As I understant your situation, your SAS server logon/initialization processing assigns a WORK library (e.g.,  J:\SASWORK\_TD26644_B000010455_) and a TEMP library (e.g., W:\_TD26644_WAPPRIB00001040_).  Your programs use the TEMP library instead of the WORK library.  Both are deleted at the end of the SAS session.  (I don't see the point but there must be one.) Jaap tells how to map the remote (server-side) WORK directory for local (desktop) access but you want to map the TEMP library for local access.  SAS allows you to map any remote library for local access. Let's say in your signon/connect program you've assigned the remote server the alias 'bigOne' and the TEMP library is named 'temp'.  In the SAS editor, in the local environment issue one of these two commands:      To use a libref name different from the remote libref when you access it from the local environment use the construct Jaap provided --           libref remTemp remote server=bigOne slibref=temp;       ** remote is not required but is a visual aid to make it even more obvious that I'm referencing a remote libref. ;      To use the same libref name as the remote when you access it locally --           library temp remote server=bigOne;     ** remote is not required. ; I don't use EG, but to make it run automatically you can add either of these librefs to the bottom of your connect program (if you can modify it) -- just be sure it is after the remote libref TEMP is assigned. You might want to look at my 2011 SGF paper "Using SAS to Move Data Between Servers". ... View more

This afternoon I needed this very solution.  Thanks to both of you, Paul and Art. John Bentley ... View more

I suspect that at least part of your performance problem is that you're using ODBC connectivity.  My experience is that that is s-l-o-w for anything more than a couple hundred thousand rows.       connect to odbc as oracle Ask around and be sure you don't have SAS Access/Oracle.  That way you can control which of the datasets is moved to the other environment depending on which is the smaller dataset. It could be that excessive time is spent creating the SAS data set A.  Check the log and if you think so, when you create data set A add a Set keep= or drop= option to the where= to insure that only the fields you need are loaded into A... loading unneeded data is a common mistake that can have a huge performance impact. data a; set b (where = (var1<= date and var2>date and var3>date); I've written a couple papers on using SAS and Oracle.  The first one deals with SAS Options that either improve or degrade performance and I think you'll find a couple things here that will be useful to you, like how to make the read buffer larger.  Each section has a quick tabular summary so that you can find the option(s) that look appropriate for your situation. http://support.sas.com/resources/papers/proceedings13/072-2013.pdf Another paper is not so appropriate to this problem but explains how to leverage database and system return codes.  If you eventually need to automate your processes then these are critical for trapping and handling errors.  http://analytics.ncsu.edu/sesug/2012/BB-03.pdf Good luck! John ... View more

I'm not very familiar with SAS's Intersect operator but from your syntax it 'joins' two stand-alone Select queries, like a Union does.  Neither query has any knowledge of the other--the query with the table aliased as ball06 has no knowledge of the query with the table aliased as ball04 so your Where critieria will not work.  That's what the error message is saying. Maybe Tom's second answer will give you the results you want. ... View more

You say you want to move 'a file'.  What kind of file is it?  If it's a SAS data set you will naturally see lots of special characters if you use any viewer/editor except SAS to open and browse the data. Your connect error sounds to me like a mainframe-side error thrown by SAS/Connect.  You need SAS/Connect on both sides, client and server (PC and mainframe).  Are you sure you have SAS/Connect installed on the mainframe?  In light of security concerns it would not surprise me if it's not installed or if you just don't have access to it.  Mainframe IT security folks can be even more paranoid than the server-side people, if that's possible.  The default connect script from SAS is very robust and should work as-is. Regarding the FTP error, FTPing doesn't convert anything.   I suggest explicity converting the file on the mainframe to either a column- or comma-delimited text file.  That way you'll know exactly what you have.  But even then you will have to use SAS's EBCDIC formats when you read the data into a PC SAS dataset.  Hey, maybe that's the problem... you're trying to import a mainframe EBCDIC file into an ASCII system on the PC.  Read SAS's Tech Note TS-642, "Reading EBCDIC Files on ASCII Systems".      https://support.sas.com/techsup/technote/ts642.html When you FTP the file, move it binary format so it lands as an exact image of the original.  If it is "large" then first subset the data on the mainframe to get a few hundred/couple thousand records for testing.  You might look at my SGF 2011 paper titled "Using SAS to Move Data between Servers".  There is a section towards the end that talks about using SAS's FTP/SFTP access method and a code example is provided.  The SAS documentation on FTP/SFTP is also very good. Good luck! John ... View more

I am not using DI studio so cannot comment on your particular issue other than to say perhaps you need to use a code window.  The code provided by DBaily on March 31st, 2011 solved my problem. ... View more

More than two years later, this fixed my problem.. no matter what I was doing the date was coming out 1/1/1960.  SAS 9.1.3 (yes, still) and Oracle 11g.  Thanks for posting! ... View more

I'm going to be presenting a paper at SGF in San Francisco next month on _reading_ Word documents into SAS and the process can be reversed to write SAS data to a Word document but I don't cover it.  The trick is setting bookmarks in Word--such as bookmarking a table--and then running SAS so that it outputs data into the bookmarked table.  You don't have to worry about how many rows are in the data set because a bookmarked table is dynamic in terms of rows. I'd rather not post the paper here but if you send me your email address I'll send you an advance copy.  john.bentley@wellsfargo.com ... View more