<?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"
	>
<channel>
	<title>Comments on: InterBase in a Multi-tier World</title>
	<atom:link href="http://blogs.teamb.com/craigstuntz/2004/06/17/441/feed/" rel="self" type="application/rss+xml" />
	<link>http://blogs.teamb.com/craigstuntz/2004/06/17/441/</link>
	<description>C# • Entity Framework • Functional Programming • MVC • Web</description>
	<pubDate>Sat, 11 Feb 2012 15:03:25 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.6.2</generator>
		<item>
		<title>By: Fred</title>
		<link>http://blogs.teamb.com/craigstuntz/2004/06/17/441/#comment-4530</link>
		<dc:creator>Fred</dc:creator>
		<pubDate>Fri, 20 Feb 2009 07:36:07 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.teamb.com/craigstuntz/2004/06/17/420/#comment-4530</guid>
		<description>Hi Craig,

this is a great post about Datasnap &#38; IB.

It would be so cool if could write a new post about Datasnap 2009 &#38; whatever rdbm. 

I'm planning to use Datasnap 2009 but there are so very few quality ressources. "Real world" exemples are really rare with datasnap. This post is very usefull for me but quite outdated.

Thanks for you blog and the quality posts.</description>
		<content:encoded><![CDATA[<p>Hi Craig,</p>
<p>this is a great post about Datasnap &amp; IB.</p>
<p>It would be so cool if could write a new post about Datasnap 2009 &amp; whatever rdbm. </p>
<p>I&#8217;m planning to use Datasnap 2009 but there are so very few quality ressources. "Real world" exemples are really rare with datasnap. This post is very usefull for me but quite outdated.</p>
<p>Thanks for you blog and the quality posts.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Craig Stuntz</title>
		<link>http://blogs.teamb.com/craigstuntz/2004/06/17/441/#comment-3326</link>
		<dc:creator>Craig Stuntz</dc:creator>
		<pubDate>Fri, 11 Jan 2008 13:43:54 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.teamb.com/craigstuntz/2004/06/17/420/#comment-3326</guid>
		<description>You can't use PacketRecords  -1 with a stateless server unless you handle the paging yourself. There is code on CodeCentral which demonstrates how to do this.</description>
		<content:encoded><![CDATA[<p>You can&#8217;t use PacketRecords  -1 with a stateless server unless you handle the paging yourself. There is code on CodeCentral which demonstrates how to do this.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Christopher Brian Jurado</title>
		<link>http://blogs.teamb.com/craigstuntz/2004/06/17/441/#comment-3325</link>
		<dc:creator>Christopher Brian Jurado</dc:creator>
		<pubDate>Fri, 11 Jan 2008 11:18:17 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.teamb.com/craigstuntz/2004/06/17/420/#comment-3325</guid>
		<description>Read ur post about multitier.
When the app server queries the data on the server side, it involves processing time to execute the query. But then only a portion of that is sent to the client if we use the packetrecords property. Assume that the server is stateless (like using TSOAPConnection) and we found a way to communicate state about w/c records to get next. There is a  problem when the next set of records are requested by the clientdataset. Since the server is stateless, the whole query was freed on the server after the first query. And it would have to execute the whole query again, and send the second portion. It wastes query processing time and that query might take long or might have thousands of records. Moreover, the data might have changed at that point and the second set may not be what was expected. Doesn't this cause problems like for reporting? And doesn't this hurt query performance on the server side?
I agree with Stateless servers. But, how can you avoid this problem in Delphi?</description>
		<content:encoded><![CDATA[<p>Read ur post about multitier.<br />
When the app server queries the data on the server side, it involves processing time to execute the query. But then only a portion of that is sent to the client if we use the packetrecords property. Assume that the server is stateless (like using TSOAPConnection) and we found a way to communicate state about w/c records to get next. There is a  problem when the next set of records are requested by the clientdataset. Since the server is stateless, the whole query was freed on the server after the first query. And it would have to execute the whole query again, and send the second portion. It wastes query processing time and that query might take long or might have thousands of records. Moreover, the data might have changed at that point and the second set may not be what was expected. Doesn&#8217;t this cause problems like for reporting? And doesn&#8217;t this hurt query performance on the server side?<br />
I agree with Stateless servers. But, how can you avoid this problem in Delphi?</p>
]]></content:encoded>
	</item>
</channel>
</rss>

