ºÝºÝߣshows by User: jefflarkin / http://www.slideshare.net/images/logo.gif ºÝºÝߣshows by User: jefflarkin / Thu, 17 Dec 2020 16:11:30 GMT ºÝºÝߣShare feed for ºÝºÝߣshows by User: jefflarkin Best Practices for OpenMP on GPUs - OpenMP UK Users Group /slideshow/best-practices-for-openmp-on-gpus-openmp-uk-users-group/240235084 openmpongpusbestpractices-201217161130
In this talk I discuss some best practices for writing OpenMP for NVIDIA GPUs. Video of this presentation can be found here: https://www.youtube.com/watch?v=9w_2tj2uD4M]]>
In this talk I discuss some best practices for writing OpenMP for NVIDIA GPUs. Video of this presentation can be found here: https://www.youtube.com/watch?v=9w_2tj2uD4M]]> Thu, 17 Dec 2020 16:11:30 GMT /slideshow/best-practices-for-openmp-on-gpus-openmp-uk-users-group/240235084 jefflarkin@slideshare.net(jefflarkin) Best Practices for OpenMP on GPUs - OpenMP UK Users Group jefflarkin In this talk I discuss some best practices for writing OpenMP for NVIDIA GPUs. Video of this presentation can be found here: https://www.youtube.com/watch?v=9w_2tj2uD4M <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/openmpongpusbestpractices-201217161130-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> In this talk I discuss some best practices for writing OpenMP for NVIDIA GPUs. Video of this presentation can be found here: https://www.youtube.com/watch?v=9w_2tj2uD4M
Best Practices for OpenMP on GPUs - OpenMP UK Users Group from Jeff Larkin
]]> 172 0 https://cdn.slidesharecdn.com/ss_thumbnails/openmpongpusbestpractices-201217161130-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 FortranCon2020: Highly Parallel Fortran and OpenACC Directives /jefflarkin/fortrancon2020-highly-parallel-fortran-and-openacc-directives fortranconopenaccjefflarkin-200714135353
Fortran has long been the language of computational math and science and it has outlived many of the computer architectures on which it has been used. Modern Fortran must be able to run on modern, highly parallel, heterogeneous computer architectures. A significant number of Fortran programmers have had success programming for heterogeneous machines by pairing Fortran with the OpenACC language for directives-based parallel programming. This includes some of the most widely-used Fortran applications in the world, such as VASP and Gaussian. This presentation will discuss what makes OpenACC a good fit for Fortran programmers and what the OpenACC language is doing to promote the use of native language parallelism in Fortran, such as do concurrent and Co-arrays. Video Recording: https://www.youtube.com/watch?v=OXZ_Wkae63Y]]>
Fortran has long been the language of computational math and science and it has outlived many of the computer architectures on which it has been used. Modern Fortran must be able to run on modern, highly parallel, heterogeneous computer architectures. A significant number of Fortran programmers have had success programming for heterogeneous machines by pairing Fortran with the OpenACC language for directives-based parallel programming. This includes some of the most widely-used Fortran applications in the world, such as VASP and Gaussian. This presentation will discuss what makes OpenACC a good fit for Fortran programmers and what the OpenACC language is doing to promote the use of native language parallelism in Fortran, such as do concurrent and Co-arrays. Video Recording: https://www.youtube.com/watch?v=OXZ_Wkae63Y]]> Tue, 14 Jul 2020 13:53:53 GMT /jefflarkin/fortrancon2020-highly-parallel-fortran-and-openacc-directives jefflarkin@slideshare.net(jefflarkin) FortranCon2020: Highly Parallel Fortran and OpenACC Directives jefflarkin Fortran has long been the language of computational math and science and it has outlived many of the computer architectures on which it has been used. Modern Fortran must be able to run on modern, highly parallel, heterogeneous computer architectures. A significant number of Fortran programmers have had success programming for heterogeneous machines by pairing Fortran with the OpenACC language for directives-based parallel programming. This includes some of the most widely-used Fortran applications in the world, such as VASP and Gaussian. This presentation will discuss what makes OpenACC a good fit for Fortran programmers and what the OpenACC language is doing to promote the use of native language parallelism in Fortran, such as do concurrent and Co-arrays. Video Recording: https://www.youtube.com/watch?v=OXZ_Wkae63Y <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/fortranconopenaccjefflarkin-200714135353-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> Fortran