Add Nodes to a System | Teradata QueryGrid - Add Nodes to a System - Teradata QueryGrid

QueryGridâ„¢ Installation and User Guide - 3.06

Deployment
VantageCloud
VantageCore
Edition
Enterprise
IntelliFlex
Lake
VMware
Product
Teradata QueryGrid
Release Number
3.06
Published
December 2024
ft:locale
en-US
ft:lastEdition
2024-12-07
dita:mapPath
ndp1726122159943.ditamap
dita:ditavalPath
ft:empty
dita:id
lxg1591800469257
Product Category
Analytical Ecosystem

Before adding nodes, if the QueryGrid Manager publicAddress or the additionalPublicAddresses are not IP addresses and if the addresses cannot be resolved through DNS, verify that one of the host names or IP addresses has been added to the host files of the nodes in the data source system.

Nodes are registered to systems defined in QueryGrid Manager by installing the tdqg-node package on the node along with the tdqg-node.json configuration file. For supported platforms, you can add the nodes through the QueryGrid Deployment and Scaling Service, refer to Automatic Deployment of QueryGrid on Scalable Clusters and QueryGrid Compatibility Matrix for more information.

For platforms that do not support or are not using the QueryGrid Deployment and Scaling Service, you can perform the install automatically or manually. The following table provides guidelines on what types of nodes to add for each system type.
System Type Node Type Considerations
Teradata TPA, HSN, FICON All Teradata nodes must be added to the system. See Configuring QueryGrid for a Teradata FICON Node for instructions on adding nodes that do not have network connectivity to the QueryGrid Manager or remote systems.
Hive Master/Head, Worker/Core/Task, Edge

All worker nodes must be added to the system. Adding a subset of worker nodes is only possible if there is a specific configuration (such as using YARN queues and node labels) where Hive only uses those nodes for QueryGrid queries.

Each non-worker node must be added to the system if at least one of the following conditions is true:

  • HiveServer2 is running on the node and QueryGrid loaderfactory JARs are added at runtime using the ADD JAR command, as opposed to being added using the Hive Auxiliary JAR Path.
  • The node is a driver node.
Spark Master/Head, Worker/Core/Task, Edge

All worker nodes must be added to the system. Adding a subset of worker nodes is only possible if there is a specific configuration (such as using YARN queues and node labels) where Spark only uses those nodes for QueryGrid queries.

For each non-worker node, the node must be added to the system if at least one of the following conditions is true:

  • Spark Thrift Server is running on the node and is the intended execution mechanism (as opposed to Spark Application) for QueryGrid.
  • The node is a driver node.
Oracle QueryGrid driver Add one or more driver nodes to the system.
BigQuery QueryGrid driver Add one or more driver nodes to the system.
Generic JDBC QueryGrid driver Add one or more driver nodes to the system.
Presto Coordinator, Worker All Presto nodes must be added to the system.