Cpre584 - User contributions [en]

Team Challenger

2013-03-12T17:13:27Z

Zhonghu: /* Wiki Contributions */

== Team Members ==

* Xinying Wang
* Qilin Li
* Zhong Hu

== Wiki Contributions ==

* Xinying Wang
:* week 1
::* Convey vector personalities offer OpenMP-like programming approach with FPGA accelerating. [http://www.fpl2012.org/Presentations/W4B2.pdf]
::* Weekly presentation slides [[Media:week1slides.pptx]]
:* week 2
::* A Sparse Matrix Personality for the Convey HC-1 [http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5771239&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D5771239]
:* week 3
::* Translate verilog version adder file to vhdl code.
::* Discuss the implementation of QR application on convey system
::* Cordic Algorithm Implementations on FPGA [http://cutler.eecs.berkeley.edu/classes/ee225c/Papers/cordic.pdf]
:* week 4
::* Convey computing with bioinformatics applications [http://hpcsociety.org/Resources/Documents/121212Kirby-CONVEY-SHPCP_121212.pdf]
:* week 5
::* Tutorial for COREGEN [http://homepages.cae.wisc.edu/~ece554/website/Xilinx/Coregen_user_guide.pdf]
* Qilin Li
:* week 1
::* Introduction to Compilers for Convey [http://www.conveycomputer.com/files/3913/5085/4426/Compiler_Data_Sheet.pdf]
::* GPGPU Programming on example of CUDA [http://panoramix.ift.uni.wroc.pl/~maq/cuda/prezentacja-cuda-eng.pdf]
::* Parallel Programming in CUDA C [http://developer.download.nvidia.com/books/cuda-by-example/cuda-by-example-sample.pdf]
:* week 2
::* [http://www.cs.colby.edu/maxwell/courses/tutorials/maketutor/ A Simple Makefile Tutorial]
:* week 3
::* [http://newbiedoc.sourceforge.net/text_editing/vim.html#MODES-CL A Simple VIM Tutorial]
:* week 4
::* [[Media:ConveySystemAdministrationGuide.pdf | Convey System Administration Guide]]
* Zhong Hu
::* CUDA Global Memory Usage & Strategy. [http://developer.download.nvidia.com/CUDA/training/cuda_webinars_GlobalMemory.pdf]
::* CUDA C++ code samples.[http://docs.nvidia.com/cuda/cuda-samples/index.html]
:* make-up for week 2
::* QR ecomposition on GPUs[http://dl.acm.org/citation.cfm?id=1513904]
::* Introduction to Householder algorithm in QR decomposition application[http://math.fullerton.edu/mathews/n2003/householdermod.html]
:* make-up for week 3
::* CUDA Memory Model[https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rja&ved=0CEoQFjAG&url=https%3A%2F%2Fhub.vscse.org%2Fsite%2Fresources%2F2010%2F06%2F00048%2FCUDA_04.pdf&ei=_dwZUY73NKji2gWqo4GYDQ&usg=AFQjCNGgxpCFWYepx6v32fTHNyRI1pOOtA&sig2=6TW0elRmtCKW3i_fAqAT_g&bvm=bv.42261806,d.b2I]
::* CUBLAS Library Usage Reference[https://www.docs.nvidia.com/cuda/pdf/CUDA_CUBLAS_Users_Guide.pdf]
:* week 4
::* system Verilog tutorial[http://www.asic-world.com/systemverilog/tutorial.html]
:* week 5
::* most commonly used CUDA wiki website[https://developer.nvidia.com/category/zone/cuda-zone]
::* Nvidia GPU versions and corresponding compute capability[http://en.wikipedia.org/wiki/CUDA#Version_features_and_specifications]
::* methods to check CUDA memory constraints in terms of different GPU versions[http://3dgep.com/?p=1913]
:* week 6
::* open source software lists for genome alignment and assembly[http://seqanswers.com/wiki/Software/list].

