Support for Apache Spark as a Service integration with Watson Machine Learning and Watson Studio is being deprecated.

This integration will be retired on June 28, 2019. We are happy to announce that this service integration has been replaced by built-in Spark environments

Benefits of Spark environments

  • Spark kernels on-demand:  Save time and energy to focus your analysis; create a Spark environment in Watson Studio and launch directly into a notebook.
  • Configurable, elastic compute: Configure your Spark environment and choose your kernel hardware configurations from Watson Studio.
  • Easily share your environment :  Your collaborators can easily use Spark environments.
  • Multiple language support :— Choose from the most popular languages for your Spark kernels (Python 3, R, Scala).

Switching to built-in Spark environments

If you currently use the Apache Spark as a Service in any of the following ways, you must switch to using built-in Spark environments:

  • Batch Deployments
  • Model Builder
  • Modeler Flows with Spark Runtime
  • Notebooks

Here are the dates you need to know

Service Retirement Announce Date: June 4, 2019

End of Support Date: June 28, 2019

For a period of 24 days after the Service Retirement Announce Date, through June 28, 2019, all existing Spark as a Service integrations with Watson Machine Learning and with Watson Studio projects will continue to be supported. After June 28, 2019, Spark as a Service will no longer be available to Watson Machine Learning or by tools in Watson Studio like Notebooks and Model Builder. Spark-powered realtime streaming predictions will no longer be available. 

We strongly recommend you update as soon as possible to work with the newer, more flexible Spark engines in Watson Machine Learning and Watson Studio.

Learn more.


More from Announcements

IBM TechXchange underscores the importance of AI skilling and partner innovation

3 min read - Generative AI and large language models are poised to impact how we all access and use information. But as organizations race to adopt these new technologies for business, it requires a global ecosystem of partners with industry expertise to identify the right enterprise use-cases for AI and the technical skills to implement the technology. During TechXchange, IBM's premier technical learning event in Las Vegas last week, IBM Partner Plus members including our Strategic Partners, resellers, software vendors, distributors and service…

Introducing Inspiring Voices, a podcast exploring the impactful journeys of great leaders

< 1 min read - Learning about other people's careers, life challenges, and successes is a true source of inspiration that can impact our own ambitions as well as life and business choices in great ways. Brought to you by the Executive Search and Integration team at IBM, the Inspiring Voices podcast will showcase great leaders, taking you inside their personal stories about life, career choices and how to make an impact. In this first episode, host David Jones, Executive Search Lead at IBM, brings…

IBM watsonx Assistant and NICE CXone combine capabilities for a new chapter in CCaaS

5 min read - In an age of instant everything, ensuring a positive customer experience has become a top priority for enterprises. When one third of customers (32%) say they will walk away from a brand they love after just one bad experience (source: PWC), organizations are now applying massive investments to this experience, particularly with their live agents and contact centers.  For many enterprises, that investment includes modernizing their call centers by moving to cloud-based Contact Center as a Service (CCaaS) platforms. CCaaS solutions…

See what’s new in SingleStoreDB with IBM 8.0

3 min read - Despite decades of progress in database systems, builders have compromised on at least one of the following: speed, reliability, or ease. They have two options: one, they could get a document database that is fast and easy, but can’t be relied on for mission-critical transactional applications. Or two, they could rely on a cloud data warehouse that is easy to set up, but only allows lagging analytics. Even then, each solution lacks something, forcing builders to deploy other databases for…