Mainframe interview questions I faced.

[GhuGhu]

Hi all,

I faced few interview questions today can anyone help me with the correct answers?

1). One PS file is having 1000 records I want to read the records from last record to first record ?

2). PS file which is having millions of records and there are some duplicate records with word of "abcdef" in that file ...I don't know the length and position I want to delete the duplicate field from the file?

[Robert Sample]

1. This question comes up fairly frequently. Mainframes do not have any way to read sequential data sets backwards, so without a significant amount of effort it cannot be done. (Sequential data sets can be on magnetic tape, for example, which only can be read in one direction.)

2. The problem description is so vague as to be farcical. You could use DFSORT to pull out the records with "abcdef" without knowing starting position. But what determines a duplicate record -- every byte is duplicated? the key is duplicated? some subset of bytes are duplicated? There's just not enough information available to do much about the question.

[GhuGhu]

1. This question comes up fairly frequently. Mainframes do not have any way to read sequential data sets backwards, so without a significant amount of effort it cannot be done. (Sequential data sets can be on magnetic tape, for example, which only can be read in one direction.)

2. The problem description is so vague as to be farcical. You could use DFSORT to pull out the records with "abcdef" without knowing starting position. But what determines a duplicate record -- every byte is duplicated? the key is duplicated? some subset of bytes are duplicated? There's just not enough information available to do much about the question.
 

I think the key was duplicated.