<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Re: Implicit or explicit pass through? in SAS Procedures</title>
    <link>https://communities.sas.com/t5/SAS-Procedures/Implicit-or-explicit-pass-through/m-p/401449#M66772</link>
    <description>&lt;P&gt;You cannot use explicit pass through because that works on the server&amp;nbsp;and you don't have the SAS data set on the server.&amp;nbsp;&lt;/P&gt;
&lt;P&gt;If you can load your table to the server, that's a different story.&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Since you're using IMPLICIT queries, you can treat them both as 'SAS tables' and your code isn't any different.&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Thu, 05 Oct 2017 17:14:58 GMT</pubDate>
    <dc:creator>Reeza</dc:creator>
    <dc:date>2017-10-05T17:14:58Z</dc:date>
    <item>
      <title>Implicit or explicit pass through?</title>
      <link>https://communities.sas.com/t5/SAS-Procedures/Implicit-or-explicit-pass-through/m-p/401444#M66771</link>
      <description>&lt;P&gt;Appreciate if someone help me with the following situation. I've a Table A in SAS which has some 2 million rows with 25 variables and&lt;/P&gt;
&lt;P&gt;I've a other table B in Netezza which 10 millions of records with 42 variables. Now I want to join these two tables and I'm not sure whether I should use Implicit or explicit pass through for better performance.&amp;nbsp; In either method, I would like to know the basic structure of SAS code to achieve this. I've done joins using only SAS datasets or only non SAS datasets but not with both.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Thu, 05 Oct 2017 16:59:17 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Procedures/Implicit-or-explicit-pass-through/m-p/401444#M66771</guid>
      <dc:creator>Babloo</dc:creator>
      <dc:date>2017-10-05T16:59:17Z</dc:date>
    </item>
    <item>
      <title>Re: Implicit or explicit pass through?</title>
      <link>https://communities.sas.com/t5/SAS-Procedures/Implicit-or-explicit-pass-through/m-p/401449#M66772</link>
      <description>&lt;P&gt;You cannot use explicit pass through because that works on the server&amp;nbsp;and you don't have the SAS data set on the server.&amp;nbsp;&lt;/P&gt;
&lt;P&gt;If you can load your table to the server, that's a different story.&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Since you're using IMPLICIT queries, you can treat them both as 'SAS tables' and your code isn't any different.&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Thu, 05 Oct 2017 17:14:58 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Procedures/Implicit-or-explicit-pass-through/m-p/401449#M66772</guid>
      <dc:creator>Reeza</dc:creator>
      <dc:date>2017-10-05T17:14:58Z</dc:date>
    </item>
    <item>
      <title>Re: Implicit or explicit pass through?</title>
      <link>https://communities.sas.com/t5/SAS-Procedures/Implicit-or-explicit-pass-through/m-p/401452#M66773</link>
      <description>&lt;P&gt;Well, sending the 2mil rows to the database should be quicker than taking 10mil rows from the database, but by number alone.&amp;nbsp; However its not that simple, there is the hardware available to each system for instance.&amp;nbsp; That is quite a few records, you may want to test with a subset and get code working before running it in total.&amp;nbsp; You may want to consult with your Netezza person.&lt;/P&gt;
&lt;P&gt;As for code, its pretty similar either way,&amp;nbsp;&lt;/P&gt;
&lt;PRE&gt;proc sql;
  connect to net (...);
  select * from connection to net (
    select ....);

  /* or to send to db*/
  execute by net(create table yourworkspace as select * from saslib.datas);

  disconnect from net;
quit;

/* Or you could setup libnames and make the SQL a bit smaller.*/&lt;/PRE&gt;
&lt;P&gt;Many examples out there on how to go one way or the other, but which would be better is hard to say and also depends a lot on what you want to do further with it, for instance if you were creating a SAS report, might make more sense to have all the data in SAS.&lt;/P&gt;</description>
      <pubDate>Thu, 05 Oct 2017 17:16:14 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Procedures/Implicit-or-explicit-pass-through/m-p/401452#M66773</guid>
      <dc:creator>RW9</dc:creator>
      <dc:date>2017-10-05T17:16:14Z</dc:date>
    </item>
    <item>
      <title>Re: Implicit or explicit pass through?</title>
      <link>https://communities.sas.com/t5/SAS-Procedures/Implicit-or-explicit-pass-through/m-p/401490#M66775</link>
      <description>&lt;P&gt;Also consider exporting to a flat file, copy that to the database server, read it into a table there, and do the join in the database. Just as one method to compare, performancewise.&lt;/P&gt;</description>
      <pubDate>Thu, 05 Oct 2017 19:33:38 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Procedures/Implicit-or-explicit-pass-through/m-p/401490#M66775</guid>
      <dc:creator>Kurt_Bremser</dc:creator>
      <dc:date>2017-10-05T19:33:38Z</dc:date>
    </item>
    <item>
      <title>Re: Implicit or explicit pass through?</title>
      <link>https://communities.sas.com/t5/SAS-Procedures/Implicit-or-explicit-pass-through/m-p/401624#M66782</link>
      <description>There ia also the oprion of uploading the SAS data and then use implicit pass through.&lt;BR /&gt;The code will be simpler and more transferable.</description>
      <pubDate>Fri, 06 Oct 2017 06:20:50 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Procedures/Implicit-or-explicit-pass-through/m-p/401624#M66782</guid>
      <dc:creator>LinusH</dc:creator>
      <dc:date>2017-10-06T06:20:50Z</dc:date>
    </item>
  </channel>
</rss>

