GPU - Revision history

⚠️Andrej domzale: /* Further reading */

2016-04-04T11:39:45Z

Neteler: /* Modules of interest to be parallelized */

2015-09-06T12:44:04Z

Modules of interest to be parallelized

← Older revision		Revision as of 12:44, 6 September 2015
Line 47:		Line 47:
	The target version will be GRASS 7 (alias SVN trunk).		The target version will be GRASS 7 (alias SVN trunk).

	* {{cmd\|v.in.ogr~~\|version=70~~}} or → <u>underlying vector library functions to build topology and spatial index</u> ←		* {{cmd\|v.in.ogr}} or → <u>underlying vector library functions to build topology and spatial index</u> ←
	* {{cmd\|v.surf.rst~~\|version=70~~}}		* {{cmd\|v.surf.rst}}
	* {{cmd\|v.vol.rst~~\|version=70~~}}		* {{cmd\|v.vol.rst}}
	** (''probably best to focus on the RST library first'')		** (''probably the best is to focus on the RST library first'')
	* {{cmd\|r.viewshed~~\|version=70~~}}		* {{cmd\|r.viewshed}}
	* {{cmd\|v.surf.bspline~~\|version=70~~}}		* {{cmd\|v.surf.bspline}}
	* {{cmd\|r.sun~~\|version=70~~}} (Seth added OpenCL support has part of his Google Summer of Code project)		* {{cmd\|r.sun}} (Seth added OpenCL support has part of his Google Summer of Code project)
	* {{cmd\|r.proj~~\|version=70~~}}		* {{cmd\|r.proj}}
	* {{cmd\|v.proj~~\|version=70~~}}		* {{cmd\|v.proj}}
	* {{cmd\|v.net~~\|version=70~~}}.* ???		* {{cmd\|v.net}}.* ???
	* raster library (typically I/O-bound)		* raster library (typically I/O-bound)
	* {{cmd\|i.rectify~~\|version=70~~}}		* {{cmd\|i.rectify}}
	* ...		* ...
	* <strike>{{cmd\|r.mapcalc\|version=70}}</strike> (already has pthreads support (but only for parsing!!); probably I/O-bound)		* <strike>{{cmd\|r.mapcalc\|version=70}}</strike> (already has pthreads support (but only for parsing!!); probably I/O-bound)

			Notes:
			** r.resamp.stats, r.resamp.filter and r.series should be readily parallelisable, but I/O is likely to be the bottleneck.
			** r.series has the advantage that the I/O is also parallelisable.

	[[Category: Hardware]]		[[Category: Hardware]]
	[[Category: Parallelization]]		[[Category: Parallelization]]

Neteler at 20:52, 2 August 2013

2013-08-02T20:52:31Z

← Older revision		Revision as of 20:52, 2 August 2013
Line 62:		Line 62:
	* <strike>{{cmd\|r.mapcalc\|version=70}}</strike> (already has pthreads support (but only for parsing!!); probably I/O-bound)		* <strike>{{cmd\|r.mapcalc\|version=70}}</strike> (already has pthreads support (but only for parsing!!); probably I/O-bound)

			[[Category: Hardware]]
	[[Category: Parallelization]]		[[Category: Parallelization]]
	~~<math>Insert formula here</math>~~

⚠️Msieczka: /* Implementation */

2013-07-02T19:17:06Z

Implementation

← Older revision		Revision as of 19:17, 2 July 2013
Line 6:		Line 6:
	--with-opencl		--with-opencl

	~~On Linux you'll~~ need ~~to use the~~ proprietary driver ~~from~~ nVidia, AMD (ATI)~~, or~~ Intel. Point --with-opencl-includes= to the directory ''above'' cl.h, and as long as libOpenCL.so is in the ldconfig search path you're ok (a symlink to /usr/local/lib might be needed). On Mac OSX OpenCL support is now built in, and the framework should be automatically detected when you use --with-opencl in the ./configure options.		You need a GPU with a proprietary driver with libOpenCL.so library and C header files to go along with it. On Linux, currently, only nVidia, AMD (ATI) and Intel meet this criteria.

