# Gpgpu? **URL:** \ **Category:** Elasticsearch\ **Created:** [February 4, 2014, 4:39pm UTC](https://discuss.elastic.co/t/gpgpu/15593 "2014-02-04T16:39:46Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![depahelix\_2](https://avatars.discourse-cdn.com/v4/letter/d/5e9695/32.png) [@depahelix\_2](https://discuss.elastic.co/u/depahelix_2)\ **Post date:** [February 4, 2014, 4:39pm UTC](https://discuss.elastic.co/t/gpgpu/15593/1 "2014-02-04T16:39:46Z")
Is there anyway to make elasticsearch take advantage of GPGPU, when available? It would be nice to have some sort of plugin for this type of thing, in the future. See here: > **[General-purpose computing on graphics processing units](https://en.wikipedia.org/wiki/General-purpose_computing_on_graphics_processing_units)** > > General-purpose computing on graphics processing units (GPGPU, rarely GPGP) is the use of a graphics processing unit (GPU), which typically handles computation only for computer graphics, to perform computation in applications traditionally handled by the central processing unit (CPU). The use of multiple video cards in one computer, or large numbers of graphics chips, further parallelizes the already parallel nature of graphics processing. In addition, even a single GPU-CPU framew Essentially, a... -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). To view this discussion on the web visit [https://groups.google.com/d/msgid/elasticsearch/7c0c7add-3b61-4b72-a51d-150618a897ef%40googlegroups.com](https://groups.google.com/d/msgid/elasticsearch/7c0c7add-3b61-4b72-a51d-150618a897ef%40googlegroups.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![Tony\_Su](https://avatars.discourse-cdn.com/v4/letter/t/8dc957/32.png) [@Tony\_Su](https://discuss.elastic.co/u/Tony_Su)\ **Post date:** [February 4, 2014, 5:17pm UTC](https://discuss.elastic.co/t/gpgpu/15593/2 "2014-02-04T17:17:54Z")
🙂 I don't even know if there is a way to run a JRE on GPU computing (I haven't looked, either). My current exposure has generally given me the impression all code is moderately low level... C, with some scripting. But, if a JRE exists, then it's an interesting option considering the movement towards GPU computing for inexpensive massive computing for highly parallel tasks. Tony On Tuesday, February 4, 2014 8:39:46 AM UTC-8, depahelix wrote: > Is there anyway to make elasticsearch take advantage of GPGPU, when > available? It would be nice to have some sort of plugin for this type of > thing, in the future. > > See here: > > [General-purpose computing on graphics processing units - Wikipedia](http://en.wikipedia.org/wiki/General-purpose_computing_on_graphics_processing_units) -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). To view this discussion on the web visit [https://groups.google.com/d/msgid/elasticsearch/eece2db6-6aca-43fc-a701-e2a004239a3d%40googlegroups.com](https://groups.google.com/d/msgid/elasticsearch/eece2db6-6aca-43fc-a701-e2a004239a3d%40googlegroups.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![Ivan](https://avatars.discourse-cdn.com/v4/letter/i/df788c/32.png) [@Ivan](https://discuss.elastic.co/u/Ivan)\ **Post date:** [February 4, 2014, 5:28pm UTC](https://discuss.elastic.co/t/gpgpu/15593/3 "2014-02-04T17:28:14Z")
The JRE can access the GPU via libraries/bindings such as JOCL: > **[jocl.org - Java bindings for OpenCL](http://www.jocl.org/)** > > This site contains Java bindings for OpenCL That said, elasticsearch can be thought of as a wrapper around Lucene. It would make sense for the Lucene layer to take advantage of the GPU. Also, for the most part, elasticsearch tends to be more memory and IO bound and not CPU. Cheers, Ivan On Tue, Feb 4, 2014 at 9:17 AM, Tony Su [tonysu999@gmail.com](mailto:tonysu999@gmail.com) wrote: > 🙂 > I don't even know if there is a way to run a JRE on GPU computing (I > haven't looked, either). > My current exposure has generally given me the impression all code is > moderately low level... C, with some scripting. > > But, if a JRE exists, then it's an interesting option considering the > movement towards GPU computing for inexpensive massive computing for highly > parallel tasks. > > Tony > > On Tuesday, February 4, 2014 8:39:46 AM UTC-8, depahelix wrote: > > > Is there anyway to make elasticsearch take advantage of GPGPU, when > > available? It would be nice to have some sort of plugin for this