Quick News Spot

Vast Data's Enter the Cosmos event: Explore AI cloud services - SiliconANGLE


Vast Data's Enter the Cosmos event: Explore AI cloud services - SiliconANGLE

What to expect at Vast Data's 'Enter the Cosmos' event: Join theCUBE Oct. 1

There is disruption underway in the cloud industry itself as businesses begin to look outside of the major providers to support private artificial intelligence and AI cloud services.

The growth of AI has led to a need for infrastructure that can rapidly optimize data flows and quickly scale resources to support next-generation enterprise workloads. Vast Data Inc. has emerged as a central player in this developing saga.

"As we progress, what we see more and more is clouds -- and not just the top three clouds that everybody knows and talks about -- we see more and more specialized AI clouds, and we see the big clouds spilling over sometimes into these more regional clouds in order to provide services to their customers faster," said Renen Hallak, co-founder and chief executive officer of Vast Data, in an exclusive interview with theCUBE, SiliconANGLE Media's livestreaming studio. "We are enabling those clouds."

TheCUBE will cover the Vast Data virtual event on October 1, offering firsthand reporting and exclusive interviews with data platform experts, analysts and industry insiders to discuss the developments driving innovation in enterprise AI and beyond. (* Disclosure below.)

Vast Data got its start by creating storage systems that could manage data's massive growth. Yet Hallak and his co-founders came to a realization that the integration of storage with robust processing capabilities would set the stage for what they have labeled "the new era of data infrastructure."

This has evolved into the Vast Data Platform, with four components designed to operationalize AI cloud services across compute environments. "We started by building a storage system to address this extreme size of data ... and the fact that you need very fast access to it for GPUs," Hallak said. "What we ended up building is not a storage system, but an operating system."

This notion of an operating system for data has evolved to support specialized infrastructure led by newer purpose-built AI cloud providers such as Genesis Cloud, Core42, CoreWeave, Lambda and Nebul. Vast Data built its platform to include processing for structured and unstructured data and to understand natural data through an embedded queryable semantic layer.

"Storage is way more than flash or disk anymore. Vast Data has taken the Data part of their name seriously in building a platform optimized to be the infrastructure for data apps," says Rob Strechay, managing director and principal analyst at theCUBE Research. "I am looking forward to the announcements Vast Data will make around Cosmos, aiming to advance data platforms for organizations and service providers building AI applications."

Another area of Vast Data's business to watch for news about during the upcoming event involves its collaboration with Nvidia Corp. and the use of data processing unit or DPU technology.

In March, Vast Data announced new AI cloud architecture in partnership with Nvidia that leveraged the chipmaker's BlueField-3 DPUs to host Vast Data's operating system. BlueField networking provides accelerated compute infrastructure for AI cloud services, and the collaboration with Nvidia enables Vast Data to embed storage and database at extremely large scale.

The result is not only a potentially significant power savings, but also a more streamlined way to manage the substantial number of GPUs currently needed to run major AI-based workloads. Data center developers can scale out rapidly without needing to reengineer systems for AI training, and cloud infrastructure provider CoreWeave is already using Vast Data's operating software on Nvidia BlueField DPUs in production, according to John Mao, vice president of business development and technology alliances at Vast Data.

"If you start with something relatively small, a few hundred GPUs, as you add 1,000 GPUs and then 10,000 GPUs, this new architecture allows those clusters to be able to scale performance of those new storage services completely linearly with no thinking and no rearchitecting," said Mao, in an interview with SiliconANGLE.

During Vast Data's "Enter the Cosmos" event, theCUBE Research's analysts will talk with data platform experts, analysts and industry insiders to discuss the developments driving innovations in enterprise AI and beyond.

Previous articleNext article

POPULAR CATEGORY

corporate

2851

tech

3144

entertainment

3422

research

1436

misc

3640

wellness

2681

athletics

3554