page-brochureware.php

IBM PowerAI developer portal

Learn about deep learning and PowerAI. Create something amazing.

IBM PowerAI Enterprise releases

IBM PowerAI Enterprise 1.1.2

Released on 11/16/2018

PowerAI Enterprise is an enterprise-class software solution for quickly and easily building an end-to-end deep learning environment for your organization. PowerAI Enterprise 1.1.2 includes all of the open source deep learning frameworks, libraries, and tools included in PowerAI 1.5.4 along with IBM Spectrum Conductor version 2.3.0, and IBM Spectrum Conductor Deep Learning Impact version 1.2.1 to give data scientists everything they need to build a distributed deep learning environment in hours rather than days or weeks—and to easily manage it as the environment grows.

Go to the IBM Knowledge Center for information about software packages and prerequisites, installing, configuring, and managing PowerAI Enterprise 1.1.2.

What’s new in PowerAI Enterprise 1.1.2

PowerAI Enterprise 1.1.2 builds upon the previous release and includes the following updates and new features:

  • All of the updates and features included in PowerAI Release 1.5.4
  • IBM Spectrum Conductor 2.30, which includes:
    Apache Spark version 2.3.1 is now publicly available and is supported by IBM Spectrum Conductor 2.3.0. The following Apache Spark versions are now built in with IBM Spectrum Conductor 2.3.0:
    • Apache Spark 2.3.1
    • Apache Spark 2.2.0
    • Apache Spark 2.1.1
    • Apache Spark 1.6.1

    See the “What’s new in IBM Spectrum Conductor ” topic in the IBM Knowledge Center to learn more.

  • IBM Spectrum Conductor Deep Learning Impact updated to version 1.2.1, which includes:
    • Features and functionality
      • Python 3 for TensorFlow
      • Hyperparameter optimization enhancements
    • Deep learning frameworks
      IBM Spectrum Conductor Deep Learning Impact 1.2.1 for Linux for POWER LE (64-bit) now supports TensorFlow 1.12 and PyTorch 1.0. TensorFlow 1.10 is no longer supported. Caffe 1.0 and IBM Caffe 1.0 continue to be supported.
    • Directories and files
      The server_dlrest.xml file now resides in $EGO_CONFDIR/../../dli/conf/dlpd. Previously this file resided in $EGO_CONFDIR../../wlp/usr/servers/dlrest/<version>
    • Environment variables
      PYTHON_VERSION is now used in Spark instance group creation to specify Python version.
    • Terminology
      dli-sig-template is now named dli-sig-template-2-2-0

    See the “What’s new in IBM Spectrum Conductor Deep Learning Impact” topic in the IBM Knowledge Center to learn more.

  • Integration with IBM Cloud Private. PowerAI Enterprise can be deployed in Docker containers in a Kubernetes-managed cluster of Power Systems servers with GPUs. To deploy your containers, you need to use IBM Cloud Private, version 3.1 to install PowerAI Enterprise. Learn about Installing PowerAI Enterprise on IBM Cloud Private in the Knowledge Center.
  • IBM PowerAI Enterprise Evaluation version. You can now try an evaluation version of IBM PowerAI Enterprise 1.1.2 for a limited time. Try it now .

Key features of PowerAI Enterprise

  • Distributed deep learning architecture that simplifies the process of training deep learning models across a cluster for faster time to results.
  • Large model support helps increase the amount of memory available for deep learning models up to 16 GB or 32 GB per network layer, enabling more complex models with larger, more high-resolution data inputs.
  • Enhanced data ingest, preparation, and transformation tools, using Apache Spark for data management. IBM® Power® Systems servers architected for artificial intelligence (AI) applications, incorporating high-bandwidth and low-latency NVLink connections between GPU accelerators for peer-to-peer communications, and directly connecting GPU accelerators to system CPUs (and system memory).
  • Powerful model development tools, including real-time training visualization and runtime monitoring of accuracy and hyper-parameter search and optimization, for faster model development.
  • Ready-to-use deep learning frameworks (TensorFlow, Caffe-BVLC, and IBM Caffe) are included.
  • Multitenant architecture designed to run deep learning, high-performance analytics, and other long-running services and frameworks on shared resources.

How to get PowerAI Enterprise 1.1.2

There are several ways to get PowerAI Enterprise 1.1.2:

Learn more

Previous PowerAI Enterprise releases

We recommend that you install the most current release of PowerAI Enterprise, however, if you have an earlier version installed, here are the release notes for your reference:

IBM PowerAI Enterprise 1.1.1

Released on: 09/14/2018

PowerAI Enterprise is an enterprise-class software solution for quickly and easily building an end-to-end deep learning environment for your organization. PowerAI Enterprise 1.1.1 includes all of the open source deep learning frameworks, libraries, and tools included in PowerAI 1.5.3 along with IBM Spectrum Conductor version 2.3.0, and IBM Spectrum Conductor Deep Learning Impact version 1.2.0 to give data scientists everything they need to build a distributed deep learning environment in hours rather than days or weeks—and to easily manage it as the environment grows.

