Moving volumes to a data reduction pool

The system supports migrating thin-provisioned volumes in standard pools to data reduction pools using volume mirroring; however, compressed volumes cannot use volume mirroring for migration. To migrate compressed volumes you must first decompress all existing compressed volumes in standard pools in the I/O group. The volumes can then be migrated to a data reduction pool using volume mirroring.

Data reduction can increase storage efficiency and performance and reduce storage costs, especially for flash storage. Data reduction reduces the amount of data that is stored on external storage systems and internal drives by reclaiming previously used storage resources that are no longer needed by host systems. To estimate potential capacity savings that data reduction technologies can provide on the system, use the Data Reduction Estimator Tool (DRET). This tool analyzes existing user workloads which are being migrated to a new system. The tool scans target workloads on all attached storage arrays, consolidates these results, and generates an estimate of potential data reduction savings for the entire system.

Go to https://www-945.ibm.com/support/fixcentral/ to find the tool and its readme. Data reduction is only supported Lenovo Storage V5030 and Lenovo Storage V5030F systems.

Note: Data Reduction Estimator Tool also provides some analysis of potential compression savings for volumes; however, it is recommended that you also use management GUI or the command-line interface to run the integrated Comprestimator Utility to gather data for potential compression savings for volumes in data reduction pools.

The system supports data reduction pools, which can contain thin-provisioned or compressed volumes. Data reduction pools also support additional capacity savings on thin-provisioned and compressed volumes by supporting data deduplication. When deduplication is specified for a thin-provisioned or compressed volume, duplicate versions of data are eliminated and not written to storage, thus saving additional capacity. Data reduction pools also contain specific volumes that track when space is freed from hosts and possible unused capacity that can be collected and reused within the storage pool. When space is freed from hosts, the process is called unmapping. Unmap is a set of SCSI commands that hosts use to indicate that allocated capacity is no longer required on a target volume. The freed space can be collected and reused on the system without the reallocation of capacity on the storage. The pool can also reclaim unused capacity in a data reduction pool and redistribute it to free extents. Reclaimable capacity is unused capacity that is created when data is overwritten, volumes are deleted, or when data is marked as unneeded by a host by using the SCSI unmap command. When a host no longer needs the data that is stored on a volume, the host system using SCSI unmap commands to release that storage from the volume. When these volumes are in data reduction pools, that space becomes reclaimable capacity and is monitored and collected and eventually redistributed back to the pool for use by the system. In the management GUI, reclaimable capacity is added to the available capacity for the data reduction pool. For standard pools, available capacity does not include any reclaimable capacity. In the command line interface, lsmdiskgrp command displays the different values that apply to data reduction and standard pools. For data reduction pools, the value for reclaimable_capacity indicates the amount of unused capacity that is available after data is reduced in the pool. Unlike with the management GUI, reclaimable_capacity is not included in the free_capacity value that is displayed in the lsmdiskgrp. Reclaimable capacity is collected as metadata and is also stored in the data reduction pool, thus using storage on the external storage system. The system periodically returns this capacity back to the pool, however, the system can use up 85% of the available logical capacity with reclaimable data, which can generate out-of-space warnings on the external storage system incorrectly. When creating data reduction pools, ensure that 15% of the total capacity that is allocated is reserved for these operations. Reclaimable capacity can be used for other volumes, which more efficiently uses existing storage resources. Monitor physical capacity of data reduction pools in the management GUI by selecting Pools > Pools. In the command-line interface, use the lsmdiskgrp command to display the physical capacity of a data reduction pool.

Support for the host SCSI unmap command is disabled by default. To enable support for a host to use SCSI unmap commands, enter the following command:
chsystem -hostunmap on
You can move thin-provisioned volumes in existing pools to data reduction pools to simplify management of reclaimed capacity. The data reduction pool tracks the unmap operations of the hosts and reallocates capacity automatically. The system supports volume mirroring to create a copy of the thin-provisioned volume in a new data reduction pool. This method creates a copy of the thin-provisioned volume in a new data reduction pool and does not disrupt host operations. For compressed volumes, volume mirroring is not supported for migration. Before you can migrate compressed volumes from a standard pool to a data reduction pool, you must decompress these volumes first. You can then use volume mirroring to create a copy of the volume and create it in a data reduction pool with compression selected as a capacity savings method. You can also select to create the new copy using deduplication.
Note: All volumes must be decompressed before the first compressed volume can be created in a data reduction pool.