type of > > thing, in the future. > > > > See here: > > [http://en.wikipedia.org/wiki/General-purpose\_computing\_on\_](http://en.wikipedia.org/wiki/General-purpose_computing_on_) > > graphics\_processing\_units > > > > -- > > You received this message because you are subscribed to the Google Groups > > "elasticsearch" group. > > To unsubscribe from this group and stop receiving emails from it, send an > > email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). > > To view this discussion on the web visit > > [https://groups.google.com/d/msgid/elasticsearch/eece2db6-6aca-43fc-a701-e2a004239a3d%40googlegroups.com](https://groups.google.com/d/msgid/elasticsearch/eece2db6-6aca-43fc-a701-e2a004239a3d%40googlegroups.com) > > . > > For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). To view this discussion on the web visit [https://groups.google.com/d/msgid/elasticsearch/CALY%3DcQCt47Rvg-2uM%3DyP%2BM8-vuJAyKCrToReQpLnyYpOxE6R%3DQ%40mail.gmail.com](https://groups.google.com/d/msgid/elasticsearch/CALY%3DcQCt47Rvg-2uM%3DyP%2BM8-vuJAyKCrToReQpLnyYpOxE6R%3DQ%40mail.gmail.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![Tony\_Su](https://avatars.discourse-cdn.com/v4/letter/t/8dc957/32.png) [@Tony\_Su](https://discuss.elastic.co/u/Tony_Su)\ **Post date:** [February 4, 2014, 5:45pm UTC](https://discuss.elastic.co/t/gpgpu/15593/4 "2014-02-04T17:45:24Z")
Hmmmm.... Looks interesting although of course for the moment is a very narrow implementation but may pave the way for more. At first glance it sounds like one could write a CUDA app in OpenCL and access the JRE installed and running on the CPU (not GPU). If I understand that correctly, then it might warrant inspection, a close look at what is running where. The link to aparapi in your reference also looks interesting although I'm not too excited about anything that waits until runtime to do the byte translation. Tony On Tuesday, February 4, 2014 9:28:14 AM UTC-8, Ivan Brusic wrote: > The JRE can access the GPU via libraries/bindings such as JOCL: > [http://www.jocl.org/](http://www.jocl.org/) > > That said, elasticsearch can be thought of as a wrapper around Lucene. It > would make sense for the Lucene layer to take advantage of the GPU. Also, > for the most part, elasticsearch tends to be more memory and IO bound and > not CPU. > > Cheers, > > Ivan > > On Tue, Feb 4, 2014 at 9:17 AM, Tony Su \<[tony...@gmail.com](mailto:tony...@gmail.com) \\>wrote: > > > 🙂 > > I don't even know if there is a way to run a JRE on GPU computing (I > > haven't looked, either). > > My current exposure has generally given me the impression all code is > > moderately low level... C, with some scripting. > > > > But, if a JRE exists, then it's an interesting option considering the > > movement towards GPU computing for inexpensive massive computing for highly > > parallel tasks. > > > > Tony > > > > On Tuesday, February 4, 2014 8:39:46 AM UTC-8, depahelix wrote: > > > > > Is there anyway to make elasticsearch take advantage of GPGPU, when > > > available? It would be nice to have some sort of plugin for this type of > > > thing, in the future. > > > > > > See here: > > > [http://en.wikipedia.org/wiki/General-purpose\_computing\_on\_](http://en.wikipedia.org/wiki/General-purpose_computing_on_) > > > graphics\_processing\_units > > > > > > -- > > > You received this message because you are subscribed to the Google Groups > > > "elasticsearch" group. > > > To unsubscribe from this group and stop receiving emails from it, send an > > > email to [elasticsearc...@googlegroups.com](mailto:elasticsearc...@googlegroups.com) \. > > > To view this discussion on the web visit > > > [https://groups.google.com/d/msgid/elasticsearch/eece2db6-6aca-43fc-a701-e2a004239a3d%40googlegroups.com](https://groups.google.com/d/msgid/elasticsearch/eece2db6-6aca-43fc-a701-e2a004239a3d%40googlegroups.com) > > > . > > > > For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). To view this discussion on the web visit [https://groups.google.com/d/msgid/elasticsearch/c37457f9-b652-4b18-bcf8-5cc54ba18609%40googlegroups.com](https://groups.google.com/d/msgid/elasticsearch/c37457f9-b652-4b18-bcf8-5cc54ba18609%40googlegroups.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [July 6, 2017, 1:52am UTC](https://discuss.elastic.co/t/gpgpu/15593/5 "2017-07-06T01:52:40Z")
Closed on July 6, 2017, 1:52am UTC