ASUS Unveils Next-Gen AI Infrastructure Solutions at SC24, Including NVIDIA-Powered AI POD and Taiwan's Largest Supercomputer

ASUS Unveils Cutting-Edge AI Infrastructure at SC24

ASUS has made a significant splash at SC24 with the unveiling of its next-generation infrastructure solutions, showcasing an extensive lineup of AI-focused servers and advanced cooling technologies 1

1

2

2

. The company's presentation highlights its commitment to propelling the future of AI through innovative hardware solutions and strategic partnerships.

AI POD: A Powerhouse for Large Language Models

At the forefront of ASUS's offerings is the AI POD, a complete rack solution equipped with the NVIDIA GB200 NVL72 platform. This system integrates GPUs, CPUs, and switches in a seamless, high-speed configuration, designed to enhance the training of trillion-parameter Large Language Models (LLMs) and enable real-time inference 1

1

2

2

. The AI POD features the NVIDIA GB200 Grace Blackwell Superchip and fifth-generation NVIDIA NVLink technology, offering both liquid-to-air and liquid-to-liquid cooling options to maximize AI computing performance.

Comprehensive Server Lineup for Diverse AI Workloads

ASUS's showcase extends beyond the AI POD, presenting a range of servers tailored for various AI and high-performance computing (HPC) scenarios:

1. ESC8000A-E13P: An HPC server compliant with NVIDIA MGX modular architecture, supporting up to eight NVIDIA H200 NVL dual-slot cards 1

   1

   .
2. ESC N8-E11V: Featuring the NVIDIA HGX H200 platform for generative AI applications 1

   1

   .
3. ESC N8A-E13: Incorporating the latest NVIDIA Blackwell platform 1

   1

   .
4. ESC A8A-E12U: Powered by AMD EPYC 9005 processors and Instinct MI325X accelerators 3

   3

   .
5. ESC I8-E11: Supporting Intel Gaudi 3 accelerators for AI workloads 3

   3

   .

Advanced Cooling Solutions for Optimal Performance

Recognizing the critical role of heat management in data centers, ASUS has partnered with global cooling solution providers to offer adaptive cooling technologies. These solutions optimize efficiency from single-cabinet setups to entire data centers, achieving up to 95% heat dissipation 1

1

2

2

. The company's approach includes comprehensive testing and verification services to ensure reliable performance under full load.

Taiwan's Largest Supercomputing Center

In collaboration with Ubilink, ASUS has completed the construction of Taiwan's largest supercomputing center. This facility, built in just three months, boasts an impressive performance capacity of 45.82 PFLOPS and has secured positions on both the global TOP500 and Green500 lists 4

4

. Notably, the center offers customers the option to choose packages that utilize only renewable energy, underlining ASUS's commitment to sustainable computing solutions 1

1

2

2

4

4

.

Unleashing Data Potential with Storage Innovations

ASUS is also addressing the growing storage demands of AI and HPC with software-defined storage (SDS) solutions. The RS501A-E12-RS12U, for instance, is designed to excel in AI computing environments, offering the speed, efficiency, and scalability needed for demanding AI workloads 3

3

.

Industry Impact and Future Outlook

ASUS's comprehensive approach to AI infrastructure, from high-performance servers to innovative cooling and storage solutions, positions the company as a key player in shaping the future of AI computing. By offering scalable and adaptable solutions, ASUS aims to empower organizations of all sizes to leverage AI technologies effectively, potentially accelerating advancements across various industries 1

1

2

2

3

3

4

4

.