Cloudera Data Science Workbench (CDSW) accelerates data science with:
- Secure self-service access to data
- On-demand compute
- Support for Python, R, and Scala
- Project dependency isolation
- Workflow automation, version control, and sharing
CDSW Release 1.1 is available today. This release includes many new features:
GPU-powered deep learning on demand. Release 1.1 extends CDSW benefits to GPU-optimized deep learning frameworks. Users can schedule and share GPU resources, such as the Amazon EC2 P2 Instance. Data scientists can train deep learning models on GPUs and deploy the model on clustered CPUs. No other data science platform combines on-premises GPUs with secure Hadoop integration.
Improved developer experience. CDSW 1.1 includes the Spark Web UI for Spark sessions. Web applications such as TensorBoard, Shiny, and Plotly appear as apps. For easy access to data in table format, CDSW offers scrollable data grids for R, Python, and Scala.
Flexible automation. With the new Jobs API, data science teams can orchestrate jobs from third-party tools. Data scientists can write jobs that work with parameters passed from the requesting system. This is an excellent way to automate analytic calculations for use in other applications.
Expanded deployment options. Security-conscious customers can deploy CDSW in offline clusters. Cloudera adds RHEL 7.3 and Oracle Enterprise Linux 7.3 to the list of supported operating systems.
CDSW Release 1.1 includes many other enhancements to security, Python support, and management.
Read about Cloudera Data Science Workbench.