Implementing a System-Determined Block Size

During allocation, DFSMSdfp assists you to assign a block size that is optimal for the device. When you allow DFSMSdfp to calculate the block size for the data set, you are using a system-determined block size. System-determined block sizes can be calculated for both system-managed and non-system-managed primary storage, VIO, and tape data sets. Use of system-determined block size for tape data sets is discussed in Optimizing Tape Usage.

These data sets must have physical sequential, partitioned, or partitioned extended organization and fixed- or variable-length record formats. Unmovable or BDAM data sets are not supported.

Using system-determined block size provides the following benefits:

Frees data sets from device dependence
You do not need to know the track capacity to allocate efficiently.
Eases migration to new devices with different architectures
No changes to existing JCL are required when installing new devices.
Optimizes space usage
The calculated block size attempts to fully use the available track.
Improves I/O performance
Although the block size is optimized for space usage, performance benefits can be realized for sequential access.
Positions you to use sequential data striping
Striped data sets contain control information in the physical block that might make user-determined block sizes inefficient. You can let the system determine an optimal block size for these data sets.
Simplifies allocation
Eliminates the need to specify the BLKSIZE parameter.

Before beginning to implement system-determined block size, you must:

Understand the external parameters that control its usage
Determine the space savings
Understand the planning considerations for its use