<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: SSIS &#8211; Avoiding the Sort Components</title>
	<atom:link href="http://www.ssistalk.com/2009/09/17/ssis-avoiding-the-sort-components/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.ssistalk.com/2009/09/17/ssis-avoiding-the-sort-components/</link>
	<description>Random thoughts and experiences with SSIS, by Phil Brammer</description>
	<lastBuildDate>Wed, 25 Jan 2012 10:27:44 -0600</lastBuildDate>
	<generator>http://wordpress.org/?v=2.9.2</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: hello</title>
		<link>http://www.ssistalk.com/2009/09/17/ssis-avoiding-the-sort-components/comment-page-1/#comment-18021</link>
		<dc:creator>hello</dc:creator>
		<pubDate>Fri, 08 Apr 2011 08:29:10 +0000</pubDate>
		<guid isPermaLink="false">http://www.ssistalk.com/?p=170#comment-18021</guid>
		<description>Its Really helpful.</description>
		<content:encoded><![CDATA[<p>Its Really helpful.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Deepak T</title>
		<link>http://www.ssistalk.com/2009/09/17/ssis-avoiding-the-sort-components/comment-page-1/#comment-9554</link>
		<dc:creator>Deepak T</dc:creator>
		<pubDate>Thu, 11 Nov 2010 03:51:01 +0000</pubDate>
		<guid isPermaLink="false">http://www.ssistalk.com/?p=170#comment-9554</guid>
		<description>This article was helpful, thanks..</description>
		<content:encoded><![CDATA[<p>This article was helpful, thanks..</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Santhosh</title>
		<link>http://www.ssistalk.com/2009/09/17/ssis-avoiding-the-sort-components/comment-page-1/#comment-7578</link>
		<dc:creator>Santhosh</dc:creator>
		<pubDate>Wed, 06 Oct 2010 05:46:35 +0000</pubDate>
		<guid isPermaLink="false">http://www.ssistalk.com/?p=170#comment-7578</guid>
		<description>Hi Phil/Guna,

Since i am very new to SSIS i need some clarifications.
Correct me if i am wrong as per the above discussion setting the IsSortProperty to true indicates Merge Join component that the data is Sorted form where as the data internally would not be sorted ???

I have used two OLE DB Source wher i have set the IsSorted Property to true and also i have set sort Key Position,the rows coming out of the Merge join here is 4 which is Incorrect.
But If use the Sort tranfomation after this OLE DB Source i am getting 13 rows which is the Proper result.

So can u please tell me the difference between the IsSortedProperty and the Sort Tranfomation.

Thank u in advance.

Santhosh Hiriyanna</description>
		<content:encoded><![CDATA[<p>Hi Phil/Guna,</p>
<p>Since i am very new to SSIS i need some clarifications.<br />
Correct me if i am wrong as per the above discussion setting the IsSortProperty to true indicates Merge Join component that the data is Sorted form where as the data internally would not be sorted ???</p>
<p>I have used two OLE DB Source wher i have set the IsSorted Property to true and also i have set sort Key Position,the rows coming out of the Merge join here is 4 which is Incorrect.<br />
But If use the Sort tranfomation after this OLE DB Source i am getting 13 rows which is the Proper result.</p>
<p>So can u please tell me the difference between the IsSortedProperty and the Sort Tranfomation.</p>
<p>Thank u in advance.</p>
<p>Santhosh Hiriyanna</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Phil Brammer</title>
		<link>http://www.ssistalk.com/2009/09/17/ssis-avoiding-the-sort-components/comment-page-1/#comment-1938</link>
		<dc:creator>Phil Brammer</dc:creator>
		<pubDate>Wed, 30 Sep 2009 03:07:50 +0000</pubDate>
		<guid isPermaLink="false">http://www.ssistalk.com/?p=170#comment-1938</guid>
		<description>Guna,
Yes, that&#039;s correct.  Only set the IsSorted property to true if your source is, well, sorted.  Actually, to be more specific, use this property only if your SortKeyPosition columns are specified as the source is actually sorted.   In other words if the source is sorted on the first column in the file, but you need to use the Merge Join component and joining on the second column in the source file, you should not use the IsSorted property, if the file is only sorted by the first column.