			The Intel Ivy Bridge HD4000 driver only supplies OpenCL multi-CPU core support for Linux. The same chip has a GPU driver for MS Windows and (probably) for Mac OSX. On Linux + Intel graphics you need a Xeon chip for driver GPU support currently.

			Point --with-opencl-includes= to the directory ''above'' cl.h, and as long as libOpenCL.so is in the ldconfig search path you're ok (a symlink to /usr/local/lib might be needed).

			On Mac OSX OpenCL support is now built in, and the framework should be automatically detected when you use --with-opencl in the ./configure options.

			OpenCL allows to utilize any number of GPUs and CPUs at the same time, or to pick what's what's wanted from the available, but any way the code has to be designed for such selection specifically.

	Currently {{cmd\|r.sun}} is the first module with any {{wikipedia\|OpenCL}} support (see [[R.sun#Development\|here]], although integration of Seth's work into svn-trunk is not yet complete.		Currently {{cmd\|r.sun}} is the first module with any {{wikipedia\|OpenCL}} support (see [[R.sun#Development\|here]], although integration of Seth's work into svn-trunk is not yet complete.

Neteler: link to r.sun-openCL

2013-05-28T08:32:03Z

link to r.sun-openCL

← Older revision		Revision as of 08:32, 28 May 2013
Line 8:		Line 8:
	On Linux you'll need to use the proprietary driver from nVidia, AMD (ATI), or Intel. Point --with-opencl-includes= to the directory ''above'' cl.h, and as long as libOpenCL.so is in the ldconfig search path you're ok (a symlink to /usr/local/lib might be needed). On Mac OSX OpenCL support is now built in, and the framework should be automatically detected when you use --with-opencl in the ./configure options.		On Linux you'll need to use the proprietary driver from nVidia, AMD (ATI), or Intel. Point --with-opencl-includes= to the directory ''above'' cl.h, and as long as libOpenCL.so is in the ldconfig search path you're ok (a symlink to /usr/local/lib might be needed). On Mac OSX OpenCL support is now built in, and the framework should be automatically detected when you use --with-opencl in the ./configure options.

	Currently {{cmd\|r.sun}} is the first module with any {{wikipedia\|OpenCL}} support, although integration of Seth's work into svn-trunk is not yet complete.		Currently {{cmd\|r.sun}} is the first module with any {{wikipedia\|OpenCL}} support (see [[R.sun#Development\|here]], although integration of Seth's work into svn-trunk is not yet complete.

	== Discussion ==		== Discussion ==

⚠️HamishBowman: /* Modules of interest to be parallelized */

2013-04-30T08:02:06Z

Modules of interest to be parallelized

← Older revision		Revision as of 08:02, 30 April 2013
Line 45:		Line 45:
	* {{cmd\|r.viewshed\|version=70}}		* {{cmd\|r.viewshed\|version=70}}
	* {{cmd\|v.surf.bspline\|version=70}}		* {{cmd\|v.surf.bspline\|version=70}}
	* {{cmd\|r.sun\|version=70}} (Seth added support has part of his Google Summer of Code project)		* {{cmd\|r.sun\|version=70}} (Seth added OpenCL support has part of his Google Summer of Code project)
	* {{cmd\|r.proj\|version=70}}		* {{cmd\|r.proj\|version=70}}
	* {{cmd\|v.proj\|version=70}}		* {{cmd\|v.proj\|version=70}}

⚠️HamishBowman: /* Modules of interest to be parallelized */

2013-04-30T08:01:50Z

Modules of interest to be parallelized

← Older revision		Revision as of 08:01, 30 April 2013
Line 45:		Line 45:
	* {{cmd\|r.viewshed\|version=70}}		* {{cmd\|r.viewshed\|version=70}}
	* {{cmd\|v.surf.bspline\|version=70}}		* {{cmd\|v.surf.bspline\|version=70}}
	* {{cmd\|r.sun\|version=70}}		* {{cmd\|r.sun\|version=70}} (Seth added support has part of his Google Summer of Code project)
	* {{cmd\|r.proj\|version=70}}		* {{cmd\|r.proj\|version=70}}
	* {{cmd\|v.proj\|version=70}}		* {{cmd\|v.proj\|version=70}}

