VDURA’s CEO Breaks Down New V5000 All-Flash Appliance

The new offering is engineered to meet the demands of computing-intensive AI workflows, CEO Ken Claffey said.


(VDURA CEO Ken Claffey)

VDURA, formerly known as Panasas, and provider of high-performance data storage solutions, announced on Tuesday the launch of its V5000 All-Flash Appliance, designed to address growing demands for complex AI applications and environments.

[RELATED: Rubrik, Pure Storage Join Forces To Deliver 'Three-Layer Defense']

The appliance is built on the VDURA V11 data platform, encompassing VDURA V11 software and is available on “the new 'F Node,' a modular 1U platform with Intelligent Client-Side Erasure Coding—delivering over 1.5PB per rack unit,” the company said in a news release

The system “delivers GPU-saturating throughput while ensuring the durability and availability of data for 24x7x365 operation conditions,” according to the news release.

The platform is built with parallel file system (PFS) architecture and supports virtually limitless GPU clustering, providing high-performance computing (HPC), storage expansion and fault tolerance.

“We've taken our latest version of our VDURA data platform, and we have now certified an all-flash hardware offering. So, you put that software and the hardware together [and you’ve] got this all-flash appliance,” VDURA CEO Ken Claffey told MES Computing.

Radium, an Nvidia cloud partner, chose VDURA to power its next-generation AI infrastructure for massive-scale model training, according to the news release.

Claffey provided more details on the new offering.

Parallel File System Advantages

Claffey broke down the benefits of a parallel file system versus a traditional file server.

“The difference with [PFS] is really ... in a traditional file server, you’ve got the file server, and you can only go as big and as fast as that file server lets you go,” Claffey said.

“[PFS] blows that all up” he said.

PFS-based systems like the V5000 can scale far beyond traditional file servers, he explained.

“It takes the data and the metadata and separates them, so you have the metadata goes over here, and your data goes to a cluster of file servers. So just like you're building a compute cluster where the cluster is a function of n number of nodes, with a parallel file system, the performance of the storage cluster can match it, because you can just add more file servers as you go,” he added.

AI And A Rebrand

A sweet spot for V5000 is AI particularly when used with resource-intensive tasks like computer-aided design and computer-aided engineering.

“With the advent of AI, if you're doing AI at any scale, you need a parallel file system. That's not my view. That's Nvidia's view. If you look at their reference architectures, they're basically saying you need fast storage, and that is defined as a parallel file system,” Claffey said.

[RELATED: InsightFinder AI Aims To Give Data Scientist Powers To IT]

“So, we kind of looked and go, OK, we've got this great core technology for HPC. How do we bring it to ... this burgeoning world of AI?”

Claffey said that shift to an AI focus was the impetus to rebrand the company to “VDURA,” a portmanteau of “velocity meets durability.”

“We were tied to a particular type of hardware appliance, frankly, and a type of architecture. We've moved everything through microservices architecture. Now we're hardware agnostic from a functionality perspective,” he said.

Who Is VDURA And The V5000 All-Flash Appliance For?

The majority of VDURA’s customer base are those who have high-performance computing demands. Many of its customers are in research and development, engineering, manufacturing and other verticals with complex computing needs, especially those who do a lot of simulation and modeling.

Is This A Midmarket Solution?

For midmarket organizations, VDURA’s V5000 solution can reduce the barriers of entry to adopting a high-performance computing platform for complex tasks, for example, if a midsized company is “an engineering-centered company that's either making a product or even providing consulting services,” he said.

He said the company has a customer in biotech with about 50 employees. “Those type of customers, they need the ability to start at a very small scale, right? They don't want to invest 10 billion ... they don't even want to invest tens of millions of dollars in infrastructure, right? And that is one of the advantages that I think we have,” he said about providing midmarket companies access to a high-performance data storage solution like V5000.