Published on December 25, 2023, 4:32 pm
Dell Expands its AI Infrastructure with AMD-Powered Server
Dell has added a new server to its high-performance computing portfolio for AI workloads. In addition to the Nvidia-powered version, customers will soon have the option to purchase a Dell PowerEdge XE9680 server equipped with eight AMD Instinct MI300X accelerators. This new server allows businesses to train and run large language models (LLMs) in-house, providing 1.5GB of high-bandwidth memory (HBM3) and over 21 petaFLOPS of performance.
One of the standout features of this server is its ability to scale systems using the global memory interconnect (xGMI) standard. Additionally, customers can connect AMD’s GPUs over an Ethernet-based AI fabric with a Dell PowerSwitch Z9664F-ON. This launch follows Dell’s introduction of a unit fitted with Nvidia H100 GPUs earlier this year.
To complement this offering, Dell has also introduced a new standard called Dell Validated Design for Generative AI with AMD. This framework provides organizations with guidance on how to build their own hardware and networking architecture for running LLMs. It helps with the integration and physical installation, optimizing performance for generative AI platforms using various frameworks such as PyTorch, TensorFlow, and OpenAI Triton – all with native support on the PowerEdge XE9680 fitted with the AMD accelerators.
What sets Dell apart from other players in the industry is its commitment to standards-based networking. By being a member of the Ultra Ethernet Consortium (UEC), Dell promotes an open approach where switches from different vendors can interoperate seamlessly on the same system. This aligns well with AMD’s advocacy for an open Ethernet for AI.
The new hardware and services that make up Dell’s latest AI initiative are expected to be available in the first half of next year.
In conclusion, Dell’s collaboration with AMD brings a powerful and scalable AI infrastructure solution to businesses. With the capability to run LLMs in-house, organizations can leverage the performance and support of AMD Instinct accelerators along with Dell’s open approach to networking and AI fabric. This advancement marks another significant step in the world of generative AI, enabling businesses to explore new possibilities and drive innovation.