Team Challenger

2013-02-19T17:30:10Z

Zhonghu: /* Wiki Contributions */

== Team Members ==

* Xinying Wang
* Qilin Li
* Zhong Hu

== Wiki Contributions ==

* Xinying Wang
:* week 1
::* Convey vector personalities offer OpenMP-like programming approach with FPGA accelerating. [http://www.fpl2012.org/Presentations/W4B2.pdf]
::* Weekly presentation slides [[Media:week1slides.pptx]]
:* week 2
::* A Sparse Matrix Personality for the Convey HC-1 [http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5771239&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D5771239]
:* week 3
::* Translate verilog version adder file to vhdl code.
::* Discuss the implementation of QR application on convey system
::* Cordic Algorithm Implementations on FPGA [http://cutler.eecs.berkeley.edu/classes/ee225c/Papers/cordic.pdf]
:* week 4
::* Convey computing with bioinformatics applications [http://hpcsociety.org/Resources/Documents/121212Kirby-CONVEY-SHPCP_121212.pdf]
* Qilin Li
:* week 1
::* Introduction to Compilers for Convey [http://www.conveycomputer.com/files/3913/5085/4426/Compiler_Data_Sheet.pdf]
::* GPGPU Programming on example of CUDA [http://panoramix.ift.uni.wroc.pl/~maq/cuda/prezentacja-cuda-eng.pdf]
::* Parallel Programming in CUDA C [http://developer.download.nvidia.com/books/cuda-by-example/cuda-by-example-sample.pdf]
:* week 2
::* [http://www.cs.colby.edu/maxwell/courses/tutorials/maketutor/ A Simple Makefile Tutorial]
:* week 3
::* [http://newbiedoc.sourceforge.net/text_editing/vim.html#MODES-CL A Simple VIM Tutorial]
:* week 4
::* [[Media:ConveySystemAdministrationGuide.pdf | Convey System Administration Guide]]
* Zhong Hu
::* CUDA Global Memory Usage & Strategy. [http://developer.download.nvidia.com/CUDA/training/cuda_webinars_GlobalMemory.pdf]
::* CUDA C++ code samples.[http://docs.nvidia.com/cuda/cuda-samples/index.html]
:* make-up for week 2
::* QR ecomposition on GPUs[http://dl.acm.org/citation.cfm?id=1513904]
::* Introduction to Householder algorithm in QR decomposition application[http://math.fullerton.edu/mathews/n2003/householdermod.html]
:* make-up for week 3
::* CUDA Memory Model[https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rja&ved=0CEoQFjAG&url=https%3A%2F%2Fhub.vscse.org%2Fsite%2Fresources%2F2010%2F06%2F00048%2FCUDA_04.pdf&ei=_dwZUY73NKji2gWqo4GYDQ&usg=AFQjCNGgxpCFWYepx6v32fTHNyRI1pOOtA&sig2=6TW0elRmtCKW3i_fAqAT_g&bvm=bv.42261806,d.b2I]
::* CUBLAS Library Usage Reference[https://www.docs.nvidia.com/cuda/pdf/CUDA_CUBLAS_Users_Guide.pdf]
:* week 4
::* system Verilog tutorial[http://www.asic-world.com/systemverilog/tutorial.html]
:* week 5
::* most commonly used CUDA wiki website[https://developer.nvidia.com/category/zone/cuda-zone]
::* Nvidia GPU versions and corresponding compute capability[http://en.wikipedia.org/wiki/CUDA#Version_features_and_specifications]
::* methods to check CUDA memory constraints in terms of different GPU versions[http://3dgep.com/?p=1913]

Team Challenger

2013-02-19T16:19:54Z

Zhonghu:

