Vast Data Joint Reference Architecture With Nvidia

Vast Data, Inc. announced a reference architecture based on Nvidia DGX A100 systems and an Universal Storage platform.

This reference architecture is designed to increase storage performance for AI use cases such as large-scale training of conversational AI models and petabyte-scale data analytics. Jointly designed, built and tested by the 2 companies, it eliminates the guesswork of building a solution by providing enterprises with a turnkey petabyte-scale AI infrastructure solution that maximizes performance without adding needless cost and complexity.

Historically, enterprises have been forced to choose between 2 specific infrastructure configurations based on workload (i.e. GPU-intensive or storage-intensive), but as applications and data science teams’ requirements evolve, that infrastructure choice may limit and negatively impact performance.

Built on the company’s recently released LightSpeed platform and universal AI system DGX A100, the architecture leverages the firm’s capabilities such as NFS-over-RDMA, NFS Multipath, and support for Nvidia GPUDirect Storage as well as a converged fabric design. The joint reference architecture delivers more than 140GB/s of throughput for both GPU-intensive and storage-intensive AI workloads.

“For the first time, enterprises, and more importantly data science teams, are no longer constrained by the limitations of rigid infrastructure configurations,” said Jeff Denworth, co-founder and CMO, Vast Data. “We’ve worked with Nvidia on this new reference architecture, built on our LightSpeed platform, to provide customers a flexible, turnkey, petabyte-scale AI infrastructure solution and to remove the variables that have introduced compromise into storage environments for decades.”

“AI workloads require specialized infrastructure, which is why we’ve worked with Vast Data, a new member of the Nvidia DGX POD ecosystem, to combine their storage expertise with our deep background in optimizing platforms for AI excellence,” said Tony Paikeday, senior director, AI systems, Nvidia, Corp. “This new reference architecture provides customers with a winning formula to achieve their AI success.“

AI-Ready ecosystem quotes
Organizations looking to take advantage of AI can access AI assessments to guide them as they build out and deploy high-throughput AI architectures based on DGX systems and the firm’s LightSpeed storage with the following resellers.

Mark III Systems
“We offer a diverse and unique portfolio of full stack technology platforms. As our team of data scientists, DevOps, and solution architects work with enterprises to build out and manage their applications and pipelines from end to end, we are constantly looking for ways to provide the performance needed for innovative organizations. We’re seeing significant interest with VAST in multi-petabyte solutions looking to increase technology adoption while removing complexity,” said Stan Wysocki, president, Mark III Systems.

Trace3
“Our customers are increasingly looking to AI to provide meaningful insights to give them a competitive advantage and deliver a better experience to their customers. At Trace3, we are helping our customers at every level of what it takes to make AI impactful. With Nvidia‘s DGX A100 and Vast‘s Universal Storage Platform, our customers can focus on the business problem and be completely free of worry about the limitations of their infrastructure. Vast Data’s LightSpeed with Nvidia‘s DGX A100 systems allow them to unleash the potential of their data – from ingest to insights,” said Matt Fornito, head, AI, Trace3.

Cambridge Computer Services
“Providing clients the resources, ideas, and expertise needed to leverage cutting-edge technologies to drive their business forward is our core tenet. We’re constantly identifying new capabilities that can help our clients optimize their workflows and overcome the cost, capacity, scale, and resilience challenges they’ve been facing for decades. Vast and Nvidia‘s joint reference architecture provides these capabilities for mission-critical enterprises that want to transcend the limitations of rigid infrastructure configurations,” said Jose Alvarez, director, industrial and scientific computing, Cambridge Computer Services.

Uclick
“We continue to work towards solving the data processing bottleneck that has emerged as a concern for many of our customers as they build out and operate their AI data centers. Vast‘s best-in-class performance combined with Nvidia‘s DGX A100 systems have paved the way for GPU-efficient computing using the industry standard NFS file system and has eliminated all bottlenecks. With this new reference architecture, we can further strengthen our AI solution lineup and provide our Korean customers with the latest and greatest as they look to maximize the value of their data,” said Namhan Eom, CEO, UClick.

Xenon Systems
“Providing our customers with the right technologies to drive outcomes for their clients has remained an important focus of ours. We recognize that data-intensive AI will always be a big part of that strategy. With Nvidia‘s DGX A100 servers and Vast‘s Universal Storage Platform we can provide a petabyte-scale AI infrastructure solution for GPU-intensive and storage-intensive AI workloads,” said Dragan Dimitrovici, CEO, Xenon Systems.

Resource:
Vast‘s GPU Reference Architecture, including specific tests showcasing both infrastructure level micro-benchmarks as well as benchmarks representative of real-world AI training workloads (registration required)