<?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/"
	xmlns:georss="http://www.georss.org/georss" xmlns:geo="http://www.w3.org/2003/01/geo/wgs84_pos#" xmlns:media="http://search.yahoo.com/mrss/"
		>
<channel>
	<title>Comments on: Coherence 3.4 Sneak Preview: Here comes C++!</title>
	<atom:link href="http://brianoliver.wordpress.com/2008/04/10/coherence-34-sneak-preview-here-comes-c/feed/" rel="self" type="application/rss+xml" />
	<link>http://brianoliver.wordpress.com/2008/04/10/coherence-34-sneak-preview-here-comes-c/</link>
	<description>Java, Oracle Coherence, Distributed Computing, Software Architecture, Extreme Transaction Processing, Apple, OS X and Life</description>
	<lastBuildDate>Fri, 06 Nov 2009 23:27:45 +0000</lastBuildDate>
	<generator>http://wordpress.com/</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Brian Oliver</title>
		<link>http://brianoliver.wordpress.com/2008/04/10/coherence-34-sneak-preview-here-comes-c/#comment-19</link>
		<dc:creator>Brian Oliver</dc:creator>
		<pubDate>Thu, 26 Jun 2008 12:55:37 +0000</pubDate>
		<guid isPermaLink="false">http://brianoliver.wordpress.com/?p=20#comment-19</guid>
		<description>The short answer is &quot;yes&quot; to your questions.  The C++ client is built very much like the .NET client (but with pure C++) and uses TCP/IP to connect to a data grid (using the *extend protocol with Portable Object Formatted objects on the wire).

In terms of relative performance between the different clients (as they are all based on the same network technology and protocols), the differences are often down to the language/platform specific optimizations and maturity of the client.  That is, while theoretically the .NET, Java and C++ should all exhibit similar performance characteristics, it turns out that maturity of the &quot;client&quot; implementation can be a differentiator in performance between features within clients.

For now it&#039;s very possible that the Java and .NET clients may out-perform the new C++ client, simply due to the maturity of the implementation.  eg: Initially when we released the .NET implementation it lagged a bit behind Java (in terms of performance), however it caught up very quickly.  In fact, by the time .NET was released as GA (Generally Available), it out-performed parts of the Java clients! 

I&#039;d expect to see the same with C++ client.  It&#039;s new and we (with customer feedback) will find new optimizations as the product is put through it&#039;s paces and matures.

While we make every effort to ensure consistent performance between the different language implementations, language specific optimizations, together with platform support, may make some clients better than others - for specific features.</description>
		<content:encoded><![CDATA[<p>The short answer is &#8220;yes&#8221; to your questions.  The C++ client is built very much like the .NET client (but with pure C++) and uses TCP/IP to connect to a data grid (using the *extend protocol with Portable Object Formatted objects on the wire).</p>
<p>In terms of relative performance between the different clients (as they are all based on the same network technology and protocols), the differences are often down to the language/platform specific optimizations and maturity of the client.  That is, while theoretically the .NET, Java and C++ should all exhibit similar performance characteristics, it turns out that maturity of the &#8220;client&#8221; implementation can be a differentiator in performance between features within clients.</p>
<p>For now it&#8217;s very possible that the Java and .NET clients may out-perform the new C++ client, simply due to the maturity of the implementation.  eg: Initially when we released the .NET implementation it lagged a bit behind Java (in terms of performance), however it caught up very quickly.  In fact, by the time .NET was released as GA (Generally Available), it out-performed parts of the Java clients! </p>
<p>I&#8217;d expect to see the same with C++ client.  It&#8217;s new and we (with customer feedback) will find new optimizations as the product is put through it&#8217;s paces and matures.</p>
<p>While we make every effort to ensure consistent performance between the different language implementations, language specific optimizations, together with platform support, may make some clients better than others &#8211; for specific features.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Vikas</title>
		<link>http://brianoliver.wordpress.com/2008/04/10/coherence-34-sneak-preview-here-comes-c/#comment-18</link>
		<dc:creator>Vikas</dc:creator>
		<pubDate>Thu, 19 Jun 2008 14:01:26 +0000</pubDate>
		<guid isPermaLink="false">http://brianoliver.wordpress.com/?p=20#comment-18</guid>
		<description>Is it like C++ client is extend client and how does it compare with Java client in terms of performance. As I understand Java client-no_local_storage communicates over TCMP whereas extend clients over TCP/IP. Again, if it is extend, then I guess between C++ and .NET, I guess the performance should be similar. By performance, I mean data transfer rate, relieving performance of the client SDK for the moment.</description>
		<content:encoded><![CDATA[<p>Is it like C++ client is extend client and how does it compare with Java client in terms of performance. As I understand Java client-no_local_storage communicates over TCMP whereas extend clients over TCP/IP. Again, if it is extend, then I guess between C++ and .NET, I guess the performance should be similar. By performance, I mean data transfer rate, relieving performance of the client SDK for the moment.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Brian Oliver</title>
		<link>http://brianoliver.wordpress.com/2008/04/10/coherence-34-sneak-preview-here-comes-c/#comment-14</link>
		<dc:creator>Brian Oliver</dc:creator>
		<pubDate>Wed, 30 Apr 2008 18:37:06 +0000</pubDate>
		<guid isPermaLink="false">http://brianoliver.wordpress.com/?p=20#comment-14</guid>
		<description>A good place to start is here;

