Smartphones, tablets, laptops, PCs, and internet of things devices generate more data today than at any time in history — close to 2.5 quintillion bytes, some say. But we’re woefully unequipped to process and analyze that data, according to Jeff Denworth, Renen Hallak, and Shachar Fienblit. They’re the cofounders of Vast Data, a company founded on the premise that enterprise storage platforms — platforms that by and large tap segregated, independent clusters that don’t explicitly share bits among themselves — are fundamentally broken. Vast is instead developing a disaggregated, shared-everything (DASE) architecture that promises not only improved robustness compared with conventional alternatives, but greater scalability at a lower operating cost.
To stake its claim in the nearly $118.52 billion big data market, Vast emerged from stealth today with $40 million in series B financing contributed by TPG Capital and existing investors Norwest Venture Partners, Dell Technologies Capital, 83 North, and Goldman Sachs. This brings the New York startup’s total capital raised to $80 million and follows a quarter in which product bookings from Ginkgo Bioworks, General Dynamics Information Technology, Zebra Medical Vision, and other customers “exceeded” rates of top enterprise tech companies.
“Storage has always been complicated. Organizations for decades have been dealing with a complex pyramid of technologies that force some tradeoff between performance and capacity,” said Hallak, who serves as CEO. “VAST Data was founded to break this and many other long-standing tradeoffs.”
To that end, Vast’s “exabyte-scale” storage architecture leverages technologies like NVMe over Fabric (designed to facilitate rapid data transfer between host machines and target solid-state storage devices) and storage-class persistent memory (memory that preserves content during a power cycle) — with app-aware data placement techniques and a large SCM write buffer optimized for low-cost quad-level flash storage — to enable management and control from a single source in real time. In practice, Hallak claims it obviates the need to juggle multiple systems across datacenters.
June 5th: The AI Audit in NYC
Join us next week in NYC to engage with top executive leaders, delving into strategies for auditing AI models to ensure fairness, optimal performance, and ethical compliance across diverse organizations. Secure your attendance for this exclusive invite-only event.
Servers in Vast’s namespace have access to all media in all clusters, ensuring that speedy NVMe media — not DRAM-based accelerators or tiered hard drive storage — service file reads and writes. Moreover, Hallak says the architecture can scale to “near-infinite” numbers because the servers don’t need to coordinate with each other, and because they aren’t impacted by the crosstalk characteristic of clustered systems.
“By applying new thinking to many of the toughest problems,” Hallak added, “we are working to simplify how customers store and access vast reserves of data in real time, leading to new insights that were not possible before.”
Those aren’t the only innovations Vast’s solution brings to the table. Its data protection system — which requires only around 2 percent overhead — increases the resiliency of clusters as they grow, and its novel data deduplication scheme discovers patterns of similarity across the entire namespace at 4,000 to 128,000 times the granularity of most compression algorithms. Additionally, Hallak says that Vast servers can be containerized and embedded into application servers, bringing NVMe over Fabrics to potentially many multiple hosts.
“Zebra [is] a fast-growing company, and we don’t have the resources to become … storage technicians,” said Eyal Toledano, chief technology officer at Zebra Medical Vision. “Vast provides a solution to all of our storage challenges by delivering performance superior to what is possible with traditional [network-attached storage] while also providing a simple, scalable appliance that requires no effort to deploy and manage.”
Vast’s storage platform can be deployed as a turnkey server and storage cluster appliance (up to dozens of petabytes in a single rack), or storage plus Vast container software that runs on customer machines (or software only). The company is extending new and existing customers a 10-year warranty.
In addition to its New York HQ, Vast has offices in Tel Aviv and San Francisco.
