The pieces of IBM Corp.’s plan to woo developers came together this morning with the release of a major upgrade to its public cloud that brings four new services for automating how information is handled inside web applications. Headlining the rollout is a repackaged version of the managed database platform that the company obtained through its acquisition of Compose Inc. last year.

The startup had carved out a respectable niche for itself helping developers provision open-source systems such as MongoDB and PostgresQL that often take a significant amount of time to set up manually. Its service, which is now offered under the name IBM Compose Enterprise, can launch an instance on the technology giant’s infrastructure in a matter of minutes and automatically scale the implementation as the volume of information inside increases over time. In the background, every new record added to the deployment is synchronized to a backup repository so as to protect against the possibility of outages.

The value proposition should prove particularly appealing to cloud startups that don’t have the capital to spare for a dedicated database administrator. However, the current roster of supported systems in Compose Enterprise doesn’t cover some of the more advanced use cases that the web-scale crowd is pursuing. Chief among them is graph analysis, which IBM is addressing with a separate managed data management service introduced as part of today’s rollout that is in turn based on the open-source TinkerPop project. It’s specifically geared toward handling unstructured information like social media posts and online buying data that contains complicated patterns.

Once they have the groundwork in place, a developer building their application on the company’s public cloud can start building analytic features like recommendation engines using IBM Predictive Analytics, the third service that rolled out this morning. It’s touted as a relatively simple option for creating machine learning models that doesn’t require knowledge of data science. An algorithm produced in the tool can be trained against records from a new information exchange the company is unveiling in conjunction that contains more than 150 free datasets.

Accompanying the launch are new versions of 15 existing services that have been rewritten to use Spark. The move is part of the $1 billion effort that IBM kicked off last year to boost the adoption of the speedy data crunching framework among traditional enterprises. The initiative previously saw the company contribute its homegrown SystemML library to the upstream community in order to facilitate better support for machine learning use cases.

Image via Geralt