Scaling In a System - Teradata Software for Azure - 6.02

Teradata® Software for Azure Installation and Administration Guide

prodname
Teradata Software for Azure
vrm_release
6.02
category
Administration
Configuration
Installation
featnum
B035-2810-128K
Teradata Database downtime is required, as scaling in reconfigures the database and migrates the premium storage. Schedule a time that minimally impacts users.
[Base, Advanced, and Enterprise tiers only] Use this procedure to scale in a system after deployment. If you attempt to scale in under one of the following scenarios, you will receive an error:
  • Beyond the base number of nodes you originally deployed.
  • Beyond the maximum storage subpool count (8 per node).
After scaling in, you will automatically receive a Server Management alert if you provisioned more HSNs than TPA nodes. You can either keep or delete the extra HSNs.
  1. [Complete once per deployment] Create a service principal.
  2. Stop the database.
    # tpareset –x -f –y stop system
  3. Run the following command on all nodes to confirm the database is down.
    # pdestate -a
    PDE state: DOWN/HARDSTOP
  4. Display all supported node configurations for your current system.
    # tdc-scale-in -d
    No changes will be applied to the system.
  5. Run tdc-scale-in node_count [options]. where: This example shows how your current configuration will change after scaling in the system from 6 to 4 nodes:
    database001-01:~ # tdc-scale-in 4 -a -p 
    
    Current Configuration:
    ===========================================================================
     Nodes:
             Node Count:    6
    ---------------------------------------------------------------------------
     CPU(Core)/Mem(GB):
             CPUs/Node:    20             CPUs Total:   120
              Mem/Node:   144              Mem Total:   864
    ---------------------------------------------------------------------------
     AMPs/PEs:
             AMPs/Node:     8             AMPs Total:    48
              PEs/Node:     2              PEs Total:    12 
    ===========================================================================
    
    
    Current system will be scaled out by [6]:
    ===========================================================================
     Nodes:
             Node Count:    6 => 4
    ---------------------------------------------------------------------------
     CPU(Core)/Mem(GB):
              CPUs/Node:   20 == 20       CPUs Total:   120 =>  80
               Mem/Node:  144 == 144       Mem Total:   864 => 576
    ---------------------------------------------------------------------------
     AMPs/PEs:
              AMPs/Node:    8 => 12        AMPs Total:    48 == 48
               PEs/Node:    2 == 2         PEs Total:     12 =>  8
    ===========================================================================
    Continue? [yes/no] yes
  6. Type yes to continue. This process takes approximately 30 minutes to complete, after which the new configuration appears under Current Configuration.
  7. [Optional] Run the following command on all nodes to confirm the database is running.
    # pdestate –a
    PDE state: RUN/STARTED
  8. [Optional] Check the diagnostic and troubleshooting logs.