> Software > Manage > Storage > HyperScale > HyperScale 1.5 > Reference Architecture > Expansion

HyperScale Reference Architecture - Adding Nodes Using Cross Mount

To scale out beyond 144 drives or 6 nodes per block, the cross mount feature should be used for expansion. Cross mounting ensures load balance between multiple mount paths using the built-in spill and fill capabilities of library mount paths.

Note

This procedure applies to HyperScale Reference Architecture 1.5. (Does not apply to HyperScale X Reference Architecture.)

Before You Begin

  • Perform the checks listed in the Before You Begin section in Setting Up HyperScale Storage Nodes.

  • Install the latest Maintenance Release version on the CommServe server and MediaAgents in the existing storage pools.

    For more information on the latest Maintenance Release version, see What's New | Feature Release 11.x | Maintenance Releases in Feature Release 11.X | List of Maintenance Releases for Feature Release 11.X.

    For more information on verifying the version information on the CommServe and MediaAgents , see Data Views - Version and Service Pack.

  • Login to the CommServe and run the following Qscript to enable this feature:

    qoperation execscript -sn setConfigParam -si MMCONFIG_WEBSCALE_CROSS_MOUNT -si 1 -si authentication_code

    Where:

    <authentication_code> is an authorization code needed to enable this feature. Contact Commvault Customer Support to obtain the authorization code.

Procedure

  1. Depending on your environment, setup additional nodes as described in one of procedures in Setting Up HyperScale Storage Nodes.

    Important

    Verify and ensure that all nodes have the same time and timezone settings.

  2. After all the nodes are setup, add the nodes to the existing HyperScale Storage Pool in the CommServe as follows:

    1. Using a web browser, launch the Command Center using the following URL:

      https://<Commserver hostname>/commandcenter

    2. Type the user name and password provided during the setup.

    3. Click Login.

    4. From the navigation, click Storage > Storage Pools.

      The Storage Pool page appears.

    5. From the Storage Pools page, click the name of the Storage Pool to which you want to the additional nodes.

    6. Click Add nodes.

      The Add nodes page appears.

    7. From the Nodes list, select the number of nodes which you want to add.

      Note

      The number of nodes that you can add to a Storage Pool depends on the Resiliency of the Storage Pool.

    8. Click Add.

      This will add the new nodes to the storage pool.

  3. Install the latest operating system updates on the newly added nodes as described in Installing Operating System Updates on Existing Nodes.

    Note

    Setup the remote cache on one of the newly added nodes. (This is in addition to the remote cache available in the existing nodes described in step 1 of Installing Operating System Updates on Existing Nodes.

Additional Notes

In a 3 node setup, the Deduplication Database (DDB) will be distributed as follows before the expansion:

Node 1

Node 2

Node 3

Partition 1

Partition 2

Partition 3

Partition 4

Partition 5

Partition 6

When the storage pool is expanded to 6 control nodes, the DDB will be automatically re-distributed across the MediaAgents as follows:

Note

As a result, jobs accessing the DDB will be automatically suspended during the process and (automatically) resumed once the DDB partitions are successfully re-distributed across all the partitions. (If there is no data in the DDB (no backups), then a 'configuration-only' move will be initiated.)

Node 1

Node 2

Node 3

Node 4

Node 5

Node 6

Partition 1

Partition 3

Partition 5

Partition 2

Partition 4

Partition 6

DDB partitions will be moved to the new node(s) when they are configured as a Control node until all the Control nodes contain only one partition in each node.

×

Loading...