== Team Members ==

* Xinying Wang
* Qilin Li
* Zhong Hu

== Wiki Contributions ==

* Xinying Wang
:* week 1
::* Convey vector personalities offer OpenMP-like programming approach with FPGA accelerating. [http://www.fpl2012.org/Presentations/W4B2.pdf]
::* Weekly presentation slides [[Media:week1slides.pptx]]
:* week 2
::* A Sparse Matrix Personality for the Convey HC-1 [http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5771239&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D5771239]
:* week 3
::* Translate verilog version adder file to vhdl code.
::* Discuss the implementation of QR application on convey system
::* Cordic Algorithm Implementations on FPGA [http://cutler.eecs.berkeley.edu/classes/ee225c/Papers/cordic.pdf]
:* week 4
::* Convey computing with bioinformatics applications [http://hpcsociety.org/Resources/Documents/121212Kirby-CONVEY-SHPCP_121212.pdf]
* Qilin Li
:* week 1
::* Introduction to Compilers for Convey [http://www.conveycomputer.com/files/3913/5085/4426/Compiler_Data_Sheet.pdf]
::* GPGPU Programming on example of CUDA [http://panoramix.ift.uni.wroc.pl/~maq/cuda/prezentacja-cuda-eng.pdf]
::* Parallel Programming in CUDA C [http://developer.download.nvidia.com/books/cuda-by-example/cuda-by-example-sample.pdf]
:* week 2
::* [http://www.cs.colby.edu/maxwell/courses/tutorials/maketutor/ A Simple Makefile Tutorial]
:* week 3
::* [http://newbiedoc.sourceforge.net/text_editing/vim.html#MODES-CL A Simple VIM Tutorial]
:* week 4
::* [[Media:ConveySystemAdministrationGuide.pdf | Convey System Administration Guide]]
* Zhong Hu
::* CUDA Global Memory Usage & Strategy. [http://developer.download.nvidia.com/CUDA/training/cuda_webinars_GlobalMemory.pdf]
::* CUDA C++ code samples.[http://docs.nvidia.com/cuda/cuda-samples/index.html]
:* make-up for week 2
::* QR ecomposition on GPUs[http://dl.acm.org/citation.cfm?id=1513904]
::* Introduction to Householder algorithm in QR decomposition application[http://math.fullerton.edu/mathews/n2003/householdermod.html]
:* make-up for week 3
::* CUDA Memory Model[https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rja&ved=0CEoQFjAG&url=https%3A%2F%2Fhub.vscse.org%2Fsite%2Fresources%2F2010%2F06%2F00048%2FCUDA_04.pdf&ei=_dwZUY73NKji2gWqo4GYDQ&usg=AFQjCNGgxpCFWYepx6v32fTHNyRI1pOOtA&sig2=6TW0elRmtCKW3i_fAqAT_g&bvm=bv.42261806,d.b2I]
::* CUBLAS Library Usage Reference[https://www.docs.nvidia.com/cuda/pdf/CUDA_CUBLAS_Users_Guide.pdf]
:* week 4
::* system Verilog tutorial[http://www.asic-world.com/systemverilog/tutorial.html]
:* week 5
::* most commonly used CUDA wiki website[https://developer.nvidia.com/category/zone/cuda-zone]
::* Nvidia GPU versions and corresponding compute capability[http://en.wikipedia.org/wiki/CUDA#Version_features_and_specifications]