Using the management GUI

To decompress volumes in standard pools in the management GUI, complete these steps:
Note: The management GUI is the preferred method of decompressing volumes.
  1. In the management GUI, select Volumes > Volumes by Pools.
  2. Select the standard pool that contains the compressed volume.
  3. Right-click the volume and select Modify Capacity Savings.
  4. On the Modify Capacity Savings page, change Compressed to Thin Provisioning.
  5. Click Modify.
To migrate the volume to a data reduction pool, complete these steps:
  1. Create a data reduction pool by completing these steps:
    1. In the management GUI, select Pools > Pools.
    2. On the Pools page, click Create.
    3. On the Create Pool page, enter a name of the pool and select Data Reduction.
    4. Click Create. Data reduction pools are created as parent pools only, not child pools.
  2. Add storage to the data reduction pool by completing these steps:
    1. In the management GUI, select Pools > Pools.
    2. Right-click the data reduction pool that you created and select Add Storage.
    3. Select from the available storage and allocate capacity to the pool. Click Assign.
  3. Create a copy of the compressed or thin-provisioned volume by completing these steps:
    1. In the management GUI, select Pools > Volumes by Pools.
    2. Select the existing pool that contains the volumes you want to migrate to the data reduction pool. Right-click the volume and select Add Volume Copy.
    3. On the Add Volume Copy page, select the data reduction pool to create a copy of the volume. Copy 1 is the original volume in the original pool, while copy 2 is created in the data reduction pool. Ensure that either thin-provisioning or compressed is selected for the capacity savings. For the option that you select, you can also select to use deduplication for the volume that you create. For example, you can create a thin-provisioned volume that also uses deduplication to remove duplicate data.
    4. Click Add.
    5. Ensure that the copies are synchronized before you proceed to the next steps. On the Volumes page, ensure that Yes is displayed for copy 2 in the Synchronized column.
      Note: The time synchronization takes to complete depends on the size of the volume and system performance. You can increase the synchronization rate by right-clicking the volume and selecting Modify Mirror Sync Rate.
  4. To complete the migration, remove the original volume copy by completing these steps:
    1. In the management GUI, select Volumes > Volumes.
    2. Right-click the volume copy and select Delete.
    3. Click Yes to confirm the deletion.

Using the command-line interface

To migrate volumes to a data reduction pool in the command-line interface, complete these steps:
  1. To create a data reduction pool, enter the following command:
    mkmdiskgrp -name pool_name -ext extent_size -mdisk mdisk_id_list -datareduction yes
    Where pool_name is the name of the pool, extent_size is the extent size of the pool, and mdisk_id_list is a list of MDisk IDs in the data reduction pool.
  2. To create copies of the volume in the data reduction pool, enter the following command:
    Compressed volume copy
    addvdiskcopy -mdiskgrp mdisk_group_name -compressed -rsize disk_size -autoexpand vdisk_name
    Thin-provisioned volume copy
    addvdiskcopy -mdiskgrp mdisk_group_name -rsize disk_size -autoexpand vdisk_name
    Thin-provisioned, deduplicated volume copy
    addvdiskcopy -mdiskgrp mdisk_group_name -rsize disk_size -autoexpand -deduplicated vdisk_name
    Compressed, deduplicated volume copy
    addvdiskcopy -mdiskgrp mdisk_group_name -compressed -rsize disk_size -autoexpand -deduplicated vdisk_name

    Where mdisk_group_name specifies the name of the data reduction pool created in #svc_icmigratedataredtoreg/d9429e136 in which the copies reside and disk_size is an integer value in megabytes (MB). The vdisk_name variable is the name of the volume that is being copied.

  3. To ensure that the copies of the volume are synchronized, enter the following command:
    lsvdisk vdisk_name
    Where vdisk_name is the name of the volume that contains the copies. In the command output, verify that the sync value is set to yes for the new volume copy, which indicates that the new volume copy is synchronized it the original copy.
  4. To complete the migration, remove the original copy of the volumes from the original pool by entering the following command:
    rmvdiskcopy -copy copy_idvdisk_name
    Where copy_id is the system-assigned identifier for the volume copy and vdisk_name is the name of the original volume.