Representing internal pool characteristics in a Data Space
The following statements express typical customer requirements for the internal pool characteristics that are shown in
Figure 111:
|
Note: Data Spaces can accommodate the requirements in the following statements.
|
▪Capacity:
– I want to calculate IBM i capacity together with other volume types.
– I want to mix different volume types (full/data-reduced, Open Systems, and IBM i) in the same pool.
– I want to estimate the Easy Tier impact on capacity with self-compressing arrays/MDisks.
▪Balanced pairs or single pools:
– I want my resources to exist in a balanced pair of pools, to distribute the processing load.
OR
– I want my resources to be consolidated in a single pool, for easier management.
▪Data Formats (CKD or FB) from multiple architectures/models (IBM Z, Open Systems, IBM i):
– I want multiple pools that clearly separate my Open Systems volumes from my IBM i volumes.
OR
– I want one FB pool or pool pair with two Data Spaces for mixed FB architectures/models: Open Systems and IBM i.
▪Performance:
– I want to assign multiple workloads (with same or different volume types) to the same pool.
– I want to specify different types of replication/mirror for different subsets of a pool.
▪Compression:
– I want a single pool whose Data Spaces define sets of compressible volumes and uncompressible volumes.
Example:
Data Spaces can model a pool scenario where highly compressible data goes to an FCM array:
– I want to define a separate Data Space (and corresponding workload) that accounts for each data reduction scheme
that exists in the pool or pool pair.
▪Portability: I want a Data Space that encapsulates aspects of my company’s approach to disaster recovery or
high-availability. That way, I can reuse the Data Space when trying solutions with different storage systems in my Storage Modeller project.
With Data Spaces, your models more accurately display customer requirements like the ones in the preceding statements. And a workload that runs against a Data Space also generates more
accurate performance assessments.