Partnerships pitch new data center architecture and simplified accelerated pipelines.
NVIDIA and VAST Data have collaborated over the design of new unveiled AI cloud architecture.
Pitched as boosting performance, service, zero-trust security and efficiency to help organisations build and scale AI services, the new architecture embeds storage and database processing services directly into AI servers and delivers linear data services designed to scale to hundreds of thousands of graphics processing units (GPUs).
VAST has also announced a new partnership with server, AI, storage, IoT and switch systems specialist Supermicro to deliver a full-stack, AI system that simplifies the creation and expansion of large-scale AI deployments.
The system offers service providers and enterprises multi-tenancy, zero-trust security and scalability, which is useful for organisations dealing with large volumes of proprietary data.
VAST Data unveils new data center architecture for the AI Factory
Designed in collaboration with NVIDIA and being deployed at CoreWeave, the new infrastructure unleashes VAST’s parallel data services architecture on NVIDIA BlueField DPUs to simplify and scale AI.
Building on NVIDIA BlueField-3 data processing unit (DPU) technology, VAST Data’s parallel system architecture makes it possible to disaggregate the entirety of VAST’s operating system natively into AI computing machinery – transforming supercomputers into AI data engines.
The NVIDIA BlueField networking platform combines robust compute power and integrated hardware accelerators to create secure and software-defined accelerated computing infrastructure for AI.
By outfitting each graphics processing unit (GPU) server with a dedicated NVIDIA BlueField DPU
running a stateless container that powers the VAST parallel services operating system, this new architecture design embeds storage and database processing services directly into AI servers and delivers true linear data services designed to scale to hundreds of thousands of GPUs.
Moreover, by removing multiple layers of x86 hardware and networking from VAST’s network-attached Data Platform infrastructure, this new AI factory architecture dramatically reduces the cost, footprint and power associated with AI data services.
“We’re extremely proud to partner with NVIDIA to help industrialise AI computing,” said Jeff Denworth, co-founder, VAST Data.
“This new architecture is the perfect showcase to express the parallelism of the VAST Data Platform. With NVIDIA BlueField-3 DPUs, we can now realise the full potential of our vision for disaggregated data centers that we’ve been working toward since the company was founded.”
This new VAST architecture – running VAST software on BlueField DPUs in the AI servers – is being tested and deployed first at CoreWeave, the specialised GPU cloud provider.
VAST and CoreWeave began partnering in 2023 to build some of the world’s most scalable AI machinery and to help many of the world’s leading LLM builders and blue-chip enterprise customers build their own AI factories.
“With VAST’s operating system, next-generation accelerated computing solutions are paired with next-generation accelerated network infrastructure, enabling enterprises and service providers to benefit from simpler, more secure experiences with high-performance systems,” said Rob Davis, Vice President of Storage Technology, NVIDIA.
Peter Salanki, VP of Engineering, CoreWeave, said: “VAST’s revolutionary architecture is a game-changer for CoreWeave, enabling us to fully disaggregate our data centers. We’re seamlessly integrating VAST’s advanced software directly into our GPU clusters.
“Leveraging NVIDIA BlueField DPUs, we’ve been at the forefront of creating sophisticated, software-defined data center abstractions. Now, by natively incorporating storage and database services onto
BlueField, we’re not just streamlining our infrastructure but we are also elevating the user experience for our customers by removing bottlenecks in the AI data computing pipeline. CoreWeave is not just keeping pace with the future of cloud data management – we are defining it.”
VAST Data and Supermicro Partner to Simplify and Accelerate AI Pipelines
New collaboration combines VAST Data software, Supermicro hardware and NVIDIA technologies to simplify deployment of hyperscale AI infrastructure.
Through this collaboration, VAST Data and Supermicro are offering service providers, hyperscale technology companies and large, data-centric enterprises with a highly performant and resilient AI solution powered by NVIDIA-Certified Systems.
The solution offers service providers and enterprises multi-tenancy, zero-trust security and scalability, making it invaluable for organisations dealing with large volumes of proprietary data and requiring agility in their operations.
“In an era where organisations globally are launching and scaling their AI initiatives at a blistering pace, staying competitive requires them to work with infrastructure providers who are proven at hyperscale,” said Renen Hallak, CEO and co-founder, VAST Data.
“Collaborating with Supermicro, we’re delivering a unified hyperscale data platform that will make it easier for service providers and enterprises to be competitive in the AI race. With a revolutionary parallel architecture that is built with enterprise data management tools for unstructured and structured data, this joint solution is the ideal platform for the modern era of AI computing.”
“Supermicro is excited to collaborate with VAST Data to deploy their VAST Data Platform,” said Charles Liang, President and CEO, Supermicro.
“Leveraging Supermicro’s NVIDIA-Certified Systems portfolio with VAST Data’s software platform combines the compute and data functions into a Supermicro platform that’s ideal for large-scale AI deployments – enabling significant Deep Learning performance.”
Click below to share this article