Understanding the HCS nondisruptive migration workflow

Nondisruptive Migration User Guide

Version
9.6.0
Audience
anonymous
Part Number
MK-92RD8086-11

The HCS nondisruptive migration workflow consists of two parts: The primary workflow is strictly for data migration, and the secondary workflow is for the migration of copy pair configurations. The following diagram shows the order of the tasks in the migration process. Each task corresponds to a selection in the HCS nondisruptive migration workflow GUI.

Note: The copy pair migration workflow can be started any time after Step 3: Unallocate Source Volumes in the primary (data migration) workflow, but it must be completed before proceeding to Step 5: Unvirtualize Source Volumes. In the interests of data protection, it is ideal to complete the workflow prior to starting Step 4: Nondisruptive Migration.

Data migration workflow

  1. Virtualize and Allocate Volumes

    As shown in the following diagram, after a virtual storage machine is created on the target storage system, the source volume is virtualized as the target volume and allocated to the same host as the source volume. This ensures that the LUN path between the source storage system and the host is duplicated on the target side.

  2. Change Host I/O Paths

    The HCS nondisruptive migration workflow prompts you to perform the following operations manually:

    • Initiate I/O between the target storage system and the host.
    • Disable I/O between the source storage system and the host.

    You must do this using path management software, OS native multipath functions, or by changing the zoning configuration. When you confirm that the switch was successful, the I/O path is changed as shown in the figure.

  3. Unallocate Source Volumes

    To prevent the host from accessing the source volume through the source storage system, the LUN path must be deleted between the source volumes and the host. After you confirm the path deletion in the GUI, HCS changes the cache mode of the target volume from cache through to write sync.

  4. Nondisruptive Migration

    In this step, the data is copied to its final destination on the target storage system. When the migration is finished, the LDEV ID and host I/O associated with the target volume are swapped with those of the target volume for data migration, as shown in the following figure.

  5. Unvirtualize Source Volumes

    When the migration is complete, you must confirm that the source volumes on the target storage system are ready to be unvirtualized, where the LUN paths between the source volume and target storage system are deleted. (The deletion of the actual source volumes is not part of the nondisruptive migration workflow but is part of the process described in Completing the migration.)

Copy pair migration workflow

The following steps comprise the secondary workflow for migrating existing ShadowImage copy pair configurations.

  1. Prepare Target Secondary Volumes

    To duplicate the copy pair configurations of the source storage system, the GUI directs you to create the (ShadowImage) secondary volumes on the target storage system. Virtual IDs are assigned to the volumes, and LUN paths are configured to ports according to the mapping defined in the migration project.

  2. Allocate Secondary Volumes

    To ensure that the backup server can access the secondary volumes on the target storage system (rather than those on the source storage system), you must delete the backup server WWN from the host group of the source storage system, and then add the backup server WWN to the host group of the target storage system.

  3. Migrate Config Files

    The configuration definition files available for migration are listed for selection in the GUI for use on the target storage system.

    Note: Repeat this step for all the configuration definition files on each pair management server if the pair management servers are in a redundant configuration. See Migration of existing ShadowImage copy pairs for a list of configuration definition file requirements.
  4. Create ShadowImage Pairs

    Using the Change Pair Status wizard, you create the necessary ShadowImage pairs in the target storage system using the migrated configuration definition files.