Team Challenger

2013-02-12T15:31:27Z

Zhonghu: /* Wiki Contributions */

== Team Members ==

* Xinying Wang
* Qilin Li
* Zhong Hu

== Wiki Contributions ==

* Xinying Wang
:* week 1
::* Convey vector personalities offer OpenMP-like programming approach with FPGA accelerating. [http://www.fpl2012.org/Presentations/W4B2.pdf]
::* Weekly presentation slides [[Media:week1slides.pptx]]
:* week 2
::* A Sparse Matrix Personality for the Convey HC-1 [http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5771239&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D5771239]
:* week 3
::* Translate verilog version adder file to vhdl code.
::* Discuss the implementation of QR application on convey system
::* Cordic Algorithm Implementations on FPGA [http://cutler.eecs.berkeley.edu/classes/ee225c/Papers/cordic.pdf]
* Qilin Li
:* week 1
::* Introduction to Compilers for Convey [http://www.conveycomputer.com/files/3913/5085/4426/Compiler_Data_Sheet.pdf]
::* GPGPU Programming on example of CUDA [http://panoramix.ift.uni.wroc.pl/~maq/cuda/prezentacja-cuda-eng.pdf]
::* Parallel Programming in CUDA C [http://developer.download.nvidia.com/books/cuda-by-example/cuda-by-example-sample.pdf]
:* week 2
::* [http://www.cs.colby.edu/maxwell/courses/tutorials/maketutor/ A Simple Makefile Tutorial]
:* week 3
::* [http://newbiedoc.sourceforge.net/text_editing/vim.html#MODES-CL A Simple VIM Tutorial]
* Zhong Hu
::* CUDA Global Memory Usage & Strategy. [http://developer.download.nvidia.com/CUDA/training/cuda_webinars_GlobalMemory.pdf]
::* CUDA C++ code samples.[http://docs.nvidia.com/cuda/cuda-samples/index.html]
:* make-up for week 2
::* QR ecomposition on GPUs[http://dl.acm.org/citation.cfm?id=1513904]
::* Introduction to Householder algorithm in QR decomposition application[http://math.fullerton.edu/mathews/n2003/householdermod.html]
:* make-up for week 3
::* CUDA Memory Model[https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rja&ved=0CEoQFjAG&url=https%3A%2F%2Fhub.vscse.org%2Fsite%2Fresources%2F2010%2F06%2F00048%2FCUDA_04.pdf&ei=_dwZUY73NKji2gWqo4GYDQ&usg=AFQjCNGgxpCFWYepx6v32fTHNyRI1pOOtA&sig2=6TW0elRmtCKW3i_fAqAT_g&bvm=bv.42261806,d.b2I]
::* CUBLAS Library Usage Reference[https://www.docs.nvidia.com/cuda/pdf/CUDA_CUBLAS_Users_Guide.pdf]
:* week 4
::* system Verilog tutorial[http://www.asic-world.com/systemverilog/tutorial.html]

Team Challenger

2013-02-12T06:11:47Z

Zhonghu: /* Wiki Contributions */

== Team Members ==

* Xinying Wang
* Qilin Li
* Zhong Hu

== Wiki Contributions ==

* Xinying Wang
:* week 1
::* Convey vector personalities offer OpenMP-like programming approach with FPGA accelerating. [http://www.fpl2012.org/Presentations/W4B2.pdf]
::* Weekly presentation slides [[Media:week1slides.pptx]]
:* week 2
::* A Sparse Matrix Personality for the Convey HC-1 [http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5771239&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D5771239]
:* week 3
::* Translate verilog version adder file to vhdl code.
::* Discuss the implementation of QR application on convey system
::* Cordic Algorithm Implementations on FPGA [http://cutler.eecs.berkeley.edu/classes/ee225c/Papers/cordic.pdf]
* Qilin Li
:* week 1
::* Introduction to Compilers for Convey [http://www.conveycomputer.com/files/3913/5085/4426/Compiler_Data_Sheet.pdf]
::* GPGPU Programming on example of CUDA [http://panoramix.ift.uni.wroc.pl/~maq/cuda/prezentacja-cuda-eng.pdf]
::* Parallel Programming in CUDA C [http://developer.download.nvidia.com/books/cuda-by-example/cuda-by-example-sample.pdf]
:* week 2
::* [http://www.cs.colby.edu/maxwell/courses/tutorials/maketutor/ A Simple Makefile Tutorial]
:* week 3
::* [http://newbiedoc.sourceforge.net/text_editing/vim.html#MODES-CL A Simple VIM Tutorial]
* Zhong Hu
::* CUDA Global Memory Usage & Strategy. [http://developer.download.nvidia.com/CUDA/training/cuda_webinars_GlobalMemory.pdf]
::* CUDA C++ code samples.[http://docs.nvidia.com/cuda/cuda-samples/index.html]
:* make-up for week 2
::* QR ecomposition on GPUs[http://dl.acm.org/citation.cfm?id=1513904]
::* Introduction to Householder algorithm in QR decomposition application[http://math.fullerton.edu/mathews/n2003/householdermod.html]
:* make-up for week 3
::* CUDA Memory Model[https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rja&ved=0CEoQFjAG&url=https%3A%2F%2Fhub.vscse.org%2Fsite%2Fresources%2F2010%2F06%2F00048%2FCUDA_04.pdf&ei=_dwZUY73NKji2gWqo4GYDQ&usg=AFQjCNGgxpCFWYepx6v32fTHNyRI1pOOtA&sig2=6TW0elRmtCKW3i_fAqAT_g&bvm=bv.42261806,d.b2I]
::* CUBLAS Library Usage Reference[https://www.docs.nvidia.com/cuda/pdf/CUDA_CUBLAS_Users_Guide.pdf]

