The Apache Flink Top-Level project has steadily been gaining traction over the last 18 months as an alternative to the more widely used Apache Spark. Now, data Artisans GmbH is ready to capitalize on Flink’s growing reputation with the release of version one of its dA Platform, the first fully supported commercial distribution of the alternative data stream processing engine.

Apache Flink was first conceived as a replacement for the MapReduce component in Hadoop, one that’s capable of digesting streaming data in addition to regular old batch data. That description makes it sound an awful lot like Spark, but really doesn’t do it justice, because Flink offers some unique capabilities that Spark doesn’t have. One of them is the ability to analyze streaming data “in-memory,” which makes its overall processing speed considerably faster than that of Spark’s.

Flink also has the considerable advantage of being able to unify the processing of historical and real-time data, thanks to its scalable framework, data Artisans says.

Apache Spark can do this too, but according to Leon Doitscher in Medium the process is somewhat “clunky” when compared to Flink because Spark was designed to handle batches of data rather than sequential data that’s constantly flowing in. “Spark Streaming” was developed as a work around, but the process still isn’t “real, real-time,” Doitscher says, because it necessitates accumulating data in batches for a short time before processing.

In contrast, “Flink is a true stream processor with low data latency on event analytics,” said data Artisans’ co-founder and Chief Executive Kostas Tzoumas during an appearance on SiliconANGLE’s TV show theCUBE at the #BigDataSV event earlier this year. “So you can have an event and it is flagged in milliseconds. Additionally, Flink takes into account that the streams you are receiving do not arrive in order.”

With those advantages in mind, data Artisans is pitching its Apache Flink-based dA Platform as a superior data stream processing solution for the “rapidly growing” base of enterprise customers who’ve already moved high-performance streaming applications into deployment.

During his appearance on theCUBE, Tzoumas said stream processing was only just now beginning to take off, driven by data from the Internet of Things (IoT) applications and a growing demand from companies hoping to make informed decisions in real-time. Nonetheless, Flink has already established a significant following, with more than 200 contributors to the project and high-profile users including Alibaba Group Holding Ltd., Capital One, Ericsson, Bouygues Telecom, ResearchGate, Zalando and Otto Group, to name just a few.data analytics

Data Artisans says its dA Platform aligns with its strategic mission to build a strong community of Apache Flink users. As such, the company’s distribution comes with 24/7/365 enterprise support, as well as on-site consulting, tailored training for specific use cases, and direct lines of communication with Flink’s original creators. In future, the company is also planning to add proprietary modules of its own to make Flink easier to deploy and use, and provide expanded access to different types of data.

“As the original creators of Apache Flink, data Artisans is uniquely qualified to help enterprises overcome any challenges they may experience with integration and deployment as they push the boundaries of this technology, and we already have a range of active service accounts,” the company said.

Check out Tzoumas’s full interview on theCUBE below: