> The easiest is probably using the SPLITBY parameter of the > OUTFIL control statement. This should work with either SyncSort or > other sort products. The SPLITBY=n parameter writes groups of records > in rotation among multiple output data sets and distributes multiple > records at a time among the OUTFIL data sets. N specifies the number of > records to split by. The following control statements will copy the > first 1,00,000 (not sure if this is a typo or if is should be 1,000,000) > to the first data set and the next 1,00,000 to the next data set and so > on. The only thing you need to be careful of is to allocate enough data > sets. If you need 6 data sets but only allocate 5, the next group after > the 5th, the one that starts with the 6,00,001st record will be written > to the fist data set again and the rotation continues. > ...
Raj, If you have access to DFSORT, you can use its SPLIT1R=n parameter instead of SPLITBY=n. Whereas SPLITBY=n rotates the records back to the first output file (which is not desirable in your case), SPLIT1R=n will continue to write the extra records to the last output file so each output file will have contiguous records from the input file. For complete details on DFSORT's SPLIT1R=n parameter, see: www.ibm.com/servers/storage/support/software/sort/mvs/peug/ For another way to split the records evenly and contiguously, see the "Split a file to n output files dynamically" Smart DFSORT Trick at: http://www.ibm.com/servers/storage/support/software/sort/mvs/tricks/ Frank Yaeger - DFSORT Development Team (IBM) - [EMAIL PROTECTED] Specialties: PARSE, JFY, SQZ, ICETOOL, IFTHEN, OVERLAY, Symbols, Migration => DFSORT/MVS is on the Web at http://www.ibm.com/storage/dfsort/ ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