has long been the language of computational math and science and it has outlived many of the computer architectures on which it has been used. Modern Fortran must be able to run on modern, highly parallel, heterogeneous computer architectures. A significant number of Fortran programmers have had success programming for heterogeneous machines by pairing Fortran with the OpenACC language for directives-based parallel programming. This includes some of the most widely-used Fortran applications in the world, such as VASP and Gaussian. This presentation will discuss what makes OpenACC a good fit for Fortran programmers and what the OpenACC language is doing to promote the use of native language parallelism in Fortran, such as do concurrent and Co-arrays. Video Recording: https://www.youtube.com/watch?v=OXZ_Wkae63Y
FortranCon2020: Highly Parallel Fortran and OpenACC Directives from Jeff Larkin
]]> 406 0 https://cdn.slidesharecdn.com/ss_thumbnails/fortranconopenaccjefflarkin-200714135353-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Early Results of OpenMP 4.5 Portability on NVIDIA GPUs & CPUs /slideshow/early-results-of-openmp-45-portability-on-nvidia-gpus-cpus/79216550 2larkindoeportability-170828134704
This talk was presented at the DOE Centers of Excellence Performance Portability Workshop in August 2017. In this talk I explore the current status of 4 OpenMP 4.5 compilers for NVIDIA GPUs and CPUs from the perspective of performance portability between compilers and between the GPU and CPU.]]>
This talk was presented at the DOE Centers of Excellence Performance Portability Workshop in August 2017. In this talk I explore the current status of 4 OpenMP 4.5 compilers for NVIDIA GPUs and CPUs from the perspective of performance portability between compilers and between the GPU and CPU.]]> Mon, 28 Aug 2017 13:47:04 GMT /slideshow/early-results-of-openmp-45-portability-on-nvidia-gpus-cpus/79216550 jefflarkin@slideshare.net(jefflarkin) Early Results of OpenMP 4.5 Portability on NVIDIA GPUs & CPUs jefflarkin This talk was presented at the DOE Centers of Excellence Performance Portability Workshop in August 2017. In this talk I explore the current status of 4 OpenMP 4.5 compilers for NVIDIA GPUs and CPUs from the perspective of performance portability between compilers and between the GPU and CPU. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/2larkindoeportability-170828134704-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> This talk was presented at the DOE Centers of Excellence Performance Portability Workshop in August 2017. In this talk I explore the current status of 4 OpenMP 4.5 compilers for NVIDIA GPUs and CPUs from the perspective of performance portability between compilers and between the GPU and CPU.
Early Results of OpenMP 4.5 Portability on NVIDIA GPUs & CPUs from Jeff Larkin
]]> 3756 3 https://cdn.slidesharecdn.com/ss_thumbnails/2larkindoeportability-170828134704-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Performance Portability Through Descriptive Parallelism /slideshow/performance-portability-through-descriptive-parallelism/61213092 larkindoeportabilityworkshop-160421230747
This is a talk from the 2016 DOE Performance Portability workshop in Glendale AZ. The purpose of this talk is to explain the concept of descriptive parallel programming and why it is one way to provide performance portability to a variety of parallel architectures.]]>
This is a talk from the 2016 DOE Performance Portability workshop in Glendale AZ. The purpose of this talk is to explain the concept of descriptive parallel programming and why it is one way to provide performance portability to a variety of parallel architectures.]]> Thu, 21 Apr 2016 23:07:47 GMT /slideshow/performance-portability-through-descriptive-parallelism/61213092 jefflarkin@slideshare.net(jefflarkin) Performance Portability Through Descriptive Parallelism jefflarkin This is a talk from the 2016 DOE Performance Portability workshop in Glendale AZ. The purpose of this talk is to explain the concept of descriptive parallel programming and why it is one way to provide performance portability to a variety of parallel architectures. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/larkindoeportabilityworkshop-160421230747-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> This is a talk from the 2016 DOE Performance Portability workshop in Glendale AZ. The purpose of this talk is to explain the concept of descriptive parallel programming and why it is one way to provide performance portability to a variety of parallel architectures.