Go to the IBM Knowledge Center for information about software packages and prerequisites, installing, configuring, and managing PowerAI Enterprise 1.1.1.

What’s new in PowerAI Enterprise 1.1.1

PowerAI Enterprise 1.1.1 builds upon the previous release and includes the following updates and new features:

  • All of the updates and features included in PowerAI Release 1.5.3
  • IBM Spectrum Conductor updated to version 2.30, which includes:
    • Product rebranding: New for 2.3.0, the product name has changed from IBM Spectrum Conductor with Spark to IBM Spectrum Conductor.
    • Apache Spark version 2.3.1 is now publicly available and is supported by IBM Spectrum Conductor 2.3.0. The following Apache Spark versions are now built in with IBM Spectrum Conductor 2.3.0:
      • Apache Spark 2.3.1
      • Apache Spark 2.2.0
      • Apache Spark 2.1.1
      • Apache Spark 1.6.1
    See the “What’s new in IBM Spectrum Conductor ” topic in the IBM Knowledge Center to learn more.
  • IBM Spectrum Conductor Deep Learning Impact updated to version 1.2.0, which includes:
    • Features and functionality
    • Deep learning frameworks
    • Directories and files
    • Environment variables
    • Terminology
    See the “What’s new in IBM Spectrum Conductor Deep Learning Impact” topic in the IBM Knowledge Center to learn more.
  • IBM PowerAI Enterprise Evaluation version. You can now try an evaluation version of IBM PowerAI Enterprise 1.1.1 for a limited time. Learn more.

Key features of PowerAI Enterprise

  • Distributed deep learning architecture that simplifies the process of training deep learning models across a cluster for faster time to results.
  • Large model support helps increase the amount of memory available for deep learning models up to 16 GB or 32 GB per network layer, enabling more complex models with larger, more high-resolution data inputs.
  • Enhanced data ingest, preparation, and transformation tools, using Apache Spark for data management. IBM® Power® Systems servers architected for artificial intelligence (AI) applications, incorporating high-bandwidth and low-latency NVLink connections between GPU accelerators for peer-to-peer communications, and directly connecting GPU accelerators to system CPUs (and system memory).
  • Powerful model development tools, including real-time training visualization and runtime monitoring of accuracy and hyper-parameter search and optimization, for faster model development.
  • Ready-to-use deep learning frameworks (TensorFlow, Caffe-BVLC, and IBM Caffe) are included. Multitenant architecture designed to run deep learning, high-performance analytics, and other long-running services and frameworks on shared resources.

How to get PowerAI Enterprise 1.1.1

There are several ways to get PowerAI Enterprise 1.1.1:

  • Install an evaluation version of PowerAI Enterprise to give it a try. If you don’t already have one, you’ll need to register for an IBMID to access the evaluation.
  • Order PowerAI Enterprise 1.1.1 from your IBM representative or authorized Business Partner.
    • To be contacted by an IBM sales representative please either complete this form: https://www.ibm.com/connect/ibm/us-en/
    • Or call a telesales associate within your region through the telephone numbers listed here: https://www.ibm.com/contact.

    When you are entitled to download the packages, download them from either:

Learn more

IBM PowerAI Enterprise 1.1.0

Available 06/19/2018

PowerAI Enterprise makes deep learning easier and faster for organizations. The solution provides workflow support for the deep learning lifecycle on a distributed architecture. PowerAI Enterprise enables teams to iterate through the deep learning cycle faster, training on more data to continuously improve the model.

Key features and benefits

PowerAI Enterprise provides these features and benefits:

  • Fast time to deploy a deep learning environment so clients can get to work immediately:
    • Simplified installation
    • Precompiled deep learning libraries, including all required files
  • Optimized performance so clients can capture value sooner:
    • Built for Power servers with NVLink CPUs and NVIDIA GPUs, delivering performance unattainable elsewhere
    • Hyper-parameter search and optimization
    • Distributed deep learning, taking advantage of parallel processing
    • Large Model Support for more complex models using higher resolution data
    • Elastic training to allocate the resources needed to optimize performance
  • Designed for enterprise deployments:
    • Multitenancy supporting multiple users and lines of business
    • End-to-end security, including role-based access controls
    • Centralized management and monitoring
  • IBM service and support for the whole solution, including the open source deep learning frameworks (such as TensorFlow and IBM Caffe and associated libraries)

Learn more