Vultr enhances global AI cloud inference with AMD Instinct™ MI300X

Author without image icon
casper.brands@jaarbeurs.nl
27 September 2024
3 min

New Cloud GPU offering ensures business agility, performance and cost efficiency for global scalability of AI-native applications with AMD Instinct™ MI300X

Amsterdam, 25 September 2024 - Vultr, the world's largest private cloud computing platform, today announced that the new AMD Instinct™ MI300X accelerator and ROCm™ open software will become available within Vultr's composable cloud solution.

The collaboration between Vultr's composable cloud infrastructure and AMD's next-generation silicon architecture opens up new frontiers for GPU-accelerated workloads from the data centre to the edge.

"Innovation thrives in an open ecosystem," said J.J. Kardwell, CEO of Vultr. "The future of enterprise AI workloads lies in open environments that enable flexibility, scalability and security. AMD accelerators offer our customers an unmatched price-performance ratio. The balance between high memory and low power consumption supports sustainability efforts and empowers them to efficiently drive innovation and growth through AI."

Building a composable cloud

With AMD ROCm open software and Vultr's cloud platform, enterprises have access to an environment for AI development and deployment. The open nature of AMD's architecture and Vultr infrastructure gives enterprises access to thousands of open source, pre-trained models and frameworks with a drop-in code experience. This creates an optimised environment for AI development, allowing projects to move forward quickly.

"We are proud of our close collaboration with Vultr as the cloud platform is designed to manage high-performance AI training and inferencing tasks and provide improved overall efficiency," said Negin Oliver, corporate vice president of business development, Data Centre GPU Business Unit, AMD. "With the use of AMD Instinct MI300X accelerators and ROCm open software for these latest deployments, Vultr's customers will benefit from a truly optimised system that can manage a wide range of AI-intensive workloads."

Designed for next-generation workloads, the AMD architecture on the Vultr infrastructure enables true cloud-native orchestration of all AI resources. AMD Instinct™ accelerators and ROCm software management tools integrate seamlessly with the Vultr Kubernetes Engine for Cloud GPU to create GPU-accelerated Kubernetes clusters that can power the most resource-intensive workloads anywhere in the world. These platform capabilities give developers and innovators the tools to build advanced AI and machine learning solutions for the most complex business challenges.

The partnership has multiple benefits:

Improved price-performance ratio: Vultr's high-performance cloud compute, accelerated by AMD GPUs, provides exceptional processing power for demanding workloads while maintaining cost efficiency.
Scalable compute power and optimised workload management: Vultr's scalable cloud infrastructure, combined with AMD's advanced processing capabilities, enables enterprises to effortlessly scale their compute power as demand increases.
Accelerated discovery and innovation in R&D: Vultr's cloud infrastructure provides the necessary compute power and scalability for developers to leverage AMD Instinct GPUs, AMD ROCm™ open software and its extensive partner ecosystem, enabling them to solve complex problems and enable faster discovery cycles and innovation.
Optimised for AI inference: Vultr's platform is optimised for AI inference, with AMD MI300X GPUs providing faster, scalable and energy-efficient processing of AI models, enabling lower latency and higher throughput.
Sustainable computing: Vultr's eco-friendly cloud infrastructure helps users achieve energy-efficient and sustainable computing in large-scale operations with efficient AI technologies from AMD.

Visit our website to learn more about Vultr's composable cloud solutions.

View information about the AMD Instinct™ MI300X accelerator here.