Performance Portability Through Descriptive Parallelism from Jeff Larkin
]]> 659 7 https://cdn.slidesharecdn.com/ss_thumbnails/larkindoeportabilityworkshop-160421230747-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 GTC16 - S6510 - Targeting GPUs with OpenMP 4.5 /slideshow/gtc16-s6510-targeting-gpus-with-openmp-45/60869651 s6510targetinggpuswithopenmp45-160413144408
These slides are from an instructor-led tutorial from GTC16. The talk discusses using a pre-release version of CLANG with support for OpenMP offloading directives to NVIDIA GPUs to experiement with OpenMP 4.5 target directives. ]]>
These slides are from an instructor-led tutorial from GTC16. The talk discusses using a pre-release version of CLANG with support for OpenMP offloading directives to NVIDIA GPUs to experiement with OpenMP 4.5 target directives. ]]> Wed, 13 Apr 2016 14:44:08 GMT /slideshow/gtc16-s6510-targeting-gpus-with-openmp-45/60869651 jefflarkin@slideshare.net(jefflarkin) GTC16 - S6510 - Targeting GPUs with OpenMP 4.5 jefflarkin These slides are from an instructor-led tutorial from GTC16. The talk discusses using a pre-release version of CLANG with support for OpenMP offloading directives to NVIDIA GPUs to experiement with OpenMP 4.5 target directives. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/s6510targetinggpuswithopenmp45-160413144408-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> These slides are from an instructor-led tutorial from GTC16. The talk discusses using a pre-release version of CLANG with support for OpenMP offloading directives to NVIDIA GPUs to experiement with OpenMP 4.5 target directives.
GTC16 - S6510 - Targeting GPUs with OpenMP 4.5 from Jeff Larkin
]]> 1582 6 https://cdn.slidesharecdn.com/ss_thumbnails/s6510targetinggpuswithopenmp45-160413144408-thumbnail.jpg?width=120&height=120&fit=bounds presentation White http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 GTC16 - S6410 - Comparing OpenACC 2.5 and OpenMP 4.5 /slideshow/gtc16-s6410-comparing-openacc-25-and-openmp-45/60869522 s6410-comparingopenacc25andopenmp45-160413144132
This talk was given at GTC16 by James Beyer and Jeff Larkin, both members of the OpenACC and OpenMP committees. It's intended to be an unbiased discussion of the differences between the two languages and the tradeoffs to each approach.]]>
This talk was given at GTC16 by James Beyer and Jeff Larkin, both members of the OpenACC and OpenMP committees. It's intended to be an unbiased discussion of the differences between the two languages and the tradeoffs to each approach.]]> Wed, 13 Apr 2016 14:41:32 GMT /slideshow/gtc16-s6410-comparing-openacc-25-and-openmp-45/60869522 jefflarkin@slideshare.net(jefflarkin) GTC16 - S6410 - Comparing OpenACC 2.5 and OpenMP 4.5 jefflarkin This talk was given at GTC16 by James Beyer and Jeff Larkin, both members of the OpenACC and OpenMP committees. It's intended to be an unbiased discussion of the differences between the two languages and the tradeoffs to each approach. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/s6410-comparingopenacc25andopenmp45-160413144132-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> This talk was given at GTC16 by James Beyer and Jeff Larkin, both members of the OpenACC and OpenMP committees. It&#39;s intended to be an unbiased discussion of the differences between the two languages and the tradeoffs to each approach.
GTC16 - S6410 - Comparing OpenACC 2.5 and OpenMP 4.5 from Jeff Larkin
]]> 1679 5 https://cdn.slidesharecdn.com/ss_thumbnails/s6410-comparingopenacc25andopenmp45-160413144132-thumbnail.jpg?width=120&height=120&fit=bounds presentation White http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 SC13: OpenMP and NVIDIA /slideshow/sc13-openmp-andnvidia/28604016 sc13openmpandnvidia-131125095324-phpapp02
This talk was given at the OpenMP booth at Supercomputing 2013. ]]>
This talk was given at the OpenMP booth at Supercomputing 2013. ]]> Mon, 25 Nov 2013 09:53:24 GMT /slideshow/sc13-openmp-andnvidia/28604016 jefflarkin@slideshare.net(jefflarkin) SC13: OpenMP and NVIDIA jefflarkin This talk was given at the OpenMP booth at Supercomputing 2013. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/sc13openmpandnvidia-131125095324-phpapp02-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> This talk was given at the OpenMP booth at Supercomputing 2013.
