It was a busy day at the Apache Software Foundation’s (ASF) offices yesterday, with not one, but two major software releases hitting the wire. The foundation announced the availability of Apache CouchDB v2.0, the popular NoSQL database, alongside the initial release of the Apache Kudu columnar storage engine that’s built to run alongside Apache Hadoop.

Apache CouchDB is one of several open-source NoSQL databases that have risen in popularity over the last couple of years, having been born from IBM Cloudant’s BigCouch project before being donated to the ASF. Used by names including Apple, Deutsche Telekom and DHL, the database is popular for its compatibility with JSON and HTTP, and because it offers a multi-master synchronization feature that makes it ideal for Big Data applications and mobile devices, the ASF says.

Now, with CouchDB 2.0, the project has finally fulfilled its original vision of bringing “reliable data sync to Big Data and Mobile in one seamless solution,” claims Jan Lehnardt, vice president of Apache CouchDB.

Apache CouchDB 2.0 packs a number of new features not seen in earlier versions, including built-in clustering for Big Data-scale and high-availability; 99 percent API compatibility compared to previous versions, which means most applications will work right out of the box; and the addition of “Mango”, a new developer-friendly querying language.

“CouchDB 2.0 is the result of several years of hard work, bringing Cloudant’s clustering code back into the project from which it sprang,” said Robert Newson, member of the Apache CouchDB Project Management Committee and Senior Database Engineer for Cloudant, IBM, in prepared comments. “2.0 reinvigorates the CouchDB project with new features and better performance. I hope the CouchDB community is as thrilled as I am to reach this milestone in our journey and hope many of you will be active participants in the journey ahead.”

Apache Kudu takes its bow

Apache Kudu is more recent addition to the ASF, first arriving on the scene at Strata/Hadoop World NYC in September 2015 before graduating as an Apache Top-Level project in July of this year.

Kudu is a columnar storage engine that’s designed to enable “fast analytics on fast data,” according to the project’s developers. The software is optimized for “lightning-fast scans,” which makes it particularly well suited for hosting time-series data, the ASF adds. The software also supports many traditional database operations, such as real-time insert, update and delete operations, and can be accessed by multiple query engines including Apache Spark, Apache Drill and Impala.

“The Apache Kudu 1.0 release represents a major milestone for the project,” said Todd Lipcon, Vice President of Apache Kudu, in a statement. “One year after the first public beta, the community is confident that Kudu is ready for production for critical business use cases.”

Along with being “production-ready,” Apache Kudu v1.0 comes with a number of features not seen in earlier, pre-release editions, including support for redundant and highly available Kudu Master nodes; support for manual management of range partitioning; rewritten integration with Apache Spark, including Spark SQL and Data Frame APIs; a client library for Python, and “substantial” performance improvements for random access and analytic workloads.

“Kudu 1.0 is the most performant, full-featured, and stable release of Kudu yet,” said Lipcon. “Kudu expands the Apache Hadoop ecosystem’s capabilities, enabling real-time data ingestion and updates while also serving high performance analytics with a substantially simplified architecture.”

Image credit: AdinaVoicu via pixabay