Monitoring | Real-time Monitoring & Cooling Optimization |
| Power Input | Bus Bar & CRPS Power Input |
| Networking | Modbus RTU/TCP |
| Additional Features | Direct Connection to Digital/Analog Sensors or PLC |
Wiwynn GB200 NVL72 Rack Solution
| Feature | Description |
| Rack Configuration | 72x GPU Single Open Rack v3 Solution |
| GPU Allocation |
|
| Networking |
|
| Rack Specifications | 440U ORv3 Rack with 1400A Bus Bar |
| Power Supply | 4x 66KW Power Shelf |
| Cooling Support | Support Liquid-to-Air / Liquid-to-Liquid CDU |
Wiwynn envisions its GB200-enabled AI racks as a pivotal solution in delivering some serious computing power while providing the adaptability needed to accommodate the diverse and evolving data center scalability, flexibility, and efficient management requirements. In partnership with NVIDIA, Wiwynn is dedicated to advancing optimized and sustainable data center solutions for the GenAI era.
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
