#Singapore #NVIDIA Blackwell #AI Server #SuperX

SuperX Unveils the All-New SuperX XN9160-B200 AI Server

SuperX AI Technology Limited has officially launched its latest flagship product, the SuperX XN9160-B200 AI Server. This cutting-edge server, equipped with NVIDIA's Blackwell B200 GPU architecture, promises to revolutionize the landscape of high-performance computing, primarily for AI training, machine learning, and high-performance computing workloads. With an impressive performance increase of up to 30 times compared to the previous H100 series, this AI server is engineered not just for functionality but also for incredible speed and efficiency.

Features and Capabilities

The XN9160-B200 is purpose-built to handle large-scale distributed AI training and inference tasks. Significantly optimized for GPU-supported processes, it excels in applications involving reinforcement learning, multimodal model training, and high-performance computing tasks like climate modeling and drug discovery. This server aims to meet the growing needs of various sectors, including finance, health, and academia, requiring formidable compute resources.

In an age where AI innovation drives competitive advantage, the XN9160-B200's performance rivals traditional supercomputers while fitting into a compact form factor. Equipped with eight NVIDIA Blackwell B200 GPUs, 6th Gen Intel® Xeon® processors, and cutting-edge memory technology, this server makes an impressive case for enterprises looking to boost their AI capabilities.

Performance Specifications

• - GPU Architecture: 8× NVIDIA Blackwell B200, utilizing fifth-generation NVLink technology to provide ultra-high inter-GPU bandwidth of up to 1.8 TB/s.
• - Memory: 1440 GB of high-bandwidth memory (HBM3E) allows for unprecedented processing power, achieving an astonishing throughput of 58 tokens per second per card on the GPT-MoE 1.8T model. In comparison, this represents up to a 15x speed increase over the previous generation H100 platform.
• - Processors: Dual Intel® Xeon® 6710E Processor (64 cores) running at 2.40 GHz enhances data processing efficiency, crucial for complex parallel computing tasks.
• - Storage Options: Features SSD and NVMe storage to facilitate rapid data access, crucial for high-load virtualization environments.
• - Power Redundancy: An advanced multi-path power redundancy solution guarantees continuous operation and reduces risks associated with power failures, ensuring the server remains functional during critical tasks.

Target Market and Use Cases

The XN9160-B200 is particularly designed for large tech companies, academic institutions, financial services, and healthcare providers. It aims to meet complex computing needs such as:

• - Tech Companies: For large-scale AI modeling and generative applications.
• - 研究機関: For intricate scientific simulations and modeling.
• - Finance and Insurance: For risk modeling and analytics to drive timely decision-making processes.
• - Pharmaceutical and Healthcare: To support drug discovery and screening tasks more efficiently.
• - Government and Environmental Agencies: For climate modeling and predictive analyses to inform policy and action.

Customer Assurance and Support

To ensure the reliability of their systems, SuperX has established thorough testing protocols. Each server undergoes over 48 hours of rigorous stress testing accompanied by multiple quality control measures. A three-year warranty is also provided along with technical support to assist enterprises in effectively navigating the expanding AI landscape.

Conclusion

The launch of the SuperX XN9160-B200 marks a pivotal advancement in AI computing solutions, catering to extensive performance demands across diverse sectors. With its unique features and commitment to innovation, SuperX is set to play a significant role in driving AI initiatives that will shape the future of technology.

For more information or inquiries regarding purchase and specifications, interested parties can reach out to the SuperX sales team.
Visit SuperX Website for complete details on product offerings and services.