ASUS Adds AMD Instinct MI350 GPU Support to ESC A8A-E12U Server

ASUS upgrades its AI server, ESC A8A-E12U, with support for AMD Instinct MI350 GPUs, boosting Artificial Intelligence and HPC performance for enterprise and research.

ASUS has announced that its flagship high-density server, the ESC A8A-E12U, has been upgraded to support the new AMD Instinct MI350 series GPUs. This development allows enterprises, research bodies, and cloud service providers to accelerate Artificial Intelligence and high-performance computing workloads with advanced capabilities, while avoiding the need for a complete infrastructure overhaul.

The AMD Instinct MI350 series GPUs, built on the fourth-generation AMD CDNA architecture, bring significant hardware enhancements. These graphics processors feature 288 GB of HBM3E memory and deliver up to 8 TB/s of bandwidth. Such specifications help expedite large-scale Artificial Intelligence model training and the handling of complex simulations with improved energy efficiency. The MI350 lineup also introduces broader compatibility with low-precision compute formats, specifically FP4 and FP6, which enhance the speed and performance of generative Artificial Intelligence, inference, and machine learning computations.

Crucially, the Instinct MI350 GPUs are engineered for backward compatibility as drop-in replacements for systems that currently utilize AMD Instinct MI300 series processors, including the MI325X. This approach offers customers a cost-effective and seamless upgrade path, minimizing resource requirements and simplifying both scaling and overall workload management. By enabling straightforward hardware upgrades and more efficient system utilization, the ESC A8A-E12U with MI350 support presents a compelling solution for organizations seeking to deploy or expand large-scale Artificial Intelligence infrastructure.

59

Impact Score

Europe and US discuss biometric data-sharing framework

European Union and US officials are negotiating a border security arrangement that could enable continuous biometric data exchanges on EU citizens. The UK says the US has also requested access to fingerprint records as part of Visa Waiver Program discussions.

Apple plans Intel 18A-P for M7 and 14A for A21

Apple is expected to use Intel’s 18A-P process for M7 chips in MacBook models and Intel’s 14A process for A21 chips in iPhones. The shift points to a broader supplier strategy as Apple moves beyond TSMC for parts of its future silicon roadmap.

Google and other chatbots surface real phone numbers

Generative Artificial Intelligence chatbots are surfacing real phone numbers and other personal details, sometimes by pulling from obscure public sources and sometimes by inventing plausible but wrong contact information. Privacy experts say users have few reliable ways to find out whether their data is in model training sets or to force its removal.

U.S. and China revisit Artificial Intelligence emergency talks

Washington and Beijing are exploring renewed talks on an emergency communication channel for Artificial Intelligence as fears grow over the capabilities of Anthropic’s Mythos model. The shift reflects rising concern in both capitals that competitive pressure is outpacing safeguards.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.