ANF-HANA-multi-partition

A PowerShell script to automate the deployment of Azure NetApp Files application volume groups for SAP HANA using a multi-partition layout.

Description

For larger HANA systems, such as the new 24TiB VMs, a single data volume does not deliver enough performance and capacity. For this scenario, SAP supports 'multiple partitions' (MP) where the SAP HANA database is striped across multiple volumes. This script automates the deployment of multiple Azure NetApp Files application volume groups (AVGs) to support SAP HANA databases striped across one or more data volume partitions.

Instructions

There are two ways to use this script:

Specify all arguments within an external file such as 'config.ps1' and run the following command:

./ANF-HANA-multi-partition.ps1 -configFile ./config.ps1

See here for an example of the external config file.

Specify all arguments inline using Powershell parameters:

.\ANF-HANA-multi-partition.ps1 -numPartitions 4 -subnetId <subnetResourceId> -ppgId <ppgResourceId> -capacityPoolId <capacityPoolResourceId> -avgAppIdentifier "SH1" -mountPoint "-mnt00001" -deployForHSR $false -sharedVolSizeGiBs 1024 -sharedVolTPutMiBps 128 -logVolSizeGiBs 512 -logVolTPutMiBps 250 -dataVolSizeGiBs 4096 -dataVolTPutMiBps 300 -cleanUponFail $true

Output Examples

See here for an example of a successful run.

See here for an exmaple of a failed run.

Order of Operations

1. Pre-check validation
   1. verify capacity pool, proximity placement group, and delegated subnet exist
   2. verify capacity pool throughput available
   3. verify capacity pool size
2. Create base AVG which includes 'shared', 'log', and temporary 'data' volumes.
3. Create additional AVGs which include 'data' and temporary 'log' volume(s).
4. Check that all resources were created successfully.
   1. Optionally, if any errors are encountered, the script will delete any resources that it did create successfully.
5. Delete temporary 'data' and 'log' volumes.
6. Display final state of application volume groups and child volumes.

Example

In the following scenario a multi-partition configuration is created with 4 data partitions.

Initially the following AVGs and child volumes are created:

• AVG: SAP-HANA-SH1-shared-log
  • volume: SH1-shared
  • volume: SH1-log-mnt00001
  • volume: SH1-data-temp
• AVG: SAP-HANA-SH1-part1-mnt00001
  • volume: SH1-data-part1-mnt00001
  • volume: SH1-log1-mnt00001-temp
• AVG: SAP-HANA-SH1-part2-mnt00001
  • volume: SH1-data-part2-mnt00001
  • volume: SH1-log2-mnt00001-temp
• AVG: SAP-HANA-SH1-part3-mnt00001
  • volume: SH1-data-part3-mnt00001
  • volume: SH1-log3-mnt00001-temp
• AVG: SAP-HANA-SH1-part4-mnt00001
  • volume: SH1-data-part4-mnt00001
  • volume: SH1-log4-mnt00001-temp

Note: bold volumes are to be deleted.

Once the AVGs and child volumes are created successfully, the temporary volumes are deleted. The AVGs and remaining volumes are as follows:

• AVG: SAP-HANA-SH1-shared-log
  • volume: SH1-shared
  • volume: SH1-log-mnt00001
• AVG: SAP-HANA-SH1-part1-mnt00001
  • volume: SH1-data-part1-mnt00001
• AVG: SAP-HANA-SH1-part2-mnt00001
  • volume: SH1-data-part2-mnt00001
• AVG: SAP-HANA-SH1-part3-mnt00001
  • volume: SH1-data-part3-mnt00001
• AVG: SAP-HANA-SH1-part4-mnt00001
  • volume: SH1-data-part4-mnt00001