AMD, Intel & Nvidia Reveal Next-Gen Server Hardware at Supermicro Innovate

Intel

The chatter around AI and Deep Learning is no longer a bunch of industry buzzwords, these AI products are in our homes, our phones, our cars, and many other aspects of our daily lives. From helping with your homework to generating images using Dali or Chat GPT, to creating code for your new software.

Most of it happens without you knowing, or when you request something from your browser assistant, but to get those responses dealt with swiftly, it takes some impressive server hardware, and both Intel and Supermicro have been working hardware to push the limits of what is possible.

With the latest generation of Intel Xeon Scalable CPUs and Habana Gaudi deep learning processors, AI training has now become faster and more efficient than ever. The speed and power of such hardware allow engineers and researchers to push the limits of their experimentation and deployment of AI models, faster, and more reliable than ever.

When it comes to dealing with datasets that can be near limitless in storage capacity, and with trillions of AI model parameters, having a powerful processor that can deal with these complex data sets is vital, and that’s where the dual-socket Xeon solutions come into their own.

With dual socket Xeon solutions from Intel and Supermicro, these new systems can house up to 8 Gaudi Deep Learning Accelerator Mezzanine Cards per 4U rackmount server tray. That allows for performance levels, and scalability like never before, to push the limits of accelerated compute performance for next-generation AI and Deep Learning systems.

Supermicro, in collaboration with Intel, not only provides the fastest AI training solutions, but they are proving to be the most reliable too. When dealing with critical and real-time systems, and huge and undoubtedly sensitive datasets for AI training models, reliability can be crucial. Especially when these systems can be used for vehicle safety systems, such as real-time navigation and accident avoidance systems in self-driving cars.