This post was originally published on this siteThe FlashArray implementation of Virtual Volumes surfaces VMs on the FlashArray as standard volume groups. The volume group being named by ...
Today Pure announced Cloud Block Store for AWS which offers a way for the Pure Storage platforms to be extended into public cloud. There are many different use cases that Cloud Block Store can be used for and the one that I focused on was around data mobility.
The data mobility use case for this demonstration involves how to asynchronously replicate an on-premises Microsoft SQL Server 2017 database running on a Pure Storage FlashArray//M20 host in Equinix to Cloud Block Store. After replicating the database volume we bring up a development instance connected to a SQL Server instance running as an Amazon EC2 instance. The entire demonstration is automated using the Pure Storage PowerShell SDK.
The following visualization defines the individual components of the demonstration.
We have an on-prem production Microsoft SQL Server instance with a database attached. The database has active insertions running against a demonstration table named CloudBlockStore. Once the active I/O is running on the database the following steps are performed as part of the on-prem automation.
(1) Take a snapshot of the volume which hosts the AdventureWorks2017 database. This will be known as the development snapshot.
(2) After creating the snapshot, we dynamically connect the on-prem FlashArray to Cloud Block Store and configure all settings to allow asynchronous replication.
(3) A protection group is created and configured with member volumes then replicated immediately to Cloud Block Store.
(4) After the replication has been initiated a new volume is created from the development snapshot and mounted to the production SQL Server instance just to show that the local snapshot is usable. New insertions are then started on that development database.
Now that the on-prem database volume has been replicated to Cloud Block Store we can begin the restore process.
(1) Using the snapshot from the replicated protection group a new volume is created which contains the development database.
(2) Once a new Pure volume has been created we mount that to the SQL Server instance running in Amazon EC2.
(3) Just as we did with the on-prem deployment we dynamically attach the database using Invoke-SqlCommand and begin writing data to the development instance.
At this point there are many different paths that can be taken such as, (1) Perform database schema changes and replicate back to the on-prem SQL Server deployment or (2) Scale out additional development EC2 instances for additional developers or testers.
Here is the demonstration video that walks through all of the different steps.
All of the PowerShell automation scripts are available on GitHub <LINK>. The cool thing about these automation scripts is that regardless of using the FlashArray or Cloud Block Store they use all of the same PowerShell cmdlets and offer a consistent automation scripting experience. These scripts can easily be used for FlashArray to FlashArray or to Cloud Block Store.
There are many more scenarios for how Cloud Block Store can be used for data mobility, disaster recovery and development. The team is very excited for the road ahead with our cloud integrations.
Read more about Pure’s cloud vision in our announcement blog. If you are interested in signing up for the Beta
Cheers — Barkz