Voltron Data Inc., a new startup focused on the open-source Apache Arrow framework, today announced that it has raised $110 million over two funding rounds.

San Francisco-based Voltron Data was founded in 2021. Since launching, it has raised a $22 million seed round and an oversubscribed $88 million Series A investment led by Walden Catalyst. The startup’s other investors include BlackRock, Anthos Capital, Battery Ventures, Coatue, Alphabet Inc.’s GV venture capital arm, Lightspeed Venture Partners, Nepenthe Capital, Redline and The Factory.

Voltron Data is reportedly developing a managed version of Apache Arrow, a popular open-source software framework used for data analytics. The startup says that Arrow is downloaded more than 42 million times per month. Amazon Web Services Inc., Microsoft Corp., Google LLC and Snowflake Inc. are among the companies that have adopted the framework. 

“Voltron Data is focused on making the data science ecosystem more efficient with modular and composable analytics building blocks,” said Voltron Data co-founder and Chief Executive Officer Josh Patterson. “Modularity and composability help us build bridges — not walls — across the analytic ecosystem.”

Arrow was created with two main goals in mind. The first is making data analytics applications faster to help organizations more quickly extract useful insights from their business information. Arrow’s other main objective is enabling applications to more easily exchange data with one another.

Different applications store the information they process in different ways. A spreadsheet tool, for example, stores information as rows and columns. An analytics application that a company uses to find ways of optimizing its operations likewise organize the data it processes in a specific format.

The way data is organized in a server’s memory directly influences the speed with which it can be processed. As a result, developers building analytics software often implement custom ways of organizing information in memory so as to speed up processing. This task requires a deal of time and effort.   

Arrow provides a ready-to-use format for organizing data in memory that spares developers the hassle of building everything from scratch. Moreover, Arrow includes optimizations that help analytics applications process data faster. One of the framework’s core features is a capability called zero-copy reads, which allows analytics applications to access information without having to make a copy of it first. Skipping this step speeds up calculations.

The other major benefit of Arrow is that it makes it easier to share data between systems. 

Different applications often store information in memory using different approaches. As a result, moving a record between two applications frequently requires changing the structure of the record. Changing the way data is organized can require a significant amount of hardware resources, especially when large volumes of information are involved. 

In contrast, if two applications both use Arrow to organize their data, they can easily exchange data without having to change its structure. The result is that information can be shared considerably faster. Moreover, developers have to write less code to manage the task of exchanging information between systems. 

Voltron Data is the largest corporate contributor to Arrow. The startup is developing a managed version of Arrow that will make the framework easier to use, according to TechCrunch.

Voltron Data currently has 100 employees. The startup plans to use its recently raised funding to double its workforce, primarily by expanding the engineering and go-to-market teams. 

Image: Voltron Data