JPPF Issue Tracker
star_faded.png
Please log in to bookmark issues
feature_request_small.png
CLOSED  Feature request JPPF-12  -  Using GPU performance
Posted Aug 08, 2012 - updated Mar 17, 2013
action_vote_minus_faded.png
0
Votes
action_vote_plus_faded.png
icon_info.png This issue has been closed with status "Closed" and resolution "RESOLVED".
Issue details
  • Type of issue
    Feature request
  • Status
     
    Closed
  • Assigned to
     lolo4j
  • Progress
       
  • Type of bug
    Not triaged
  • Likelihood
    Not triaged
  • Effect
    Not triaged
  • Posted by
     lolo4j
  • Owned by
    Not owned by anyone
  • Category
    Integration
  • Resolution
    RESOLVED
  • Priority
    High
  • Targetted for
    icon_milestones.png JPPF 3.3
Issue description
* Created by SF.net user unaimatakenny *

Processing power of GPUs for parallel processing of tasks by the nodes. The problems that we can have is, that each of the companies that manufacture graphics cards has its own API. NVIDEA has its own support for Java, called JCUDA, may be a possible starting point. I not that very familiar with this. But the idea is that tasks are distributed by each of the cores of GPUs. I must admit that this is not a simple task, but once seen JPPF scalability and the ease of use, it would be nice to see in the future this feature implemented.

#1
Comment posted by
 lolo4j
Aug 13, 06:59
I just came across an article on Slashdot about project rootbeer1.

From the README on their home page:

The Rootbeer GPU Compiler makes it easy to use Graphics Processing Units from within Java.
Rootbeer is more advanced than CUDA or OpenCL Java Language Bindings. With bindings the developer must serialize complex graphs of objects into arrays of primitive types. With Rootbeer this is done automatically.
Also with language bindings, the developer must write the GPU kernel in CUDA or OpenCL. With Rootbeer a static analysis of the Java Bytecode is done (using Soot) and CUDA code is automatically generated.
This is something we should definitely look into for harnessing GPU capabilities from JPPF.

Additionally the project is now licensed under the MIT license terms, which is defintily compatible with Apache 2.0 (it used to be under the GPL)
#6
Comment posted by
 lolo4j
Mar 13, 10:14
A few links from this forum post:

#8
Comment posted by
 lolo4j
Mar 17, 12:03
This is now implemented as a new sample in the samples pack. CHnages committed to SVN reviison 2657

The issue was updated with the following change(s):
  • This issue has been closed
  • The status has been updated, from Confirmed to Closed.
  • This issue's progression has been updated to 100 percent completed.
  • The resolution has been updated, from Not determined to RESOLVED.
  • Information about the user working on this issue has been changed, from lolo4j to Not being worked on.