Seismic UNIX Data Flow

SU consists of many, relatively small, independent programs that communicate through a common data format (modified SEGY).

Contrast this mode of operation to Landmark ProMAX, which is a “single”, large program consisting of many sub-programs that pass data through in-memory links in a processing flow.

SU uses files and pipes to pass data between programs.

SU programs read and write files via UNIX re-direction (< and >).

SU programs read and write to each other via UNIX pipes ( | ).

An SU program “thinks” that it is reading data from standard input, i.e. the keyboard on your workstation, and that it is writing data to standard output, i.e. the screen, but UNIX re-direction allows stdin and stdout to be actual files.

Syntax:

sugethw < x.su Input to sugethw is from x.su

sufilter > y.su Output from sufilter is to y.su

suwind < x.su > y.su Input suwind is from x.su, and output is to y.su

The obvious disadvantage of using all these files is that disk space is cheap, but not free.

suwind <x.su | sufilter | susort | sufrac >y.su

Pipes are both input and output structures, and don’t use any permanent disk space.