Simple job with data manipulation

From MediaWiki

(Difference between revisions)
Jump to: navigation, search
Line 33: Line 33:
2. Enter directory hello.SE
2. Enter directory hello.SE
-
  cd hello.SE
+
  [ngrkic@ui jobovi]$ cd hello.SE
3. List directory
3. List directory
-
  ll hello.SE
+
  [ngrkic@ui jobovi]$ ll hello.SE
4. See content of files
4. See content of files

Revision as of 13:51, 9 December 2011

This is simple single job submitting example with data management. In this case, WN will download source code and executing script from SE and after successful ending of job, result will be uploaded on SE.

hello.SE.tgz

Content of uploaded files:

hello.c

#include <stdio.h>

int main()
{
 printf ("hello world!!\n");
 return 0;
}

hello.sh

#Compile C file

gcc hello.c -o hello

#Execute hello program

./hello


1. Extract files

[ngrkic@ui jobovi]$ tar xvzf hello.SE.tgz

2. Enter directory hello.SE

[ngrkic@ui jobovi]$ cd hello.SE

3. List directory

[ngrkic@ui jobovi]$ ll hello.SE

4. See content of files

[ngrkic@ui hello.SE]$ cat hello.jdl 
Executable = "run.sh";
Arguments = "";
StdOutput = "std.out";
StdError = "std.err";
InputSandbox = {"run.sh"};
OutputSandbox = {"std.out", "std.err"};
Requirements =  RegExp("ce64.ipb.ac.rs*",other.GlueCEUniqueID);
[ngrkic@ui hello.SE]$ cat run.sh 
#Downloading source code from SE
lcg-cp lfn:/grid/aegis/nikola/hello.SE/hello.c 'pwd'

#Downloading script that compiles source code and execute program
lcg-cp lfn:/grid/aegis/nikola/hello.SE/hello.sh 'pwd'

#runing compile & execute script
./hello.sh

#Uploads output to SE
lcg-cr -l lfn:/grid/aegis/nikola/hello.SE/std.out file:`pwd`/std.out

5.

Personal tools