Data is driving the enterprise. How many times has that sentence been written? But it’s true. Not only is data driving the enterprise, but it’s also being generated at unprecedented rates in a seemingly infinite number of formats. The successful data-driven business is the one that can find the proverbial needle in the haystack in the shortest amount of time.
It’s in this context that high-performance storage vendor VAST Data made a major announcement earlier this week, unveiling the VAST Data Platform, which is aimed at enabling organizations to gain faster and deeper insights from all the data across the enterprise. What exactly was announced, and what does this mean for enterprise IT? I’ll explain below.
Data — the enterprise opportunity and the challenge it presents
While data drives the enterprise, it is sometimes hard to tell if it is driving it in the right direction—sources that it sometimes seems impossible to collect, aggregate, transform and make sense of all the data while it’s still relevant.
As if that challenge isn’t big enough, let’s throw data locality into the equation as well. Data resides everywhere, of course—on the edge, in the cloud and scattered across the enterprise. As more and more data is generated and that data becomes more complex, it gets increasingly difficult to extract real intelligence from it for business users and data scientists.
More than that, as datasets grow from terabyte to petabyte scale and beyond, enterprises need an easier way to store and organize their data to optimize it for deep learning and deep analysis. Think highly performant storage and data lakehouse technology consolidated into a single, integrated offering.
Finally, this unprecedented growth in data is largely unstructured. And this is the essence of the enterprise challenge. How can my organization take all these images or text files and get them into a state where that can be queryable?
Companies across life sciences, media, financial services, transportation and other industries are dealing with this problem/opportunity every day. And this is the challenge VAST is aiming to solve.
VAST Data Platform
When VAST talks about enabling the enterprise, it does so with three guiding principles:
- Embrace standards. Allow organizations to leverage the technology (and knowledge) they have in house to the greatest extent.
- Simplify infrastructure. I think about storage infrastructure and how siloed it can become.
- Fix the hard problems without compromise. Forget about the tradeoffs that companies traditionally make such as price v. performance or scale v. complexity. Rethink the equation, enabling customers to solve the tough challenges without introducing other challenges.
Out of this thinking comes the VAST Data Platform. This solution promises consolidation of storage, data management and the tools necessary to simplify data collection, synthesis and continuous analysis across the enterprise.
The VAST Data Platform comprises four components I’ll cover at a high level. First, the VAST DataStore is the company’s highly scalable storage solution that some of the largest (and most performance-hungry) organizations already use.
The new VAST DataBase is one of the key elements of the Data Platform solution. Natively integrated into the system, DataBase includes the constructs of a traditional database, data warehouse and data lake in a single database management system (DBMS). Yes, DataBase is a DBMS.
VAST DataEngine creates a single global compute framework for a data environment, both on-premises and in the cloud. This means a single compute framework that consolidates the entire data environment. DataEngine is slated to release in 2024.
Finally, the VAST DataSpace delivers what one would expect from a global namespace—simplicity, redundancy and mobility. All resources are presented and managed as a single resource, with global enforcement of policies.
What this means for enterprise IT
A successful modern business makes the right bets faster than its competitors. This could be a pharmaceutical company working on a new drug, a lab working on the cure for a pandemic or a financial institution that can anticipate and be the first to respond to market fluctuations and trends. In each of these examples, accurately interpreting data is critical to success. Also, in each case, the more data that can be incorporated, the better.
The model that supports such use cases goes beyond the traditional machine learning we’ve become familiar with and moves into a deep learning use case. I’m talking about GPU-based neural networks that train on exabytes worth of data for inferencing. These environments continuously train as more and more data is generated and applied to the training models.
This is a real-world scenario that is playing out every day in enterprises around the globe. And these organizations benefit greatly from this consolidation of the AI stack. While such solutions are expensive, the performance they deliver can help create a return on investment that makes the upfront costs seem trivial.
While some businesses are still figuring out how to use AI effectively across the enterprise, a few verticals seem to be leading the market, including life sciences, media, transportation and financial services. Companies in these verticals that are already well into their AI journeys should see tangible and measurable value in what VAST is doing.
VAST’s moves here make sense. The lines between storage, data management and computing are becoming increasingly blurred. And each element is so dependent on the others that whichever company can deliver data management, storage, computing and a richness of functionality should be able to differentiate itself in the market.
Is it easier for a data management company to come down the stack versus a storage company moving up the stack? That’s a question some ask, but I don’t think it’s a that question can be quickly or easily answered. It really depends on the company and the people in the company. I do believe that VAST has a clear vision of where it wants to go (and grow) as a company and is executing against a well thought strategy.
Forget about the ”Are they are a storage company or data management company” question and think about whether the company understands data. Again, I think VAST gets it.
VAST’s data management vision is the future of the market. The company’s success in the high-performance computing space has given it an understanding of what’s needed in terms of architecture to impact performance and cost at scale. I’m certain competitors are sure to follow.
As usual, I’ll be tracking how VAST Data Platform lands in the market, and I’ll be sure to update the story over the following quarters. Stay tuned.