SC13: OpenMP and NVIDIA from Jeff Larkin
]]> 1253 6 https://cdn.slidesharecdn.com/ss_thumbnails/sc13openmpandnvidia-131125095324-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Refactoring Applications for the XK7 and Future Hybrid Architectures /jefflarkin/refactoring-applications-for-the-xk7-and-future-hybrid-architectures refactoringforxk7-130516131118-phpapp02
This presentation was originally given at CUG 2013.]]>
This presentation was originally given at CUG 2013.]]> Thu, 16 May 2013 13:11:18 GMT /jefflarkin/refactoring-applications-for-the-xk7-and-future-hybrid-architectures jefflarkin@slideshare.net(jefflarkin) Refactoring Applications for the XK7 and Future Hybrid Architectures jefflarkin This presentation was originally given at CUG 2013. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/refactoringforxk7-130516131118-phpapp02-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> This presentation was originally given at CUG 2013.
Refactoring Applications for the XK7 and Future Hybrid Architectures from Jeff Larkin
]]> 1560 6 https://cdn.slidesharecdn.com/ss_thumbnails/refactoringforxk7-130516131118-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Optimizing GPU to GPU Communication on Cray XK7 /slideshow/optimizing-gpu-to-gpu-communication-on-cray-xk7/21276343 optimizinggputogpucommunicationoncrayxk7-130516130556-phpapp02
When developing an application for Cray XK7 systems, optimization of compute kernels is only a small part of maximizing scaling and performance. Programmers must consider the effect of the GPU’s distinct address space and the PCIe bus on application scalability. Without such considerations applications rapidly become limited by transfers to and from the GPU and fail to scale to large numbers of nodes. This paper will demonstrate methods for optimizing GPU to GPU communication and present XK7 results for these methods. This presentation was originally given at CUG 2013.]]>
When developing an application for Cray XK7 systems, optimization of compute kernels is only a small part of maximizing scaling and performance. Programmers must consider the effect of the GPU’s distinct address space and the PCIe bus on application scalability. Without such considerations applications rapidly become limited by transfers to and from the GPU and fail to scale to large numbers of nodes. This paper will demonstrate methods for optimizing GPU to GPU communication and present XK7 results for these methods. This presentation was originally given at CUG 2013.]]> Thu, 16 May 2013 13:05:56 GMT /slideshow/optimizing-gpu-to-gpu-communication-on-cray-xk7/21276343 jefflarkin@slideshare.net(jefflarkin) Optimizing GPU to GPU Communication on Cray XK7 jefflarkin When developing an application for Cray XK7 systems, optimization of compute kernels is only a small part of maximizing scaling and performance. Programmers must consider the effect of the GPU’s distinct address space and the PCIe bus on application scalability. Without such considerations applications rapidly become limited by transfers to and from the GPU and fail to scale to large numbers of nodes. This paper will demonstrate methods for optimizing GPU to GPU communication and present XK7 results for these methods. This presentation was originally given at CUG 2013. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/optimizinggputogpucommunicationoncrayxk7-130516130556-phpapp02-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> When developing an application for Cray XK7 systems, optimization of compute kernels is only a small part of maximizing scaling and performance. Programmers must consider the effect of the GPU’s distinct address space and the PCIe bus on application scalability. Without such considerations applications rapidly become limited by transfers to and from the GPU and fail to scale to large numbers of nodes. This paper will demonstrate methods for optimizing GPU to GPU communication and present XK7 results for these methods. This presentation was originally given at CUG 2013.
Optimizing GPU to GPU Communication on Cray XK7 from Jeff Larkin
]]> 1166 4 https://cdn.slidesharecdn.com/ss_thumbnails/optimizinggputogpucommunicationoncrayxk7-130516130556-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation White http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Progress Toward Accelerating CAM-SE /slideshow/progress-toward-accelerating-camse/9228826 progresstowardacceleratingcam-se-110912140718-phpapp01
]]>
]]> Mon, 12 Sep 2011 14:07:12 GMT /slideshow/progress-toward-accelerating-camse/9228826 jefflarkin@slideshare.net(jefflarkin) Progress Toward Accelerating CAM-SE jefflarkin <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/progresstowardacceleratingcam-se-110912140718-phpapp01-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br>
Progress Toward Accelerating CAM-SE from Jeff Larkin
]]> 601 2 https://cdn.slidesharecdn.com/ss_thumbnails/progresstowardacceleratingcam-se-110912140718-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 HPCMPUG2011 cray tutorial /slideshow/hpcmpug2011-cray-tutorial/8358217 hpcmpug2011craytutorial-110620010019-phpapp01
Maximizing Application Performance on the Cray XE6]]>
