Terracotta Discussion Forums (LEGACY READ-ONLY ARCHIVE)

If I've created a queue which I've declared shared (root) which all my clusters will be reading objects instances from to process, will changes to the instances inside the queue read by each cluster also be replicated to the other clusters ? If so is there any way to disable this behavior ? All I want to implement is a single shared queue where each cluster removes objects from this queue and processes them independently.

If there is any documentation that expains these types of details please send me the link... I'm having a hard time finding any Documentation other than higher level overviews and basic samples.

thanks again,
Nick.

Hi,

I have a blog awhile back demonstrating this usecase. Take a look at the workLoad array in the Main class. In the CalculateNode.start(), I remove the element in the array and process it independently.

This app was written for the old config file format so you might run into config XML problem if you try to run it 'as is' with the current version.

http://blog.terracottatech.com/archive/2006/06/dso_mandelbrot.html

Hope that helps,

Hung-