Aetina Unveils Revolutionary SuperEdge NVIDIA MGX Server By Elets News Network - 04 June 2024

nvidia

Aetina, a subsidiary of the Innodisk Group and a leader in edge AI solutions, proudly announces the launch of its innovative SuperEdge NVIDIA MGX short-depth server, model AEX-2UA1, at Computex 2024. This unveiling marks a significant expansion of Aetina’s product portfolio, transitioning from specialized edge devices to comprehensive AI server solutions.

The AEX-2UA1 is among the first x86-CPU short-depth NVIDIA MGX servers introduced to the market, representing a pivotal milestone in the Innodisk Group’s AI roadmap. This compact yet powerful server is designed to deliver maximum performance for edge computing, enabling enterprises to manage sensitive data securely with private large language models (LLMs) and other enterprise AI applications. Sectors such as finance, healthcare, and 5G telecommunications, where data privacy and space optimisation are critical, stand to benefit significantly from this cutting-edge technology.

As generative AI transforms industries, enterprises handling sensitive data face increased operational and compliance risks with cloud-based LLM services. The AEX-2UA1 addresses these concerns by offering an on-premises solution for private LLM AI training and inferencing. This ensures that models trained on sensitive data remain secure and compliant with strict industry regulations.

Powered by NVIDIA and Intel technologies, the AEX-2UA1 supports two double-deck GPUs with NVIDIA NVLink bridges and a single Intel® Xeon 6 processor, enhancing LLM training and inferencing performance. The server’s advanced network solutions, including NVIDIA BlueField-3 DPUs, SuperNICs, and ConnectX-7 NICs, facilitate efficient server-to-server and server-to-storage communication, optimising parallel processing for enterprise edge AI systems.

The AEX-2UA1’s x86 short-depth design, front access, and compact form factor make it ideal for space-constrained edge environments, supporting both rack and non-rack deployments. This versatility allows for deploying diverse AI-driven applications wherever needed at the edge.

Also Read | Lenovo and NVIDIA strengthen AI capabilities with innovative hybrid AI solutions

Compliant with the NVIDIA MGX architecture, the AEX-2UA1 offers a modular server architecture that provides maximum flexibility and forward compatibility. This includes support for future NVIDIA Blackwell GPUs, ensuring greater scalability for enterprises.

James Su, Director of Software R&D at Aetina, stated, “The AEX-2UA1 not only marks Aetina’s advancements in product portfolio extension but also serves as an ideal entry system for enterprises in need of LLM deployment. It offers low barriers to entry compared to standard GPU servers, streamlining the development and deployment of AI applications.”

The launch of the AEX-2UA1 reinforces Aetina’s position as a driving force in AI innovation within the Innodisk Group, leveraging cross-industry integration and insights to transform cutting-edge innovations into practical endpoint solutions. This move facilitates the widespread adoption of edge AI across diverse markets.

Mark your calendar for the Elets India AI Summit, taking place on July 11-12, 2024. This summit aims to foster a strong AI ecosystem across various sectors, empowering both public and private organizations to harness tech-driven innovations. Attendees can expect valuable insights and excellent networking opportunities.”

Related News