Lightbend Fast Data Platform

1.2.0, June 29, 2018


Lightbend Fast Data Platform is an integrated suite of tools and services for building and running fast data (streaming) systems, either on-premise or hosted in the cloud. Fast Data Platform delivers value three ways:

  • An accelerated on-ramp for building streaming data systems: Configuration and installation tools to get you up and running quickly.
  • Expert guidance, so you pick the streaming components best suited for your project: Bundled sample applications, documentation, and access to Lightbend’s Developer Assist provide quick, expert answers to all your questions.
  • Intelligent monitoring and management: Assure cluster reliability at scale for peace of mind.


The 1.2.0 release includes the following components, divided into supported and certified. Let’s first define terms:

  • Supported: A component that is supported is one that Lightbend has committed to fixing problems encountered within the component as per Lightbend’s standard support and maintenance policy.
  • Certified: A component that is certified has been tested by Lightbend or one of our partners for interoperability with Lightbend Fast Data Platform, but Lightbend disclaims all warranties and makes no promises about the component’s operation or production readiness. Support for the integration between a certified component and Lightbend Fast Data Platform will be provided on a “best effort” basis only. Support for the certified component itself may be available through a third-party.
  • Unsupported/Uncertified: Customers may need to run components that are unsupported and uncertified. This is allowed on Fast Data Platform and Lightbend will assist customers as is reasonable by standard Lightbend practices. However, Lightbend will not guarantee to fix platform bugs specific to an unsupported/uncertified component and Lightbend will not commit to service level agreements (SLAs) for such a component.

The 1.2.0 release includes the following supported components:

Note: See the Release Notes for more details. Not all features of all components are fully supported.

In addition, the 1.2.0 release includes installation tools in Fast Data Platform Manager for the following certified components:

See the Fast Data Platform Overview for an explanation of the terms supported, certified, and preview. In a nutshell, they are defined as follows:

Term Description
Supported Full production support
Certified Production ready (we use them ourselves), but offered without production support
Preview Possible candidate for supported or certified, but not yet considered production ready

Finally, the Next Steps section of the the Installation Guide describes the tools and instructions we provide for installing the following additional certified components, used by our Sample Applications:

Lightbend Fast Data Platform requires Mesosphere DC/OS 1.10.4 or greater. We recommend version 1.11.X, which is required if you wish to leverage Kubernetes. For production clusters, Enterprise DC/OS is required, configured in permissive mode, as opposed to strict mode. (Support for strict mode is planned.) Development and test clusters can use Opens Source DC/OS.

We picked DC/OS for its enterprise-class features combined with the modern, flexible resource management capabilities of Mesos. With DC/OS, you can install third-party components that supplement Fast Data Platform as required for your needs, including persistence stores like Cassandra and Elasticsearch. These components are available in the DC/OS Catalog (formerly known as the Universe).

Fast Data is a rapidly evolving ecosystem. To balance the desire to use the “latest and greatest” software, vs. the need for stability and reliability, Fast Data Platform introduces “previews” of some components, where they are not yet supported for production use, either because the components are immature or because the integrations with DC/OS are immature.

In this release, the following preview component is offered:

  • Hive on Spark: Based on customer feedback, Fast Data Platform 1.1.0 introduced “preview” support for Hive using Spark as the execution back end. This build is based on Hive 3.0.0.

Please provide feedback about these components, such as whether or not you intend to use them in production, as well as other components you would like to use, so we can plan accordingly to meet your needs!

About This Documentation

This documentation includes the following:

All the documentation for version 1.2.0 is distributed with Fast Data Platform and is also available at To see the documentation for the latest release, go to

The 1.2.0 release provides installation and monitoring tools for Spark, Flink, Kafka, Hive, Lightbend Enterprise Suite, and HDFS on DC/OS. The monitoring tools include the cloud-based monitoring solution from OpsClarity, which gathers operational telemetry from your cluster and gives you a global, graphical view of your cluster and its health.

Here is the release history for Fast Data Platform. More details can be found in the release notes for each release, which are the links in this table:

Release Date Highlights
V1.2.0 June 29, 2018 Production hardening, bug fixes and enhancements throughout the product
V1.1.0 February 23, 2018 Kafka 1.0, Hive on Spark Preview, general improvements
V1.0.3 January 18, 2018 Bug fixes in sample applications and various robustness improvements
V1.0.2 January 9, 2018 Bug fixes in sample applications (regressions)
V1.0.1 January 5, 2018 Improved documentation, bug fixes, refinements to sample applications
V1.0.0 October 30, 2017 General availability, production ready

Interested in trying the current release? Sign up here or contact your Lightbend account manager.