Depending on your virtualization platform (VMware, Hyper-V, etc.), create new datastores with the desired size and configuration. These should be created with enough free space to accommodate the data you plan to move

For VMware: Use Storage vMotion to migrate virtual machines from the existing 190 TB datastore to the new ones. This process allows you to move VMs without downtime.
For Hyper-V: Use Live Migration to move VMs to the new datastores. Again, this can be done without service interruptions.

If you are not using a virtualization platform that supports live migrations, you might need to use data migration tools to move data between datastores. Tools like Veeam, Zerto, or similar solutions can help facilitate the transfer.

Once you have moved the data to the new datastores, you can resize the original 190 TB datastore to a smaller size to free up space. Depending on your storage system, this process may vary.

Keep an eye on the performance of the new datastores, and regularly monitor the storage usage. Adjust the sizes of the datastores as needed to accommodate future growth.

Remember that splitting a datastore is a significant operation, and there is always a risk involved. Careful planning, testing, and having backups in place are crucial to minimize the potential for data loss or service disruptions. It's advisable to involve experienced IT personnel or administrators who are familiar with your specific virtualization and storage environment.

Splitting a 190 TB Sangfor datastore with 160 TB occupied space can be a complex task, but it is possible with careful planning and execution. Here's a step-by-step guide on how to split the datastore:

    Preparation:
        Backup: Make sure you have a complete backup of your existing datastore before proceeding. This will ensure that you can restore your data in case of any issues during the splitting process.
        Free Space: Ensure that you have at least 30 TB of free space available on your storage system. This space will be required to temporarily hold the data while you split the datastore.
        Network Bandwidth: Ensure that you have sufficient network bandwidth to handle the data transfer during the splitting process. This can be a bottleneck, so make sure you have adequate bandwidth to avoid slow transfer speeds.

    Data Migration:
        Identify Split Point: Determine the point at which you want to split the datastore. This could be based on file size, file type, or any other criteria that makes sense for your data organization.
        Create New Datastore: Create a new datastore to hold the data that will be split from the existing datastore. This datastore should have sufficient capacity to accommodate the data being moved.
        Migrate Data: Use Sangfor's built-in data migration tools to migrate the data from the existing datastore to the new datastore. This process will copy the data from the source datastore to the destination datastore without affecting the data itself.

    Datastore Splitting:
        Stop Services: Stop all services that are using the existing datastore to prevent any data inconsistencies during the splitting process.
        Unlink Datastore: Unlink the existing datastore from the Sangfor cluster. This will remove the datastore from the cluster's management and allow you to modify it directly.
        Shrink Datastore: Use Sangfor's datastore management tools to shrink the existing datastore to the desired size. This will free up space on the storage system for the new datastore.
        Expand New Datastore: Use Sangfor's datastore management tools to expand the new datastore to the desired size. This will make it ready to receive the data that was migrated in the previous step.
        Link Datastores: Link the new datastore to the Sangfor cluster. This will make it available to the cluster and allow applications to access it.
        Start Services: Start all services that were stopped in step 3. This will allow applications to resume normal operation using the split datastores.

    Verification:
        Data Integrity: Verify the data integrity of both the existing datastore and the new datastore. This ensures that the data was copied correctly during the migration process.
        Application Functionality: Test the functionality of your applications to ensure that they are working correctly after the datastore split. This includes data access, performance, and any other relevant aspects.

Remember that splitting a large datastore can be a time-consuming and resource-intensive process. It is important to plan carefully and execute each step meticulously to minimize downtime and ensure data integrity. If you are not comfortable performing the splitting process yourself, it is recommended to engage a qualified Sangfor administrator or consultant to assist you.

Hello Team Members, very good suggestion from all of you.
There is a restriction that we can not make datastore more than 190 TB with all flash disks.
This was the only reason to split the datastore, as new SSDs have been installed but unable to expand existing datastore.
This issue will be resolved in the next version that is HCI 691. so this solution ll be perfect, only  we need to wait on the release date. Enjoy using sangfor HCI.