Configure a DS8000 storage system
for an
IBM
Z
host by
creating count key data (CKD) pools, logical subsystems (LSSs), and volumes. After you complete
these steps, configure the host to connect to the volumes.
About this task
Note: LSS is synonymous with logical control unit (LCU) and subsystem identification (SSID).
Procedure
- Create a pair of CKD pools, one for each processor node,
by using the Create Pool Pair action in the Arrays
by Pool page.
The DS8000
Storage Management GUI automatically
creates a pair of pools, which balances the workload between the two
nodes.
- In the Create Pool Pair window, select
the number of arrays that you want to assign to the pool pair, according
to your performance needs as determined in the storage planning process.
If you have multiple drive classes on your storage system, decide
how many of each drive class you want in each pool. An array consists
of eight drives. The DS8000
Storage Management GUI configures
the selected arrays for CKD storage and distributes them evenly between
the two pools.
Note: If you prefer to balance arrays manually,
you can use the Assign action in the Arrays
by Pool page.
-
Create CKD volumes by using the Create Volumes action in either the
Volumes page or the Volumes by Pool page. Use the quick
volume creation option to conveniently create volumes with standard provisioning and to create
aliases. Use the custom volume creation option to create volumes with thin provisioning and specify
logical subsystems (LSSs). For more information, see Creating volumes for IBM Z hosts by using quick creation
or Creating volumes for IBM Z hosts by using custom creation.
-
Add aliases, also known as parallel access volumes (PAVs).
PAVs enable concurrent write operations to the same volume.
- If Dynamic PAV or HyperPAV are enabled on the storage system, add aliases to the LSS.
- If Dynamic PAV and HyperPAV are not enabled, add aliases to the volumes that require them.
You can add aliases by using one of the following options:
- The quick volume creation option in Create Volumes window, which is
available from the Create Volumes action in either the
Volumes page or the Volumes by Pool page.
- The Manage Aliases action, which is available by selecting a volume on
the Volumes, Volumes by Pool,Volumes by
LSS, and Volumes by Host pages. This action is also available by
selecting an LSS on the Volumes by LSS page.
For more details about aliases and the procedures for configuring them on an
IBM Z
host, see
the online product documentation about PAVs.
- Set the I/O port protocols of the I/O ports that the host
uses to communicate with the storage system. Use the Modify
I/O Port Protocols action on the System page to set the I/O port protocol to FICON®.
-
If you are using a zHyperLink connection, use the zHyperLink tab on the
System page to enable the connection.
Note: By default, the zHyperLink facility is disabled in z/OS®. To enable z/OS for zHyperLink read or
write processing, use the SETIOS MVS™ system command after an
IPL, or use the IECIOSxx PARMLIB to enable zHyperLink processing during IPL.
- SETIOS MVS Command
- To enable zHyperLink read and write processing on the IBM
z host system after an IPL, enter the following MVS
commands:
SETIOS ZHPF=YES
SETIOS ZHYPERLINK,OPER={READ|WRITE|ALL}
For
more information about the SETIOS MVS command, refer to the
z/OS
MVS System Commands.
- IECIOSxx parmlib
- To enable zHyperLink read and write processing during an IPL, include the following statements
in the IECIOSxx PARMLIB member on the IBM z host
system:
ZHPF=YES
ZHYPERLINK,OPER={READ|WRITE|ALL}
For
more information about the IECIOSxx PARMLIB member, refer to the z/OS
MVS Initialization and Tuning Reference.
-
Configure the host so that it connects to the new CKD volumes on the storage system.
What to do next
After configuration is complete, you can migrate data from another storage system or a host
to the new storage system. For details and procedures, see the online product
documentation about migrating data.