This article is part of a VB special issue. Read the full series here: The metaverse – How close are we?
Enterprises are fast adopting data warehouse technologies, which offer a system for reporting and analyzing mission-critical data. At their best, data warehouses allow companies to process, transform, and leverage data from a range of sources within the organization for decision-making. For example, data warehousing can make the practice of data mining possible, assisting businesses in looking for data patterns that can lead to higher sales and profits — at least in theory.
The data warehouse market remains red-hot, with Yellowbrick estimating that it’ll be worth over $30 billion globally by 2025. But companies — whether due to pandemic headwinds or otherwise — are encountering roadblocks in implementing key data warehouse components. According to a recent BARC Research survey, the most pressing issues for enterprises adopting data warehouse technologies are a lack of agility in the development process, expensive software licenses, and an inability to provide the required data.
Vendors like Firebolt propose a solution in cloud data warehouses, which migrate the management of data warehouses from on-premises infrastructure to cloud datacenters. Firebolt — one of the larger suppliers in the space — allows for the scaling up and down of compute resources across public clouds like Amazon Simple Storage Service, letting customers work with data in common file formants to prep it for analytics.
Moving to the cloud
The data warehouse market is on the cusp of switching from traditional deployment methods to becoming majority cloud-based, surveys show. A 2019 survey from Denodo found that more than half of businesses have a data warehouse in the cloud, a steep uptick from several years ago. Those embracing cloud solutions cite benefits like improved scalability, elasticity, and performance (in some cases) as well as more data storage.
Firebolt, which was founded in 2019 by Ariel Yaroshevich, Eldad Farkash, and Saar Bitner, claims that its cloud data warehouse software can support “over terabyte scale” data analytics. In fact, Firebolt claims to be the first company to deliver a “sub-second” data analytics experience regardless of the size and usage patterns of a customer’s data.
“[I] was the founder and CTO of Sisense. In 2013, Bitner joined as CMO of Sisense and eventually became GM Israel. [We] saw firsthand how, with terabytes of data, cloud data warehouses could not scale to deliver the performance and efficiency companies needed to power their business intelligence tools and analytics,” CEO Farkash told VentureBeat via email. “Companies that need to analyze data, whether for internal or customer facing purposes, are challenged with processing ever-growing amounts of data, while providing a great analytics experience.”
Firebolt offers a decoupled infrastructure that lets customers match analytics and data extract, load, and transform workloads with available compute resources. The company’s technology can automatically handle aspects like performance optimization, storage management, and server management, ostensibly simplifying data warehouse usage.
Firebolt allows programmatic control of its platform through APIs and software development kits for developers to create data applications. Structured query language — the standard language used in managing data in relational databases — can be used to control everything in the platform.
“[C]urrent cloud data warehouse deployments struggle with performance and high costs when it comes to analytics at terabyte scale, limiting their ability to address the needs of modern data applications,” Farkash added. “The growing pressing need for a data warehousing solution that efficiently manages huge volumes of data has become a top priority for software and data engineers. With Firebolt, the compromise taking place today when analytics is run over big data, can go away, so data can once again be fresh, granular, and allow the analysis of the entire historical data if needed.”
Firebolt competes with a range of cloud data warehouse vendors, including Snowflake, Microsoft (through Azure Synapse Analytics), Amazon (Redshift), and Google (Big Query). But through Firebolt’s technology, Farkash believes that data warehousing can evolve from serving internal business intelligence scenarios to a broader range of applications. The bulk of Firebolt’s “few dozen” paying customers, which include SimilarWeb, Explorium, and AppsFlyer, use Firebolt for customer-facing and operational data apps as well as business intelligence.
Cloud data warehouses don’t come without drawbacks, including a potential lack of governance, increased latency, and visibility (depending on the vendor). As data management provider Astera notes, organizations might struggle to balance resources and grant permissions efficiently, resulting in unnecessary (and costly) load on the system and creating bottlenecks that could have been avoided. Companies might also lack a proper structure for access control, opening up source systems to access by unauthorized users.
But this hasn’t tempered investors’ appetite for cloud data warehouse technologies. Firebolt today announced that it raised $100 million in a series C financing round, bringing the company’s total capital raised to $269 million at a $1.4 billion valuation. Alkeon Capital led the newest round with participation from Sozo Ventures, Glynn Capital, Zeev Ventures, Angular Ventures, Dawn Capital, Bessemer Venture Partners, K5 Global, and TLV Partners.
“The digital transformation we’re experiencing in every part of our lives due to COVID-19 has translated into very significant data growth rates for businesses. [T]hey’re already on the lookout for solutions that can support [this] growth,” Farkash said. “Being a software-as-a-service platform, Firebolt eliminates many of the [associated] labor intensive tasks and delivers ‘super powers’ to the data and engineering teams … The capabilities we’ll be working on [in the future] will enable connectivity with other technologies in the modern data stack, so Firebolt can become more of a plug and play option for technology companies. We [also plan to] continue to invest in making Firebolt even more efficient and developer-friendly.”
Firebolt has doubled in size in the last six months to 200 team members, Farkash says — the vast majority of them engineers. The company plans to hire 100 to 150 additional employees in 2022.
VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Learn More