⚠️HamishBowman: howto

2013-04-30T08:00:58Z

howto

← Older revision		Revision as of 08:00, 30 April 2013
Line 1:		Line 1:
	__TOC__		__TOC__

			== Implementation ==

			In GRASS 7 you can ./configure GRASS with:
			--with-opencl

			On Linux you'll need to use the proprietary driver from nVidia, AMD (ATI), or Intel. Point --with-opencl-includes= to the directory ''above'' cl.h, and as long as libOpenCL.so is in the ldconfig search path you're ok (a symlink to /usr/local/lib might be needed). On Mac OSX OpenCL support is now built in, and the framework should be automatically detected when you use --with-opencl in the ./configure options.

			Currently {{cmd\|r.sun}} is the first module with any {{wikipedia\|OpenCL}} support, although integration of Seth's work into svn-trunk is not yet complete.

			== Discussion ==

	Comments from the mailing list concerning GRASS and GPU parallelization:		Comments from the mailing list concerning GRASS and GPU parallelization:

⚠️HamishBowman: ll

2013-04-30T07:55:48Z

← Older revision		Revision as of 07:55, 30 April 2013
Line 10:		Line 10:
	* As I understand it, CUDA is 100% dependent on the closed-source binary driver from nVidia and works on their video cards alone. Which is fine for today for people with nVidia hardware using their binary video card driver. If nVidia decides in a couple of years to stop supporting CUDA, your old card, your specific OS or distro, your OS or distro version+cpu type, or if they go out of business or are bought/sold to another company who is not interested, any code based on it becomes useless. For this reason code written for an open platform such as OpenCL, even if less advanced, seems to have a brighter long-term future. -- ''HB''		* As I understand it, CUDA is 100% dependent on the closed-source binary driver from nVidia and works on their video cards alone. Which is fine for today for people with nVidia hardware using their binary video card driver. If nVidia decides in a couple of years to stop supporting CUDA, your old card, your specific OS or distro, your OS or distro version+cpu type, or if they go out of business or are bought/sold to another company who is not interested, any code based on it becomes useless. For this reason code written for an open platform such as OpenCL, even if less advanced, seems to have a brighter long-term future. -- ''HB''

	* Support for double precision floating point values must be retained for calculations which deal with positional data. For elevation and radiometric data floating point precision may be enough.		* Support for double precision floating point values must be retained for calculations which deal with positional data (as sub-meter precision for lat/long exceeds single-precision floating poing). For elevation and radiometric data floating point precision may be enough.

	== Further reading ==		== Further reading ==

Neteler: r.mapcalc: already has pthreads support but only for parsing! i.rectify has been modernized. referring to G7 now

2013-01-30T20:15:39Z

r.mapcalc: already has pthreads support but only for parsing! i.rectify has been modernized. referring to G7 now

← Older revision		Revision as of 20:15, 30 January 2013
Line 28:		Line 28:
	The target version will be GRASS 7 (alias SVN trunk).		The target version will be GRASS 7 (alias SVN trunk).