Maximizing Application Performance on the Cray XE6]]> Mon, 20 Jun 2011 01:00:16 GMT /slideshow/hpcmpug2011-cray-tutorial/8358217 jefflarkin@slideshare.net(jefflarkin) HPCMPUG2011 cray tutorial jefflarkin Maximizing Application Performance �on the Cray XE6 <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/hpcmpug2011craytutorial-110620010019-phpapp01-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> Maximizing Application Performance �on the Cray XE6
HPCMPUG2011 cray tutorial from Jeff Larkin
]]> 1998 7 https://cdn.slidesharecdn.com/ss_thumbnails/hpcmpug2011craytutorial-110620010019-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 CUG2011 Introduction to GPU Computing /slideshow/cug2011-introduction-to-gpu-computing/8068998 larkintutorial-110523100034-phpapp01
]]>
]]> Mon, 23 May 2011 10:00:33 GMT /slideshow/cug2011-introduction-to-gpu-computing/8068998 jefflarkin@slideshare.net(jefflarkin) CUG2011 Introduction to GPU Computing jefflarkin <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/larkintutorial-110523100034-phpapp01-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br>
CUG2011 Introduction to GPU Computing from Jeff Larkin
]]> 1349 1 https://cdn.slidesharecdn.com/ss_thumbnails/larkintutorial-110523100034-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Maximizing Application Performance on Cray XT6 and XE6 Supercomputers DOD-MOD Users Group 2010 /slideshow/maximizing-application-performance-on-cray-xt6-and-xe6-supercomputers-dodmod-users-group-2010/4488835 maximizingapplicationperformance-100613133819-phpapp01
This presentation gives basic information about optimizing applications for Cray XT6 and XE6 Supercomputers.]]>
This presentation gives basic information about optimizing applications for Cray XT6 and XE6 Supercomputers.]]> Sun, 13 Jun 2010 13:38:14 GMT /slideshow/maximizing-application-performance-on-cray-xt6-and-xe6-supercomputers-dodmod-users-group-2010/4488835 jefflarkin@slideshare.net(jefflarkin) Maximizing Application Performance on Cray XT6 and XE6 Supercomputers DOD-MOD Users Group 2010 jefflarkin This presentation gives basic information about optimizing applications for Cray XT6 and XE6 Supercomputers. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/maximizingapplicationperformance-100613133819-phpapp01-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> This presentation gives basic information about optimizing applications for Cray XT6 and XE6 Supercomputers.
Maximizing Application Performance on Cray XT6 and XE6 Supercomputers DOD-MOD Users Group 2010 from Jeff Larkin
]]> 2322 7 https://cdn.slidesharecdn.com/ss_thumbnails/maximizingapplicationperformance-100613133819-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 May2010 hex-core-opt /slideshow/may2010-hexcoreopt/4040005 may2010-hex-core-opt-100510122121-phpapp02
]]>
]]> Mon, 10 May 2010 12:21:15 GMT /slideshow/may2010-hexcoreopt/4040005 jefflarkin@slideshare.net(jefflarkin) May2010 hex-core-opt jefflarkin <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/may2010-hex-core-opt-100510122121-phpapp02-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br>
May2010 hex-core-opt from Jeff Larkin
]]> 464 2 https://cdn.slidesharecdn.com/ss_thumbnails/may2010-hex-core-opt-100510122121-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 A Comparison of Accelerator Programming Models /slideshow/a-comparison-of-accelerator-programming-models/3921001 olcf-seminar-100430111657-phpapp01
This talk compares the pros and cons of GPU programming with CUDA C, CUDA Fortran, PGI accelerator directives, CAPS HMPP directives, and OpenCL. It was presented on April 30, 2010 in the OLCF Seminar Series at Oak Ridge National Lab. The original presentation is also available on prezi (http://prezi.com/5ogkgcw9qske/) and a video of the presentation will be posted elsewhere once available.]]>
This talk compares the pros and cons of GPU programming with CUDA C, CUDA Fortran, PGI accelerator directives, CAPS HMPP directives, and OpenCL. It was presented on April 30, 2010 in the OLCF Seminar Series at Oak Ridge National Lab. The original presentation is also available on prezi (http://prezi.com/5ogkgcw9qske/) and a video of the presentation will be posted elsewhere once available.]]> Fri, 30 Apr 2010 11:16:52 GMT /slideshow/a-comparison-of-accelerator-programming-models/3921001 jefflarkin@slideshare.net(jefflarkin) A Comparison of Accelerator Programming Models jefflarkin This talk compares the pros and cons of GPU programming with CUDA C, CUDA Fortran, PGI accelerator directives, CAPS HMPP directives, and OpenCL. It was presented on April 30, 2010 in the OLCF Seminar Series at Oak Ridge National Lab. The original presentation is also available on prezi (http://prezi.com/5ogkgcw9qske/) and a video of the presentation will be posted elsewhere once available. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/olcf-seminar-100430111657-phpapp01-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> This talk compares the pros and cons of GPU programming with CUDA C, CUDA Fortran, PGI accelerator directives, CAPS HMPP directives, and OpenCL. It was presented on April 30, 2010 in the OLCF Seminar Series at Oak Ridge National Lab. The original presentation is also available on prezi (http://prezi.com/5ogkgcw9qske/) and a video of the presentation will be posted elsewhere once available.
