topic Re: PC vs MAINFRAME in SAS Programming

PC vs MAINFRAME

Jaheuk — Mon, 04 Nov 2013 10:29:00 GMT

dear all,

after a remote submit I frequently use code below to view the remote data:

. . . . code . . .

%sysrput
pad=%SYSFUNC(PATHNAME(WORK));

ENDRSUBMIT;

libname rwork "&pad." server=prdsys;

HOW do I code the opposite?

something like this . . . but what servername to use . . . ??

%syslput pad=%bquote(%SYSFUNC(PATHNAME(WORK)));

RSUBMIT;

%put &pad. ;

libname rwork "&pad." server=XXXXXX ;

. . . code . . .

GreetZ,

Herman

Re: PC vs MAINFRAME

LaurentdeWalick — Mon, 04 Nov 2013 14:08:25 GMT

You can't ...

... unless you are running a SAS/CONNECT server on your local machine. I don't expect that will be the case. If you are running SAS/CONNECT locally, than the servername is the hostname of your local machine

Re: PC vs MAINFRAME

LinusH — Mon, 04 Nov 2013 14:34:32 GMT

Just out of curiosity, what are you trying to achieve?

Re: PC vs MAINFRAME

Jaheuk — Mon, 04 Nov 2013 15:13:47 GMT

>>> all code is submittted on SAS PC,

because of large datasets i do a remote submit on mainframe,

afterwards i read in the created subset on pc and process it

so from pc view this rwork libname works

BUT,

now if have a dataset on pc and i want, from mainframe, read it in directly (without doing a proc upload)

Re: PC vs MAINFRAME

LinusH — Mon, 04 Nov 2013 15:41:19 GMT

PROC UPLOAD will always be more efficient than reading a data set via a remote libname (exception: a very narrow where-clause on an indexed variable - but I doub't that your work-tables are indexed..?)