GEOCLUSTER© is CGG's seismic processing software limited version for test/demonstration purpose.
Components:
1- An executable binary file installed on the CE:
yuc.exe: GEOCLUSTER core executable for Linux, takes a GSL (Geocluster Seismic Language) file as input.
2- A script file installed on the CE:
run.sh: the shell script used to launch the seismic job and manage output files. Takes the GSL file as argument;
The following link is created at install: /usr/bin/Egeode-GCT -> /opt/EGEODE/GCT/run.sh
Example (provided that /usr/bin/ is in $PATH):
> Egeode-GCT fanmo.gsl
Create a private working directory for the seismic job; it is removed after job completion.
It manage 3 output files :
- log is the system execution messages that is useful only if the job crash, for investigation and debugging
purpose.
- err : error file;
- list contains output results from the seismic processing modules; The Geocluster user on the grid will want
to transfer this file to his local site.
3- The user's seismic job file coded in the GSL (Geocluster Seismic Language) to be submitted with the jdl:
fanmo.gsl: an example of a seismic job coded in the GSL.
This is where the geophysicist builds the processing sequence;
You would need the Ceocluster documentation in order to understand this kind of files;
This is just to give you a brief description;
The syntax is column sensitive:
1 to 2: '* ' a star followed by a space means a module name is to be read in column 3 to 7;
'**' 2 stars mean this line is a comment line;
' ' 2 spaces mean this is a continuation line for current module's parameters.
2 to 7: Geocluster seismic module name
9 to 10: seismic module option
15 to 16: seismic module input buffer
23 to 24: seismic module output buffer
31 to 80: seismic module parameters; May continue on next line starting at column 3.
The GSL file is divided in 2 parts:
The 1rst part provide auxiliary data for the job (LIBRI) velocity laws, water bottom data, mute data;
The 2nd part contains the seismic modules within a loop (BOUCL...FINBO); the module DAGEN generates
synthetic seismic traces that are delivered into the buffer EA; module FANMO (Hyperbolic, non-hyperbolic
or liner move out) takes the seismic trace as input in buffer EA and output the processed seismic trace
in buffer EB; module LISTE print the contents of trace EB in the job output listing file "list$$" (cf script
file "run" below).
Here is the list of the modules currently available with this version:
* LIBRI VI, * LIBRI FD, * LIBRI MU, * LIBRI UM, * LIBRI CN, * RUNET, * DAGEN, * CORRE, * DUPLI, * MODET,
* LISTE, * TRANS, * HISTA, * EVERY, * SCALE, * SELEC, * FANMO, * FILTR, * MUTES, * STACK
This example file is given the name "fanmo.gsl" after the main processing module FANMO; but there is no rule
for naming, it can be given any name. The user of Geocluster on the grid should be able to edit existing GSL
files or create new ones.