	* {{cmd\|v.in.ogr}} or → <u>underlying vector library functions to build topology and spatial index</u> ←		* {{cmd\|v.in.ogr\|version=70}} or → <u>underlying vector library functions to build topology and spatial index</u> ←
	* {{cmd\|v.surf.rst}}		* {{cmd\|v.surf.rst\|version=70}}
	* {{cmd\|v.vol.rst}}		* {{cmd\|v.vol.rst\|version=70}}
	** (''probably best to focus on the RST library first'')		** (''probably best to focus on the RST library first'')
	* {{~~AddonCmd~~\|r.viewshed}}		* {{cmd\|r.viewshed\|version=70}}
	* {{cmd\|v.surf.bspline}}		* {{cmd\|v.surf.bspline\|version=70}}
	* {{cmd\|r.sun}}		* {{cmd\|r.sun\|version=70}}
	* {{cmd\|r.proj}}		* {{cmd\|r.proj\|version=70}}
	* {{cmd\|v.proj}}		* {{cmd\|v.proj\|version=70}}
	* {{cmd\|v.net}}.* ???		* {{cmd\|v.net\|version=70}}.* ???
	* raster library (typically I/O-bound)		* raster library (typically I/O-bound)
			* {{cmd\|i.rectify\|version=70}}
	* ...		* ...
	* <strike>{{cmd\|r.mapcalc}}</strike> (already has pthreads support; probably I/O-bound)		* <strike>{{cmd\|r.mapcalc\|version=70}}</strike> (already has pthreads support (but only for parsing!!); probably I/O-bound)
	* <strike>{{cmd\|r.los}}</strike> (to be replaced by r.viewshed after last few bugs are fixed)
	* <strike>{{cmd\|i.rectify}}</strike> (internal code to be replaced by GDALwarp API; note that for the polynomial calcs it's partly the same exact code)


	[[Category: Parallelization]]		[[Category: Parallelization]]
	<math>Insert formula here</math>		<math>Insert formula here</math>

← Older revision		Revision as of 11:39, 4 April 2016
Line 42:		Line 42:

	* OpenCL podcasts: http://www.macresearch.org/opencl		* OpenCL podcasts: http://www.macresearch.org/opencl

			* Parallel GRASS GIS modules for viewshed and Fresnel analysis running on CUDA GPU, test project, http://s51mo.net/fresnel/

	== Modules of interest to be parallelized ==		== Modules of interest to be parallelized ==

← Older revision		Revision as of 12:44, 6 September 2015
Line 47:		Line 47:
	The target version will be GRASS 7 (alias SVN trunk).		The target version will be GRASS 7 (alias SVN trunk).

	* {{cmd\|v.in.ogr~~\|version=70~~}} or → <u>underlying vector library functions to build topology and spatial index</u> ←		* {{cmd\|v.in.ogr}} or → <u>underlying vector library functions to build topology and spatial index</u> ←
	* {{cmd\|v.surf.rst~~\|version=70~~}}		* {{cmd\|v.surf.rst}}
	* {{cmd\|v.vol.rst~~\|version=70~~}}		* {{cmd\|v.vol.rst}}
	** (''probably best to focus on the RST library first'')		** (''probably the best is to focus on the RST library first'')
	* {{cmd\|r.viewshed~~\|version=70~~}}		* {{cmd\|r.viewshed}}
	* {{cmd\|v.surf.bspline~~\|version=70~~}}		* {{cmd\|v.surf.bspline}}
	* {{cmd\|r.sun~~\|version=70~~}} (Seth added OpenCL support has part of his Google Summer of Code project)		* {{cmd\|r.sun}} (Seth added OpenCL support has part of his Google Summer of Code project)
	* {{cmd\|r.proj~~\|version=70~~}}		* {{cmd\|r.proj}}
	* {{cmd\|v.proj~~\|version=70~~}}		* {{cmd\|v.proj}}
	* {{cmd\|v.net~~\|version=70~~}}.* ???		* {{cmd\|v.net}}.* ???
	* raster library (typically I/O-bound)		* raster library (typically I/O-bound)
	* {{cmd\|i.rectify~~\|version=70~~}}		* {{cmd\|i.rectify}}
	* ...		* ...
	* <strike>{{cmd\|r.mapcalc\|version=70}}</strike> (already has pthreads support (but only for parsing!!); probably I/O-bound)		* <strike>{{cmd\|r.mapcalc\|version=70}}</strike> (already has pthreads support (but only for parsing!!); probably I/O-bound)

			Notes:
			** r.resamp.stats, r.resamp.filter and r.series should be readily parallelisable, but I/O is likely to be the bottleneck.
			** r.series has the advantage that the I/O is also parallelisable.