Team Challenger

2013-02-12T06:09:47Z

Zhonghu: /* Wiki Contributions */

== Team Members ==

* Xinying Wang
* Qilin Li
* Zhong Hu

== Wiki Contributions ==

* Xinying Wang
:* week 1
::* Convey vector personalities offer OpenMP-like programming approach with FPGA accelerating. [http://www.fpl2012.org/Presentations/W4B2.pdf]
::* Weekly presentation slides [[Media:week1slides.pptx]]
:* week 2
::* A Sparse Matrix Personality for the Convey HC-1 [http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5771239&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D5771239]
:* week 3
::* Translate verilog version adder file to vhdl code.
::* Discuss the implementation of QR application on convey system
::* Cordic Algorithm Implementations on FPGA [http://cutler.eecs.berkeley.edu/classes/ee225c/Papers/cordic.pdf]
* Qilin Li
:* week 1
::* Introduction to Compilers for Convey [http://www.conveycomputer.com/files/3913/5085/4426/Compiler_Data_Sheet.pdf]
::* GPGPU Programming on example of CUDA [http://panoramix.ift.uni.wroc.pl/~maq/cuda/prezentacja-cuda-eng.pdf]
::* Parallel Programming in CUDA C [http://developer.download.nvidia.com/books/cuda-by-example/cuda-by-example-sample.pdf]
:* week 2
::* [http://www.cs.colby.edu/maxwell/courses/tutorials/maketutor/ A Simple Makefile Tutorial]
:* week 3
::* [http://newbiedoc.sourceforge.net/text_editing/vim.html#MODES-CL A Simple VIM Tutorial]
* Zhong Hu
:* CUDA Global Memory Usage & Strategy. [http://developer.download.nvidia.com/CUDA/training/cuda_webinars_GlobalMemory.pdf]
:* CUDA C++ code samples.[http://docs.nvidia.com/cuda/cuda-samples/index.html]
* make-up for week 2
*QR ecomposition on GPUs[http://dl.acm.org/citation.cfm?id=1513904]
*Introduction to Householder algorithm in QR decomposition application[http://math.fullerton.edu/mathews/n2003/householdermod.html]
* make-up for week 3
*CUDA Memory Model[https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rja&ved=0CEoQFjAG&url=https%3A%2F%2Fhub.vscse.org%2Fsite%2Fresources%2F2010%2F06%2F00048%2FCUDA_04.pdf&ei=_dwZUY73NKji2gWqo4GYDQ&usg=AFQjCNGgxpCFWYepx6v32fTHNyRI1pOOtA&sig2=6TW0elRmtCKW3i_fAqAT_g&bvm=bv.42261806,d.b2I]
*CUBLAS Library Usage Reference[https://www.docs.nvidia.com/cuda/pdf/CUDA_CUBLAS_Users_Guide.pdf]

Team Challenger

2013-02-12T06:08:40Z

Zhonghu:

