Defining large sequential data sets

Sequential data sets can become large and span many volumes of storage hardware. However, z/OS® Data Facilities Storage Management Subsystem (DFSMS) supports more than 65,535 tracks per volume of sequential data sets, which enables more data to be stored on fewer volumes. IMS supports large format data sets.

Sequential data sets are used in IMS primarily for:
  • GSAM databases
  • Sequential OSAM databases (DSORG=PS - physical sequential)
  • Long and short message queues
  • Online log data sets (OLDS)
  • External trace data sets
  • Terminal data sets with UNITYPE = DISK or SPOOL

Migration to large format data sets can be done only at data set creation time.

The following process shows one possible approach to defining large sequential data sets:

Procedure

  1. Determine which existing GSAM data sets that use the BSAM access method and which sequential OSAM data sets span more than one volume because they are larger than 65,535 tracks.
  2. Determine if you have hardware that supports more than 65,535 tracks.
  3. Add the DSNTYPE=LARGE parameter to the DD statement in the JCL that defines these data sets.
  4. Allocate the new data sets.
  5. Bring the new data sets online.
    • If you are using the new data sets for an online data set (OLDS) or a message queue, perform a cold start of IMS to bring the new data sets online.
    • If you are using the new data sets for a database, use a database reorganization process (unload and reload) to bring the new data sets online.

DD statement examples

Here are a few DD statement examples that include the DSNTYPE=LARGE statement that is required to define large format sequential data sets:
DD statement for GSAM data sets
//ODASD    DD UNIT=SYSDA,VOL=SER=LRGVS1,DISP=(,KEEP),
//         SPACE=(CYL,(4200,5)),DSN=IMSTESTL.GSAM.VARIAB01,
//         DSNTYPE=LARGE,
//         DCB=(LRECL=800,BLKSIZE=80,BUFNO=5,RECFM=FB)
DD statement for OSAM data sets
//AJOSAMDB DD  DSN=IMSTESTL.AJOSAMDB,UNIT=SYSDA,
//             DISP=(NEW,CATLG),DSNTYPE=LARGE,
//             SPACE=(CYL,(4500,100)),VOL=SER=LRGVS1
DD statement for OSAM data sets on an extended address volume (EAV)
//AJOSAMDB DD  DSN=IMSTESTL.AJOSAMDB,UNIT=SYSDA,
//             DISP=(NEW,CATLG),DSNTYPE=LARGE,
//             SPACE=(CYL,(4500,100)),VOL=SER=EAV001,EATTR=OPT
Restriction: EAVs are available only in z/OS V1.12 and later.
DD statement to allocate a large spool data set
//INIT01   EXEC PGM=IEBGENER  
//SYSUT2   DD  DSN=IMSTESTL.IMS01.SPOOL1,UNIT=SYSDA,  
//         VOL=SER=DSHR10,SPACE=(TRK,(66000)),DISP=(,CATLG),  
//         DCB=(RECFM=VBM,LRECL=137,BLKSIZE=27998),  
//         DSNTYPE=LARGE
DD statement to allocate a spool data set on an EAV volume
//INIT01   EXEC PGM=IEBGENER  
//SYSUT2   DD  DSN=IMSTESTL.IMS01.SPOOL2,UNIT=SYSDA,  
//         VOL=SER=EAV001,SPACE=(CYL,(25)),DISP=(,CATLG),  
//         DCB=(RECFM=VBM,LRECL=137,BLKSIZE=27998),  
//         EATTR=OPT