If your flat file or database source is not sorted, then setting IsSorted to True can have negative effects when trying to use components like the Merge or Merge Join.</description>
		<content:encoded><![CDATA[<p>Guna,<br />
Yes, that&#8217;s correct.  Only set the IsSorted property to true if your source is, well, sorted.  Actually, to be more specific, use this property only if your SortKeyPosition columns are specified as the source is actually sorted.   In other words if the source is sorted on the first column in the file, but you need to use the Merge Join component and joining on the second column in the source file, you should not use the IsSorted property, if the file is only sorted by the first column.</p>
<p>If your flat file or database source is not sorted, then setting IsSorted to True can have negative effects when trying to use components like the Merge or Merge Join.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Guna</title>
		<link>http://www.ssistalk.com/2009/09/17/ssis-avoiding-the-sort-components/comment-page-1/#comment-1937</link>
		<dc:creator>Guna</dc:creator>
		<pubDate>Tue, 29 Sep 2009 22:35:06 +0000</pubDate>
		<guid isPermaLink="false">http://www.ssistalk.com/?p=170#comment-1937</guid>
		<description>Phil,
The IsSorted is available for flat files, but that is only useful if the data on the flat file is sorted, right?
what if the flat file data is not sorted? then we can&#039;t specify on ORDER BY, and so we have to use the Sort component.
Correct me if I&#039;m wrong?
thanks.</description>
		<content:encoded><![CDATA[<p>Phil,<br />
The IsSorted is available for flat files, but that is only useful if the data on the flat file is sorted, right?<br />
what if the flat file data is not sorted? then we can&#8217;t specify on ORDER BY, and so we have to use the Sort component.<br />
Correct me if I&#8217;m wrong?<br />
thanks.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Phil Brammer</title>
		<link>http://www.ssistalk.com/2009/09/17/ssis-avoiding-the-sort-components/comment-page-1/#comment-1935</link>
		<dc:creator>Phil Brammer</dc:creator>
		<pubDate>Tue, 29 Sep 2009 16:28:02 +0000</pubDate>
		<guid isPermaLink="false">http://www.ssistalk.com/?p=170#comment-1935</guid>
		<description>Ryan,
That is not true - you can still use the above techniques for flat files.  The IsSorted property is available for flat files as well.</description>
		<content:encoded><![CDATA[<p>Ryan,<br />
That is not true &#8211; you can still use the above techniques for flat files.  The IsSorted property is available for flat files as well.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ryan Shirley</title>
		<link>http://www.ssistalk.com/2009/09/17/ssis-avoiding-the-sort-components/comment-page-1/#comment-1932</link>
		<dc:creator>Ryan Shirley</dc:creator>
		<pubDate>Mon, 28 Sep 2009 17:01:02 +0000</pubDate>
		<guid isPermaLink="false">http://www.ssistalk.com/?p=170#comment-1932</guid>
		<description>Thanks for the post, Phil.  I agree that sorting on the database side is faster and always preferred.  Unfortunately you still have to resort to the Sort component when using flat files as a data source.  The one bonus is that the Sort component also lets you remove duplicate values in your sort key.  So, if you are trying to do a &#039;SELECT DISTINCT&#039; from a flat file, Sort is your friend.</description>
		<content:encoded><![CDATA[<p>Thanks for the post, Phil.  I agree that sorting on the database side is faster and always preferred.  Unfortunately you still have to resort to the Sort component when using flat files as a data source.  The one bonus is that the Sort component also lets you remove duplicate values in your sort key.  So, if you are trying to do a &#8216;SELECT DISTINCT&#8217; from a flat file, Sort is your friend.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: James Rowland-Jones</title>
		<link>http://www.ssistalk.com/2009/09/17/ssis-avoiding-the-sort-components/comment-page-1/#comment-1903</link>
		<dc:creator>James Rowland-Jones</dc:creator>
		<pubDate>Sun, 20 Sep 2009 09:55:48 +0000</pubDate>
		<guid isPermaLink="false">http://www.ssistalk.com/?p=170#comment-1903</guid>
		<description>Hi Phil,

Great post and a very important area to highlight.  As I see it the big problem with the SORT component is that the bigger the sort the slower it gets - and the performance degradation isn&#039;t linear it&#039;s n log n.  Invariably users don&#039;t find this out till they hit the full production volume by which time the Solution has been built and tested.  The existence of a SORT inside a dataflow is definitely a &quot;smell&quot; for me.

How I wish the SSIS team would buy NSORT for the product...

Cheers, James</description>
		<content:encoded><![CDATA[<p>Hi Phil,</p>
<p>Great post and a very important area to highlight.  As I see it the big problem with the SORT component is that the bigger the sort the slower it gets &#8211; and the performance degradation isn&#8217;t linear it&#8217;s n log n.  Invariably users don&#8217;t find this out till they hit the full production volume by which time the Solution has been built and tested.  The existence of a SORT inside a dataflow is definitely a &#8220;smell&#8221; for me.</p>
<p>How I wish the SSIS team would buy NSORT for the product&#8230;</p>
<p>Cheers, James</p>
]]></content:encoded>
	</item>
</channel>
</rss>