	[[Category: Hardware]]		[[Category: Hardware]]
	[[Category: Parallelization]]		[[Category: Parallelization]]

← Older revision		Revision as of 08:02, 30 April 2013
Line 45:		Line 45:
	* {{cmd\|r.viewshed\|version=70}}		* {{cmd\|r.viewshed\|version=70}}
	* {{cmd\|v.surf.bspline\|version=70}}		* {{cmd\|v.surf.bspline\|version=70}}
	* {{cmd\|r.sun\|version=70}} (Seth added support has part of his Google Summer of Code project)		* {{cmd\|r.sun\|version=70}} (Seth added OpenCL support has part of his Google Summer of Code project)
	* {{cmd\|r.proj\|version=70}}		* {{cmd\|r.proj\|version=70}}
	* {{cmd\|v.proj\|version=70}}		* {{cmd\|v.proj\|version=70}}

← Older revision		Revision as of 08:01, 30 April 2013
Line 45:		Line 45:
	* {{cmd\|r.viewshed\|version=70}}		* {{cmd\|r.viewshed\|version=70}}
	* {{cmd\|v.surf.bspline\|version=70}}		* {{cmd\|v.surf.bspline\|version=70}}
	* {{cmd\|r.sun\|version=70}}		* {{cmd\|r.sun\|version=70}} (Seth added support has part of his Google Summer of Code project)
	* {{cmd\|r.proj\|version=70}}		* {{cmd\|r.proj\|version=70}}
	* {{cmd\|v.proj\|version=70}}		* {{cmd\|v.proj\|version=70}}

← Older revision		Revision as of 20:15, 30 January 2013
Line 28:		Line 28:
	The target version will be GRASS 7 (alias SVN trunk).		The target version will be GRASS 7 (alias SVN trunk).

	* {{cmd\|v.in.ogr}} or → <u>underlying vector library functions to build topology and spatial index</u> ←		* {{cmd\|v.in.ogr\|version=70}} or → <u>underlying vector library functions to build topology and spatial index</u> ←
	* {{cmd\|v.surf.rst}}		* {{cmd\|v.surf.rst\|version=70}}
	* {{cmd\|v.vol.rst}}		* {{cmd\|v.vol.rst\|version=70}}
	** (''probably best to focus on the RST library first'')		** (''probably best to focus on the RST library first'')
	* {{~~AddonCmd~~\|r.viewshed}}		* {{cmd\|r.viewshed\|version=70}}
	* {{cmd\|v.surf.bspline}}		* {{cmd\|v.surf.bspline\|version=70}}
	* {{cmd\|r.sun}}		* {{cmd\|r.sun\|version=70}}
	* {{cmd\|r.proj}}		* {{cmd\|r.proj\|version=70}}
	* {{cmd\|v.proj}}		* {{cmd\|v.proj\|version=70}}
	* {{cmd\|v.net}}.* ???		* {{cmd\|v.net\|version=70}}.* ???
	* raster library (typically I/O-bound)		* raster library (typically I/O-bound)
			* {{cmd\|i.rectify\|version=70}}
	* ...		* ...
	* <strike>{{cmd\|r.mapcalc}}</strike> (already has pthreads support; probably I/O-bound)		* <strike>{{cmd\|r.mapcalc\|version=70}}</strike> (already has pthreads support (but only for parsing!!); probably I/O-bound)
	* <strike>{{cmd\|r.los}}</strike> (to be replaced by r.viewshed after last few bugs are fixed)
	* <strike>{{cmd\|i.rectify}}</strike> (internal code to be replaced by GDALwarp API; note that for the polynomial calcs it's partly the same exact code)


	[[Category: Parallelization]]		[[Category: Parallelization]]
	<math>Insert formula here</math>		<math>Insert formula here</math>