Configuring the Cluster

TIBCO ComputeDB has three main components - Locator, Server, and Lead.

The Lead node embeds a Spark driver and the Server node embeds a Spark Executor. The server node also embeds a Snappy store.

TIBCO ComputeDB cluster can be started with the default configurations using script sbin/snappy-start-all.sh. This script starts up a locator, one data server, one lead node and the Hive Thrift server. However, TIBCO ComputeDB can be configured to start multiple components on different nodes.

Also, each component can be configured individually using configuration files. In this section, you can learn how the components can be individually configured and also learn about various other configurations of TIBCO ComputeDB.

The following topics are covered in this section:

• Configuration Files

  • Configuring Locators

  • Configuring Leads

  • Configuring Data Servers

  • Configuring TIBCO ComputeDB Smart Connector

  • Environment Settings

    • Hadoop Provided Settings

    • TIBCO ComputeDB Command Line Utility

  • Logging

• List of Properties

• Firewalls and Connections