http://wiki.tangosol.com/display/COH33UG/Defining+a+Data+Grid

A solid definition of Data Grids (and of course  Coherence).</description>
		<content:encoded><![CDATA[<p>A good place to start is here;</p>
<p><a href="http://wiki.tangosol.com/display/COH33UG/Defining+a+Data+Grid" rel="nofollow">http://wiki.tangosol.com/display/COH33UG/Defining+a+Data+Grid</a></p>
<p>A solid definition of Data Grids (and of course  Coherence).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Neville Franks</title>
		<link>http://brianoliver.wordpress.com/2008/04/10/coherence-34-sneak-preview-here-comes-c/#comment-13</link>
		<dc:creator>Neville Franks</dc:creator>
		<pubDate>Sun, 27 Apr 2008 22:16:20 +0000</pubDate>
		<guid isPermaLink="false">http://brianoliver.wordpress.com/?p=20#comment-13</guid>
		<description>Brian, Where can I read about what Coherence is? 

FYI I&#039;m a C++ dev, not Java.</description>
		<content:encoded><![CDATA[<p>Brian, Where can I read about what Coherence is? </p>
<p>FYI I&#8217;m a C++ dev, not Java.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Andrew Roberts</title>
		<link>http://brianoliver.wordpress.com/2008/04/10/coherence-34-sneak-preview-here-comes-c/#comment-8</link>
		<dc:creator>Andrew Roberts</dc:creator>
		<pubDate>Thu, 10 Apr 2008 20:29:35 +0000</pubDate>
		<guid isPermaLink="false">http://brianoliver.wordpress.com/?p=20#comment-8</guid>
		<description>This is excellent news. It&#039;s insane just how many Java based HPC products get their C++ interface wrong! This makes the product both a chore to develop for and deploy.

Yes the C++ API for Coherence has been a very long time coming but that&#039;s because you guys really take the time to do things properly. I look forward to the final release!

PS: Apologies we could not make it to the SIG. Are the materials available anywhere?</description>
		<content:encoded><![CDATA[<p>This is excellent news. It&#8217;s insane just how many Java based HPC products get their C++ interface wrong! This makes the product both a chore to develop for and deploy.</p>
<p>Yes the C++ API for Coherence has been a very long time coming but that&#8217;s because you guys really take the time to do things properly. I look forward to the final release!</p>
<p>PS: Apologies we could not make it to the SIG. Are the materials available anywhere?</p>
]]></content:encoded>
	</item>
</channel>
</rss>
