Comments on: Merging Data And Processing: Why it doesn’t “just work” http://www.benstopford.com/2009/08/30/the-trials-of-merging-data-and-processing-in-coherence-why-it-doesnt-quite-just-work/#utm_source=feed&utm_medium=feed&utm_campaign=feed Gently flexing the grid Wed, 15 May 2013 12:40:53 +0000 http://wordpress.org/?v=2.9.1 hourly 1 By: best cccam server for spiderbox http://www.benstopford.com/2009/08/30/the-trials-of-merging-data-and-processing-in-coherence-why-it-doesnt-quite-just-work/comment-page-1/#comment-40112 best cccam server for spiderbox Wed, 15 May 2013 12:40:53 +0000 http://www.benstopford.com/?p=73#comment-40112 I am not sure where you're getting your information, but good topic. I needs to spend some time learning more or understanding more. Thanks for great info I was looking for this information for my mission. I am not sure where you’re getting your information, but good topic. I needs to spend some time learning more or understanding more. Thanks for great info I was looking for this information for my mission.

]]> By: ben http://www.benstopford.com/2009/08/30/the-trials-of-merging-data-and-processing-in-coherence-why-it-doesnt-quite-just-work/comment-page-1/#comment-2931 ben Wed, 30 May 2012 07:03:44 +0000 http://www.benstopford.com/?p=73#comment-2931 Thanks for the comments guys Thanks for the comments guys

]]> By: Guilherme Melo http://www.benstopford.com/2009/08/30/the-trials-of-merging-data-and-processing-in-coherence-why-it-doesnt-quite-just-work/comment-page-1/#comment-2277 Guilherme Melo Wed, 21 Mar 2012 13:41:55 +0000 http://www.benstopford.com/?p=73#comment-2277 Henry, Take a look at the 3.7 backing map and entry processor documentation, as now if you access the backing map with the new getBackingMapEntry method you will lock those entries and they will be part of the same transaction as the entry processor (provided they are in the same partition). Great post Ben. Henry,

Take a look at the 3.7 backing map and entry processor documentation, as now if you access the backing map with the new getBackingMapEntry method you will lock those entries and they will be part of the same transaction as the entry processor (provided they are in the same partition).

Great post Ben.

]]> By: Henry http://www.benstopford.com/2009/08/30/the-trials-of-merging-data-and-processing-in-coherence-why-it-doesnt-quite-just-work/comment-page-1/#comment-2199 Henry Fri, 16 Mar 2012 02:28:40 +0000 http://www.benstopford.com/?p=73#comment-2199 Hi Ben, It's really a good article. I am working on a securities trading system that we need to update multiple child caches inside a EntryProcessor using backing map, which is invoked against the parent cache for providing concurrency control. I realized that the change to parent cache will be rolled back by coherence automatically if any exception thrown inside EntryProcessor, but not the changes on those child caches made through backing map. I wonder should we really worry about exception in this case as it is just updating local JVM memory (ignoring extreme case such as sudden crash of JVM)? If exception should be considered, what I can think is to keep before-image at the start of EntryProcessor and rollback by coding if exception occur. Is there any better idea? Thanks a lot. Hi Ben,
It’s really a good article. I am working on a securities trading system that we need to update multiple child caches inside a EntryProcessor using backing map, which is invoked against the parent cache for providing concurrency control.

I realized that the change to parent cache will be rolled back by coherence automatically if any exception thrown inside EntryProcessor, but not the changes on those child caches made through backing map. I wonder should we really worry about exception in this case as it is just updating local JVM memory (ignoring extreme case such as sudden crash of JVM)?

If exception should be considered, what I can think is to keep before-image at the start of EntryProcessor and rollback by coding if exception occur. Is there any better idea?

Thanks a lot.

]]>