Canonical, the publisher of Ubuntu and the provider of open source security, support, and services, is releasing Charmed Spark, an advanced solution for Apache Spark that accelerates data engineering across both public cloud and private data centers.
Useful for diverse data processing applications—such as predictive analytics, data warehousing, machine learning (ML) data preparation, and extract-transform-load (ETL)—Charmed Spark is engineered to run Spark on Kubernetes, empowering cloud-native portability across environments.
“Enterprise data engineers want Apache Spark with the ease and long-term security commitment of Ubuntu”, said Mark Shuttleworth, CEO of Canonical. “Charmed Spark is the first of many Canonical open source data solutions designed for reliability and multi-cloud operation. Every production deployment is warranted for ten years compliance and security maintenance”.
Charmed Spark offers support for Apache Spark 3, along with its enhanced Python integration and richer Spark-SQL feature set, according to Canonical.
The solution also comes with spark8t Python SDK and command line tooling for streamlined workflows with Spark on Kubernetes. The Spark container image—built on Ubuntu 22.04 LTS—enables users to leverage an out-of-the-box runtime for producing Spark applications that run on Kubernetes clusters.
Charmed Spark’s capabilities are further amplified by the spark-history-server-server-k8s operator, which allows administrators to rapidly deploy, configure, and operate the Spark History Server on a Kubernetes cluster via Juju.
Also included within Charmed Spark is the Charmed Spark OCI image, backed by Ubuntu Pro enterprise support and security maintenance subscription.
Customers can purchase 24/7 or weekday enterprise support on a per-node basis via the Ubuntu Pro + Support plan—which covers the full span of applications in the Canonical Data Fabric suite of data processing solutions.
To learn more about Charmed Spark, please visit https://canonical.com/.