Deployment

• Sparkling Water Backends
  • Internal Backend
  • External Backend
• Importing H2O MOJOs from H2O-3
  • Starting a Scoring Environment
  • Loading and Usage of H2O-3 MOJO Model
  • Exporting the loaded MOJO model using Sparkling Water
  • Importing the previously exported MOJO model from Sparkling Water
  • Accessing additional prediction details
  • Customizing the MOJO Settings
  • Methods available on MOJO Model
• Importing MOJO Pipelines from Driverless AI
  • Requirements
  • Starting a Scoring Environment
  • Loading and Usage of Driverless AI MOJO Model
  • Predictions Format
  • Customizing the MOJO Settings
  • Troubleshooting
• Deploying PySparkling Pipeline Models

Amazon EMR

• Use Sparkling Water with Amazon EMR
• Use Sparkling Water with Amazon EMR from the Edge Node
• Start Sparkling Water on Amazon EMR using our Terraform Template

Kubernetes

• Running Sparkling Water in Kubernetes
  • Internal Backend
  • Manual Mode of External Backend
  • Automatic Mode of External Backend

Databricks

• Running Sparkling Water on Databricks Azure Cluster
• Running RSparkling on Databricks Azure Cluster
• Running PySparkling on Databricks Azure Cluster
• Download H2O Logs from Databricks Cloud

Google Cloud Dataproc

• Running Sparkling Water on Google Cloud Dataproc
  • Sample Script for Sparkling Water Job

HD Insight

• Using Sparkling Water with Microsoft Azure HDInsight - Beta
  • H2O Artificial Intelligence for Azure HDInsight

Windows

• Use Sparkling Water in Windows Environments
• Use RSparkling in Windows Environments
  • Prepare Spark Environment
  • Prepare R Environment
  • Test the Functionality
  • Troubleshooting
  • References

Zeppelin

• Sparkling Water and Zeppelin
  • Launch Zeppelin with Sparkling Water
  • Using Zeppelin