Teradata recently announced Kylo, data lake management software meant to help businesses build and manage data pipelines. Kylo, which is open source software, addresses three top data lake issues: long build times, high engineering costs, and low user adoption.

Using Kylo is a three-step process. First, users ingest batch or streaming data with a graphical user interface (GUI), and a schema based on the data generates automatically. Second, customers prepare their data with 100+ built-in transformations and drag-and-drop joins. Third, once preparation is complete, the data is ready to publish. Kylo also tracks metadata and data lineage so data stewards can see what data is available and explore that data’s history.

“Teradata is positioned to be a top-tier player in the data lake and the open source worlds,” says Dave Wells, Director of Eckerson Group’s data management practice. “They have funding, name recognition, an established customer base, and lots of reach.” With Kylo, Teradata strengthens the value of its big data practice and earns Teradata credibility as an open source contributor.

Teradata built Kylo because many big data projects start with a long, expensive data lake build, according to company officials. Kylo is designed to help Teradata customers deploy data lakes faster with a return on investment.

Market Impact  

Kylo competes with Podium and Zaloni. All three offer a “user-friendly application layer with data ingest, data preparation, and discovery abilities,” said Mark Shainman, marketing director at Teradata. However, Kylo’s strength comes from its extensibility. Kylo leverages Apache NiFi so customers can build new pipelines easily. Kylo also comes with a plug-in architecture, so developers can extend Kylo’s capabilities to fit their needs.  

Kylo was designed with self-service in mind. Shainman explains, “Kylo is completely oriented to a data analyst skill-set. The design goal behind Kylo was to enable user self-service and remove the need for IT and software developers.” Kylo offers features oriented towards business users and requires no coding skills or Hadoop experience.

 Kylo support, offered by Think Big Analytics, has three levels: foundation, standard, and enterprise (24x7). Foundation is free for six non-production months. Prices are determined by Hadoop cluster size.