== Team Members ==

* Xinying Wang
* Qilin Li
* Zhong Hu

== Wiki Contributions ==

* Xinying Wang
:* week 1
::* Convey vector personalities offer OpenMP-like programming approach with FPGA accelerating. [http://www.fpl2012.org/Presentations/W4B2.pdf]
::* Weekly presentation slides [[Media:week1slides.pptx]]
:* week 2
::* A Sparse Matrix Personality for the Convey HC-1 [http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5771239&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D5771239]
:* week 3
::* Translate verilog version adder file to vhdl code.
::* Discuss the implementation of QR application on convey system
::* Cordic Algorithm Implementations on FPGA [http://cutler.eecs.berkeley.edu/classes/ee225c/Papers/cordic.pdf]
* Qilin Li
:* week 1
::* Introduction to Compilers for Convey [http://www.conveycomputer.com/files/3913/5085/4426/Compiler_Data_Sheet.pdf]
::* GPGPU Programming on example of CUDA [http://panoramix.ift.uni.wroc.pl/~maq/cuda/prezentacja-cuda-eng.pdf]
::* Parallel Programming in CUDA C [http://developer.download.nvidia.com/books/cuda-by-example/cuda-by-example-sample.pdf]
:* week 2
::* [http://www.cs.colby.edu/maxwell/courses/tutorials/maketutor/ A Simple Makefile Tutorial]
:* week 3
::* [http://newbiedoc.sourceforge.net/text_editing/vim.html#MODES-CL A Simple VIM Tutorial]
* Zhong Hu
:* CUDA Global Memory Usage & Strategy. [http://developer.download.nvidia.com/CUDA/training/cuda_webinars_GlobalMemory.pdf]
:* CUDA C++ code samples.[http://docs.nvidia.com/cuda/cuda-samples/index.html]
* make-up for week 2
*QR ecomposition on GPUs[http://dl.acm.org/citation.cfm?id=1513904]
*Introduction to Householder algorithm in QR decomposition application[http://math.fullerton.edu/mathews/n2003/householdermod.html]
* make-up for week 3
*CUDA Memory Model[https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rja&ved=0CEoQFjAG&url=https%3A%2F%2Fhub.vscse.org%2Fsite%2Fresources%2F2010%2F06%2F00048%2FCUDA_04.pdf&ei=_dwZUY73NKji2gWqo4GYDQ&usg=AFQjCNGgxpCFWYepx6v32fTHNyRI1pOOtA&sig2=6TW0elRmtCKW3i_fAqAT_g&bvm=bv.42261806,d.b2I]
*CUBLAS Library Usage Reference[docs.nvidia.com/cuda/pdf/CUDA_CUBLAS_Users_Guide.pdf]

Team Challenger

2013-01-22T22:20:58Z

Zhonghu: /* Wiki Contributions */

== Team Members ==

* Xinying Wang
* Qilin Li
* Zhong Hu

== Wiki Contributions ==

* Xinying Wang
:* Convey vector personalities offer OpenMP-like programming approach with FPGA accelerating. [http://www.fpl2012.org/Presentations/W4B2.pdf]
* Qilin Li
* Zhong Hu
:* CUDA Global Memory Usage & Strategy. [http://developer.download.nvidia.com/CUDA/training/cuda_webinars_GlobalMemory.pdf]
:* CUDA C++ code samples.[http://docs.nvidia.com/cuda/cuda-samples/index.html]

Team Challenger

2013-01-22T22:18:25Z

Zhonghu: /* Wiki Contributions */

Team Challenger

2013-01-22T22:16:57Z

Zhonghu: /* Wiki Contributions */

== Team Members ==

* Xinying Wang
* Qilin Li
* Zhong Hu

== Wiki Contributions ==

* Xinying Wang
:* Convey vector personalities offer OpenMP-like programming approach with FPGA accelerating. [http://www.fpl2012.org/Presentations/W4B2.pdf]
* Qilin Li
* Zhong Hu
:* CUDA Global Memory Usage & Strategy.[developer.download.nvidia.com/CUDA/training/cuda_webinars_GlobalMemory.pdf]