Migrating SAP Environments from other cloud providers

[GS03i]

Hi, I've got several customers trying to move their SAP workloads to the Cloud, all SAP blogs recommend migrating by installing new sap instances and then exporting data from current environment and importing it into new instances, is there any easiest and faster way to achieve this? is it posible to use azure migrate tool to migrate SAP Workloads to the cloud? ( ive been told it is not recomended to use AMT (Azure Migrate Tool) because it wont check DB consistency at target VM....

[#2]

Hi @GS03i,

This is a tricky question and I would love to ear the guys from Microsoft about it to...

:)

Pedro

[#3 · Accepted solution]

@GS03i While technically possible to do so for the simpler landscapes SAPs guidance has been to build and copy. For production landscapes thats definitely the way to go. But for smaller deployment and cases where the DB is not running an HA configuration you could use tools like Azure Migrate: Server Migration as well

[#4]

HA is not provided in the same way in Cloud and OnPrem environment, so for production SAP system that are mostly HA; you have to go through a re-install + export/import or replication.

Regarding DB integrity, if you can stop the DB, you can run checksum on source and target to compare if database files have been corrupted.

However, ASR like AzureMigrate have a lot of VM requirements and you should dig more in details that part before using them.

[#5]

@bsiva thx 4 ur reply, just to be sure I get it right, if for example i have a simple landscape including Prod, QA, DEV. Where APP and DB are installed on same server with no HA or Replication. I could use Azure Migrate: server migration...am i right?

[#6]

@GS03i Yes. That is correct. If you use Azure Migrate's agentless migration option for VMware or Hyper-V VMs, the tool shuts the VM down and performs one final sync before migrating which ensures that the DB will be consistent. If using the agent-based approach, you can stop the database prior to the migration and wait for a subsequent data sync to ensure the DB is kept consistent.