A Comparison of Accelerator Programming Models from Jeff Larkin
]]> 1952 3 https://cdn.slidesharecdn.com/ss_thumbnails/olcf-seminar-100430111657-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Cray XT Porting, Scaling, and Optimization Best Practices /slideshow/cray-xt-porting-scaling-and-optimization-best-practices/2674254 optimization-091208093428-phpapp02
]]>
]]> Tue, 08 Dec 2009 09:34:18 GMT /slideshow/cray-xt-porting-scaling-and-optimization-best-practices/2674254 jefflarkin@slideshare.net(jefflarkin) Cray XT Porting, Scaling, and Optimization Best Practices jefflarkin <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/optimization-091208093428-phpapp02-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br>
Cray XT Porting, Scaling, and Optimization Best Practices from Jeff Larkin
]]> 788 2 https://cdn.slidesharecdn.com/ss_thumbnails/optimization-091208093428-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 XT Best Practices /slideshow/xt-best-practices/2433947 sdscxtbestpractices-091105190710-phpapp02
These slides are a series of "best practices" for running on the Cray XT line of supercomputers. This talk was presented at the HPCMP meeting at SDSC on 11/5/2009]]>
These slides are a series of "best practices" for running on the Cray XT line of supercomputers. This talk was presented at the HPCMP meeting at SDSC on 11/5/2009]]> Thu, 05 Nov 2009 18:52:58 GMT /slideshow/xt-best-practices/2433947 jefflarkin@slideshare.net(jefflarkin) XT Best Practices jefflarkin These slides are a series of "best practices" for running on the Cray XT line of supercomputers. This talk was presented at the HPCMP meeting at SDSC on 11/5/2009 <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/sdscxtbestpractices-091105190710-phpapp02-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> These slides are a series of &quot;best practices&quot; for running on the Cray XT line of supercomputers. This talk was presented at the HPCMP meeting at SDSC on 11/5/2009
XT Best Practices from Jeff Larkin
]]> 1235 3 https://cdn.slidesharecdn.com/ss_thumbnails/sdscxtbestpractices-091105190710-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Practical Examples for Efficient I/O on Cray XT Systems (CUG 2009) /slideshow/larkin2009slides/1469541 larkin2009slides-124291213059-phpapp01
]]>
]]> Thu, 21 May 2009 08:23:40 GMT /slideshow/larkin2009slides/1469541 jefflarkin@slideshare.net(jefflarkin) Practical Examples for Efficient I/O on Cray XT Systems (CUG 2009) jefflarkin <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/larkin2009slides-124291213059-phpapp01-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br>
Practical Examples for Efficient I/O on Cray XT Systems (CUG 2009) from Jeff Larkin
]]> 542 2 https://cdn.slidesharecdn.com/ss_thumbnails/larkin2009slides-124291213059-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 https://cdn.slidesharecdn.com/profile-photo-jefflarkin-48x48.jpg?cb=1623789715 HPC application profiling and optimization specialist with experience in MPP, cluster, and grid computing platforms. jefflarkin.com https://cdn.slidesharecdn.com/ss_thumbnails/openmpongpusbestpractices-201217161130-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/best-practices-for-openmp-on-gpus-openmp-uk-users-group/240235084 Best Practices for Ope... https://cdn.slidesharecdn.com/ss_thumbnails/fortranconopenaccjefflarkin-200714135353-thumbnail.jpg?width=320&height=320&fit=bounds jefflarkin/fortrancon2020-highly-parallel-fortran-and-openacc-directives FortranCon2020: Highly... https://cdn.slidesharecdn.com/ss_thumbnails/2larkindoeportability-170828134704-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/early-results-of-openmp-45-portability-on-nvidia-gpus-cpus/79216550 Early Results of OpenM...