Scaling Out a System - Teradata Software for Azure

Teradata Vantage™ on Azure (DIY) Installation and Administration Guide

Product

Teradata Vantage on Azure

Release Number

5.01

Published

July 2018

Language

English (United States)

Last Update

2018-07-18

dita:mapPath

kmk1523992471627.ditamap

dita:ditavalPath

TeradataAzure_PubCloud_5.01_5.01.01.ditaval

dita:id

B035-2810

lifecycle

Product Category

Cloud

Teradata Database downtime is required because scaling out reconfigures the database and migrates the premium storage. Schedule a time that minimally impacts users.
Increase the Azure service limits.
Before scaling out a system, ensure you have enough IP addresses. Each new VM requires 1 IP address from each subnet. A total of 3 IP addresses per VM are required. The available IP addresses per subnet in a VNet are in Virtual network > Subnets.
For BYOL, make sure you have enough TCores before you scale out the system.

Use this procedure for any of the following:

If you did not initially scale out a system during deployment
If you previously scaled out a system during deployment and want to scale out again after deployment

It takes about 20 minutes for a 2-node system and 30 minutes for a 4-node system, depending on the number of premium storage volumes and Azure provisioning time.

Provide service principal details and enter responses to the prompts. See Azure Documentation Center for how to create the service principal.
1. Run tdc-feature-init without arguments.
```
# tdc-feature-init
```
2. Enter the client ID of your service principal.
3. Enter the secret key of your service principal.
4. Enter the tenant of your service principal.
5. Enter your Azure subscription ID.
Credentials validation succeeded displays. Service principal details are written to /etc/opt/teradata/tdconfig/service_principal_config.json.
[First time you scale out] Check if your system can be scaled out:
```
# tdc-unfold -d
```

Stop the database.

# tpareset –x –y stop for scaling out

Verify the database is in a DOWN/HARDSTOP state.
```
# pdestate -a
```
PDE state: DOWN/HARDSTOP Putting the database in this state may take several minutes.

Enter one of the following commands to start the scale out process.

Command	Description
# tdc-unfold 2x	Doubles the current node count. For example, if the current system has 4 nodes, the scaled out system will have 8 nodes.
# tdc-unfold 4x	Quadruples the current node count. For example, if the current system has 4 nodes, the scaled out system will have 16 nodes.

The output displays how the configuration changes after scaling out the system, as shown in the following example:

Current system will be unfolded by [2x]:
===========================================================================
 Nodes:
         Node Count:    2 => 4
---------------------------------------------------------------------------
 CPU(Core)/Mem(GB):
          CPUs/Node:   16 == 16       CPUs Total:   32 => 64
           Mem/Node:   65 == 65        Mem Total:  130 => 260
---------------------------------------------------------------------------
 AMPs/PEs:
          AMPs/Node:   20 => 10       AMPs Total:   40 == 40
           PEs/Node:    2 == 2         PEs Total:    4 => 8
===========================================================================
Continue? [yes/no] yes

Enter yes. When the process completes, the new configuration appears under Current Configuration.
[Optional] Check the status of the database.
```
# pdestate –a
```
[Optional] Check the logs for troubleshooting.