A Beta to Help You Create a Data and AI Platform on Your Terms

Steven Astorino
Inside Machine learning
3 min readJun 29, 2020

Although more organizations are learning about the value of centralizing data management across hybrid multiclouds and infusing data science and AI into the infrastructure, not everyone has been able to do it. Maybe they lack the resources or capital expenditure. Or maybe they just don’t feel ready. These are the kinds of organizations we had in mind when we set out to create an as-a-Service version of our integrated data and AI platform, Cloud Pak for Data — an offering that is currently available as beta.

I discussed Cloud Pak for Data during my Think 2020 session, and the Cloud Pak for Data as-a-Service approach which is designed to help benefit organizations in several ways:

  • Faster time to value, as organizations have access to the power of a data and AI platform but with no installation, management, or updating required.
  • Access to a catalog of predefined services1 that can be provisioned with a “point and click” approach.
  • A multitenant service managed on IBM Cloud with multicloud connection capabilities.
  • Available as a “pay-as-you-go” service or subscription model with the ability to scale up and down as needs evolve.
  • Built-in security.
  • Features to help users with compliance requirements.

Of the many attributes of the as-a-Service version, “use case” examples will help organizations understand how services can be integrated together to support specific disciplines. These are designed to help organizations get started fast. For example, a Utilities Payment Risk Prediction use case can inform the user what services and tools they will be using, such as Watson Knowledge Catalog, Watson Studio, Watson Machine Learning, and Rstudio. During a use case, users will be informed of what is happening and why, through a series of context-sensitive pop-ups. Users will also be advised when to provision and organize resources into a project in preparation for testing and deployment phases.

For those new to Cloud Pak for Data, this “as-a-service” option is designed to help users get started quickly. For existing Cloud Pak for Data customers who also choose to subscribe to this as a Service beta offering, their current environment remains the same, but they can utilize the new as-a-service platform and new user experiences to try out new services or spin up smaller projects. Watson Studio, Watson Knowledge Catalog, and Watson Machine Learning on Cloud clients using this beta keep the same core capabilities, but with the advantage of the new experience. For other Data and AI cloud services customers interested in the beta, the core capabilities stay the same and these organizations can choose to integrate their services into the new experience if desired.

This article provides a glimpse of what is intended to be in the Cloud Pak for Data as a Service offering. If this has sparked your interest, there are two options to consider:

  1. If you want to be one of the first to use the offering you can apply to join the Beta Program which your experience of using the offering and providing feedback will help drive the product forward.
  2. Apply to join the Sponsor User Program which offers participation in early design discussions and feedback through workshops, calls, etc., to influence the user experience, feature and function directions, helping shape the future of the Cloud Pak for Data as a service offering.

For more information and to keep up to date on Cloud Pak for Data click this link

– Steve

--

--

Steven Astorino
Inside Machine learning

Vice President of Development, Data and AI. Tweets and opinions are my own https://stevenastorino.com