Overview of IBM Spectrum Conductor Deep Learning Impact 1.2.2

IBM Spectrum Conductor Deep Learning Impact 1.2.2 is an add-on to IBM Spectrum Conductor 2.3.0 that provides deep learning capabilities to your IBM Spectrum Conductor environment.

IBM Spectrum Conductor Deep Learning Impact provides robust, end-to-end workflow support for deep learning application logic. This includes the complete lifecycle management from installation and configuration, data ingest and preparation, building, optimizing, and training and testing the model.

IBM Spectrum Conductor Deep Learning Impact is built on IBM Spectrum Conductor, a highly available and resilient multitenant distributed framework, providing Apache Spark and deep learning application lifecycle support, centralized management and monitoring, end-to-end security, and support from IBM.

Figure 1. IBM Spectrum Conductor Deep Learning Impact architecture overview

IBM Spectrum Conductor Deep Learning Impact supports the TensorFlow, PyTorch, and Caffe deep learning frameworks. Supported frameworks, like TensorFlow and Caffe which are installed during installation have full use of cluster resources and the cluster management console.

IBM Spectrum Conductor Deep Learning Impact also provides the capability to bring any framework and run it on the cluster from a command line interface. Learn more about this capability, see Using frameworks via command line interface.

IBM Spectrum Conductor Deep Learning Impact also utilizes IBM Spectrum Scale or NFS for shared storage of deep learning working user data.