Create a New SMF Dataset

It’s not uncommon to be faced with the situation where one of our z/OS mainframe systems requires an extra SMF (System Management Facility) dataset in order to avoid losing any SMF data to be recorded.

To create a new SMF dataset we rely on the use of the DEFINE command of the Access Method Services utility. This can be done either via TSO or batch job.

The following JCL is an example of a batch job to define a new SMF dataset:


//CREATSMF EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=A
//SYSIN DD *
DEFINE CLUSTER  (NAME(SYS1.MANx)             +
VOLUME(xxxxxx)               +
NONINDEXED                   +
CYLINDERS(nnn)               +
REUSE                        +
RECORDSIZE(4086,32767)       +
SPANNED                      +
SPEED                        +
CONTROLINTERVALSIZE(nnnn)    +
SHAREOPTIONS(2))
/*

Bear in mind that the VOLUME parameter is only required if the SMF dataset you’re allocating is not SMS managed.

There are other parameters that can be used when allocating a new SMF dataset. To know more about this you can either check IBM manual ‘DFSMS Access Method Services for Catalogs (SC26-7394-xx) or via TSO command HELP DEFCL.

 

Explanation:

When defining a new SMF data set the following options need to be part of the DEFINE command:

• REUSE indicates that the dump program can clear the SMF dataset
• CONTROLINTERVALSIZE indicates the size of the SMF buffer. For the SMF data sets, you can specify any value between 0.5 (512 bytes) and 26K (26624 bytes), with certain restrictions.
• SHAREOPTIONS has two values that define the level of sharing. The cross-region value must be 2. This indicates that sharing occurs with reading and serialization occurs with writing. The cross-system value is allowed to default.
  • NONINDEXED indicates that the entries are entry-sequenced.
  • SPANNED indicates that the records can span control intervals.
  • SPEED indicates that the data set will not be preformatted by VSAM while IFASMFDP is preformatting. (If SPEED is not selected, VSAM and SMF preformat concurrently.)

 

Recommendations:

• Allocate each SMF data set as a single-extent VSAM cluster on a single volume
• Catalog it in the master catalog
• Do not specify secondary space for any of the cluster’s components
• Specify the same control interval size for all SMF data sets that a particular system will use

Note that before you can use the newly created SMF dataset, the dataset needs to be preformatted with dummy records by using the IFASMFDP program.