Remove the BY clause from the SQL statements. This should eliminate the sort.

The file can then be output and processed in any way needed......

HTH.
<sbip>
>You could split the input file into parts by number of records or the 
>high part of the key, sort each part, then merge the parts (or 
>concatenate the datasets in key range order).

The data to be sorted is the result of complex DB2 queries, which determine 
what records to read from dozens or even hundreds of different data sets. These 
records are then handed over to DF-Sort internally. So, there is no 
intermediate file I could split.
</snip>

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to