A collaborative data science experience optimized for HPE Ezmeral ML Ops

Hewlett Packard Enterprise and Dataiku accelerate data science within enterprises

+ show more
Solution brief

As organizations expand their data-driven initiatives, the demands on data teams increase. Data scientists require access to massive amounts of data, a variety of analytic tools and functions, and multiple different data processing and model training engines to unlock the value of their data.

  • In a dynamic business environment where speed is everything, business and data analysts need access to curated datasets and the ability to rapidly experiment with a variety of tools and model development frameworks. Enterprises need to ensure minimal ramp-up time by providing their data analytics teams with intuitive interfaces to analyze and interpret the data and test hypotheses.

    On the other hand, ML architects and data engineers need access to enterprise-grade tools to gather, store, protect, and make the data available for their data science teams. IT needs to ensure that all of the different data users have access to the right tool for every job in the most efficient and cost-effective way possible while working within shrinking or static IT budgets.


    To address these challenges Hewlett Packard Enterprise and Dataiku are teaming up to deliver an enterprise-ready software solution that accelerates time to value of data science projects by bringing speed and agility to the model development process.


    The solution is built on two top-level data science products—Dataiku DSS, a collaborative visual data science tool, and HPE Ezmeral ML Ops—a container-based software platform that simplifies development, deployment, and management of machine learning models and brings DevOps agility and speed to the data science lifecycle.


Figure 1. Dataiku DSS optimized for HPE Ezmeral ML Ops

  • Dataiku DSS
    • Data connectivity: Seamless connectivity to any data, no matter where it’s stored or in what format.
    • Speed: Faster data cleaning, wrangling, mining, and visualization.
    • Accelerated data science: The latest in machine learning technology (including AutoML and deep learning) all in one place and ready to be operationalized with automation environments, scenarios, and advanced monitoring.
    • Flexible interface: Every step in the data‑to-insights process can be done in code or with a visual interface.
    • Security: Enterprise-level security with fine‑grained access rights.


  • HPE Ezmeral ML Ops
    • Rapid provisioning of Dataiku DSS environments: Data science teams can quickly spin up environments with the right infrastructure for their use case.
    • Scalable training environments with secure access to Big Data: Scalable environments for development and test or production workloads with secure access to shared enterprise data sources on-premises or in cloud-based storage.
    • Flexible, scalable, endpoint deployment: HPE Ezmeral ML Ops deploys the model’s native runtime image into a secure, highly available, load-balanced, and containerized endpoint. An integrated model registry enables version tracking and seamless updates to models in production.
    • End-to-end visibility across the ML pipeline: Complete visibility into runtime resource usage such as GPU, CPU, and memory utilization. Ability to track, measure, and report model performance along with third-party integrations track accuracy and interpretability.
    • Security and control: HPE Ezmeral ML Ops software provides multitenancy and data isolation to ensure logical separation between different Dataiku users. HPE Ezmeral ML Ops integrates with enterprise security and authentication mechanisms such as LDAP, Active Directory, and Kerberos.
    • Hybrid deployment: With HPE Ezmeral ML Ops, you can run Dataiku DSS on-premises, or in any public cloud.

    The joint solution of HPE Ezmeral ML Ops and Dataiku addresses these challenges by empowering everyone on the data team to:

    • Analyze anything—using their preferred analytic tools and engines
    • Empower anyone—using their preferred interface from notebooks and scripts to visual data pipelines and dashboards
    • Deploy anywhere—on public clouds, private clouds, on-premises, or in hybrid environments


    HPE Ezmeral ML Ops simplifies deployment of Dataiku by providing rapid access to compute resources and data for ML model training. Once the model is built, HPE Ezmeral ML Ops provides scalable endpoint deployment of native model runtime.

Learn more at


Download the PDF

Active Directory is either a registered trademark or trademark of Microsoft Corporation in the United States and/or other countries. All third-party marks are property of their respective owners.