Investors have provided StarTree Inc. with $24 million in funding to help the startup launch a commercial version of Apache Pinot, a speedy open-source data platform that can ingest millions of pieces of information per second.

Pinot was developed by StarTree’s founders during their time at LinkedIn. The startup’s $24 million funding round, announced today, follows an earlier $4 million raise that included LinkedIn as well as numerous prominent enterprise software executives. Among the participants were Confluent Inc. co-founder Neha Narkhede, GitHub Chief Technology Officer Jason Warner and DataStax Inc. Chief Executive Officer Chet Kapoor. 

StarTree sees a growing trend of companies building analytics features into their products to differentiate their feature sets. Uber Technologies Inc., for example, has built an analytics dashboard into its UberEats service that helps restaurant owners measure sales. StarTree believes more companies will add such data-centric features to their software and that they’ll need a new type of backend data system to power those features.

Enter Pinot. The startup says that the platform has capabilities that makes it particularly well-suited to support user-facing analytics features in consumer apps and other services. The database returns query answers with millisecond latency, according to StarTree, which makes it easier to provide the quick response times that users generally expect from modern software. Pinot can moreover support millions of users, which is essential for consumer tech firms whose applications are used by a large number of individual customers. 

StarTree claims that Pinot also addresses two other common requirements of user-facing analytics tools. One is support for so-called multidimensional data and the other is the ability to analyze historical records alongside real-time information.  

A retailer’s store management system, for example, might have one dashboard that visualizes the past quarter’s revenues and another showing new purchases from the last 30 minutes. Displaying new purchases requires pulling transaction logs in near real-time from the retailer’s network of stores. For an application to analyze such real-time records together with historical information such as quarterly revenue figures, it needs a backend data system that can support both types of data. 

Pinot’s support for multidimensional data, another selling point touted by StarTree, means it can process complex records containing multiple fields. Even a store purchase log can contain numerous individual details, such as what was bought, where, when and whether the customer used a discount code. Such detailed information can be a valuable source of insights for the kind of analytics applications StarTree is hoping to power with Pinot.

The startup says that its commercial version of the database will be delivered from the cloud in the form of an “analytics-as-a-service” platform. StarTree promises a “one-click experience from data to insights” on its website, which hints that the platform will have features to make Pinot easier to use. Many cloud versions of open-source projects provide capabilities to simplify operations for customers, particularly the initial deployment and day-to-day maintenance tasks such as patching.

Open-source projects have given rise to multiple major players in the enterprise data management market. Confluent, which sells a  commercial version of Apache Kafka, recently filed to go public after previously achieving a $4.5 billion private valuation. DataStax, the startup commercializing the Apache Cassandra database, has raised $190 million from investors.

Bain Capital Ventures led StarTree’s new $24 million round. GGV Capital and returning investor CRV participated too.

Photo: Unsplash