Hewlett Packard Enterprise (HPE) will buy BlueData in a move to expand its artificial intelligence (AI) and big data offerings. The companies didn’t disclose the terms of the deal, which is expected to close by Jan. 31, 2019.

Founded in 2012, BlueData makes a software platform that uses Docker containers to make it easier for companies to deploy large-scale machine learning and big data analytics environments. It’s based in Santa Clara, California, and has raised $39 million in three funding rounds.

HPE says BlueData’s platform will complement its high-density Apollo servers, which are designed for big-data workloads, and related professional services. The two companies first partnered in 2016 to run BlueData software on top of HPE’s infrastructure.

The companies announced the acquisition today at HPE Discover in Madrid, which is happening this week.

In other news out of HPE Discover in Madrid, HPE rolled out four edge products as part of CEO Antonio Neri’s pledge to invest $4 billion in edge technologies and services over the next four years.

This includes the Edgeline OT Link Platform, an open platform that automates the interplay between operational technologies (OT) and IT-based applications at the edge. With this product, “HPE is not in the business of an IoT software platform,” said Dr. Tom Bradicich, VP & GM of servers and IoT systems at HPE. “The software allows a graphical drop and drag GUI interface to literally connect IoT systems at the edge.”

The second product is Edgeline systems management software designed for edge locations. Bradicich said it’s “edge optimized” to run in remote locations like oil rigs with intermittent connectivity and ensure enterprise-grade reliability and security.

Next up is Edgeline EL300 Converged Edge System. This is a converged system that combines OT link modules — these are HPE-developed adapters that connect to range of HPE and third-party OT systems. OT link will also integrate field programmable gate array (FPGA) modules so customers can connect to any industrial input/output device. This system also runs Edgeline systems management software.

And finally, new Edgeline Field Application Engineering Services are available from HPE Pointnext to help customers plan, build, and customize OT link-based IoT and cyber-physical systems. Pointnext is HPE’s technology services division.

Also today, HPE announced advances in what it calls “intelligent storage.” The vendor will soon offer storage built with storage class memory (SCM) and NVMe in its Nimble and 3PAR storage systems. HPE says this will lower latency up to 2X and is up to 50 percent faster than all-flash arrays with NVMe solid state drives.

It’s also building intelligence into these new storage systems to optimize SCM to enable real-time processing for latency-sensitive applications and mixed workloads – like AI and machine learning or high-speed transactions – at scale.

This will be available December for 3PAR and is expected in 2019 for Nimble Storage.

The company also expanded its AI management tool InfoSight into workload management by adding machine-learning capabilities. Cross-stack recommendations for HPE Nimble Storage extend the InfoSight recommendation engine to the virtualization layer. This moves it beyond the basic predictive analytics tools to provide AI-driven guidance across the infrastructure stack regarding how best to optimize customers’ environments and where to place their data.

Additionally, an AI-driven resource planner helps optimize workload placement based on available resources.

And new performance insights for HPE 3PAR storage reduces time spent troubleshooting by using machine-learning to self-diagnose performance bottlenecks. This service is delivered on-premises, and thus extends HPE InfoSight to sites that have restricted access to the cloud.

HPE also announced a partnership with hyperconverged secondary storage vendor Cohesity. The original equipment manufacturer (OEM) partnership will see Cohesity software running on HPE Apollo and ProLiant servers.

This aims to address the problem of secondary data sprawl by consolidating all secondary data and apps on a software-defined, web-scale platform that extends from on-premises infrastructure to the cloud and remote or branch offices.