z/OS DFSMS Using Data Sets
Previous topic | Next topic | Contents | Contact z/OS | Library | PDF


Conversion of an Indexed Sequential Data Set

z/OS DFSMS Using Data Sets
SC23-6855-00

Important: Before migrating to z/OS V1R7 or a later release, convert your indexed sequential data sets to key-sequenced data sets.

Access method services is used to convert an indexed-sequential data set to a key-sequenced data set. If a master and/or user catalog has been defined, define a key-sequenced data set with the attributes and performance options you want. Then use the access method services REPRO command to convert the indexed-sequential records and load them into the key-sequenced data set. VSAM builds the index for the key-sequenced data set as it loads the data set.

Each volume of a multivolume component must be on the same type of device; the data component and the index component, however, can be on volumes of devices of different types.

When you define the key-sequenced data set into which the indexed sequential data set is to be copied, you must specify the attributes of the VSAM data set for variable and fixed-length records.

For variable-length records:

  • VSAM record length equals ISAM DCBLRECL-4.
  • VSAM key length equals ISAM DCBKEYLE.
  • VSAM key position equals ISAM DCBRKP-4.

For fixed-length records:

  • VSAM record length (average and maximum must be the same) equals ISAM DCBLRECL (+ DCBKEYLE, if ISAM DCBRKP is equal to 0 and records are unblocked).
  • VSAM key length equals ISAM DCBKEYLE.
  • VSAM key position equals ISAM DCBRKP.

To learn the attributes of the ISAM data set you can use a program such as ISPF/PDF, ISMF, or the IEHLIST utility. IEHLIST can be used to get the dump format of the format 1 DSCB. The layout is in z/OS DFSMSdfp Advanced Services. The RKP is at X'5B'.

The level of sharing permitted when the key-sequenced data set is defined should be considered. If the ISAM program opens multiple DCBs pointing to different DD statements for the same data set, a share-options value of 1, which is the default, permits only the first DD statement to be opened. See Cross-Region Share Options for a description of the cross-region share-options values.

JCL is used to identify data sets and volumes for allocation. Data sets can also be allocated dynamically.

If JCL is used to describe an indexed sequential data set to be converted to VSAM using the access method services REPRO command, include DCB=DSORG=IS. Use a STEPCAT or JOBCAT DD statement as described in Retrieving an Existing VSAM Data Set to make user catalogs available; you can also use dynamic allocation.

With ISAM, deleted records are flagged as deleted, but are not actually removed from the data set. To avoid reading VSAM records that are flagged as deleted (X'FF'), code DCB=OPTCD=L. If your program depends on a record's only being flagged and not actually removed, you might want to keep these flagged records when you convert and continue to have your programs process these records. The access method services REPRO command has a parameter (ENVIRONMENT) that causes VSAM to keep the flagged records when you convert.

Go to the previous page Go to the next page




Copyright IBM Corporation 1990, 2014