Onehouse Emerges from Stealth to Deliver Data Lakes in ‘Months, Not Years’

Onehouse, a data lakehouse management company, has emerged from stealth today with $8 million in seed funding from investment firms Greylock Ventures and Addition.

Onehouse’s cloud-native managed lakehouse service is based on Apache Hudi, a data platform created by Onehouse Founder and CEO Vinoth Chandar while working at Uber in 2016. Hudi (pronounced like “hoodie”) is a data lake technology that allows for stream processing and data clustering and optimization techniques in Apache Hadoop-compatible cloud stores and distributed file systems.

Data lakes are popular with organizations who have data needs beyond a traditional warehouse, which is an architecture that is easy to use but can become extremely costly as the scale of a company’s data and AI/ML workloads increases, especially in exabyte scale operations.

Chandar encountered these challenges while building a data lake at Uber, as the company needed a high performance, large scale solution capable of supporting AI and ML workloads in near real-time due to the rapid pace of its ride sharing enterprise. Hudi was created to give core warehouse and database functionality to a data lake, a setup which gave rise to the term “lakehouse.”

Data lakes are not without their unique problems, including constantly tuning data layouts, large-scale concurrency controls, fast data ingestions, stale or unreliable data, and data deletions. Building a data lake can be risky because it can take a large investment of time and resources and there is often not enough talent available for the highly skilled engineering teams needed to develop it. Also, lake projects can become stalled or abandoned due to a lack of standardized, high quality data infrastructure centered around lakehouse technologies.

(Source: Onehouse)

“Onehouse is a cloud-native, managed foundation for your lakehouse that automatically ingests, manages and optimizes your data for faster processing,” said Chandar in a blog post. “Onehouse is not another query engine, but a self-managing data layer that seamlessly interoperates with any of the popular query engines [e.g., Apache Spark, Trino, Presto] or data/table formats and vendors out there.”

Chandar notes that Onehouse is not an enterprise version of Hudi, but that it leverages Hudi’s capabilities in order to enable incremental data processing much faster than traditional batch processing. The company claims it can help organizations build data lakes in “minutes, not months” while saving money and retaining rights to their own data without being bound to individual vendors.

“The data lakehouse is the future of data lakes, providing customers the ease of use of a data warehouse with the cost and scale advantages of a data lake,” said Greylock Partner Jerry Chen. “Apache Hudi is already the de facto starting point for modern data lakes and today Onehouse makes data lakes easily accessible and usable by all customers.”

In Search of the Modern Data Stack

The Apache Software Foundation Announces Apache Hudi as a Top-Level Project

The post Onehouse Emerges from Stealth to Deliver Data Lakes in ‘Months, Not Years’ appeared first on Datanami.