{"id":165657,"date":"2023-03-17T09:00:55","date_gmt":"2023-03-17T09:00:55","guid":{"rendered":"\/knowledge\/forums\/topic\/running-2021-r1-or-r2-mechanical-apdl-gpu-acceleration-on-windows-or-linux-with-nvidia-tesla-kepler-architecture-gpus-such-as-k40c-and-k20xm-either-fails-or-continues-without-gpu-acceleration-with\/"},"modified":"2023-07-31T12:34:34","modified_gmt":"2023-07-31T12:34:34","slug":"running-2021-r1-or-r2-mechanical-apdl-gpu-acceleration-on-windows-or-linux-with-nvidia-tesla-kepler-architecture-gpus-such-as-k40c-and-k20xm-either-fails-or-continues-without-gpu-acceleration-with","status":"publish","type":"topic","link":"https:\/\/innovationspace.ansys.com\/knowledge\/forums\/topic\/running-2021-r1-or-r2-mechanical-apdl-gpu-acceleration-on-windows-or-linux-with-nvidia-tesla-kepler-architecture-gpus-such-as-k40c-and-k20xm-either-fails-or-continues-without-gpu-acceleration-with\/","title":{"rendered":"Running 2021 R1 or R2 Mechanical APDL GPU acceleration on Windows or Linux with NVIDIA Tesla Kepler architecture GPUs, such as K40c and K20xm, either fails or continues without GPU acceleration, with below error or warning:    *** ERROR ***                           CP =       3.107   TIME= 16:32:26  A generic error has occurred when using the GPU accelerator capability.   Error code = 209 which translates to: no kernel image is available       for execution on the device.  Please rerun without using the GPU          accelerator capability.  Please send the data leading to this             operation to your technical support provider, as this will allow          ANSYS, Inc to improve the program.  Error during call to matrix factor routine on GPU!!  Error code = 209.  *** WARNING ***                         CP =      14.225   TIME= 15:02:00  An unexpected error has occurred while using the GPU accelerator          capability.  This feature is being disabled.  Please send the data        leading to this operation to your technical support provider, as this     will allow ANSYS, Inc to improve the program."},"content":{"rendered":"<p>Support for the Kepler cards was dropped in Mechanical APDL 2021 R1 GPU acceleration.   If upgrading to latest recommended GPU (such as A100, GV100) is not an immediate option, try below workaround: \u00e2\u20ac\u00a2 Linux: Rename v211\/ansys\/lib\/linx64\/nvidia folder, copy (cp -R) v202\/ansys\/lib\/linx64\/nvidia to v211\/ansys\/lib\/linx64.  \u00e2\u20ac\u00a2 Windows: Rename v211ansysbinwinx64GPUNVIDIA folder, copy v202ansysbinwinx64GPUNVIDIA folder to v211ansysbinwinx64GPU.   Above workaround is to downgrade from CUDA library files from version 11.0 to 10.0 for 2021 R1 or R2, to enable GPU acceleration with Kepler series GPU.<\/p>\n","protected":false},"template":"","class_list":["post-165657","topic","type-topic","status-publish","hentry","topic-tag-2021-r1","topic-tag-apdl","topic-tag-gpu","topic-tag-hpc-parallel","topic-tag-installation-licensing-systems","topic-tag-kepler","topic-tag-mechanical-apdl-sys"],"aioseo_notices":[],"acf":[],"custom_fields":[{"0":{"_wp_page_template":["default"],"_bbp_last_active_time":["3\/15\/2023 20:20"],"_bbp_forum_id":["153867"],"_bbp_author_ip":["23.56.168.180"],"_btv_view_count":["1308"],"siebel_km_number":["2066860"],"product_version":["2021 R1"],"km_published_date":["2021-11-30T00:00:00.000Z"],"family":["Installation\/Licensing\/Systems"],"application_name":["Mechanical APDL - SYS"]},"test":"articlesansys-com"}],"_links":{"self":[{"href":"https:\/\/innovationspace.ansys.com\/knowledge\/wp-json\/wp\/v2\/topics\/165657","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/innovationspace.ansys.com\/knowledge\/wp-json\/wp\/v2\/topics"}],"about":[{"href":"https:\/\/innovationspace.ansys.com\/knowledge\/wp-json\/wp\/v2\/types\/topic"}],"version-history":[{"count":0,"href":"https:\/\/innovationspace.ansys.com\/knowledge\/wp-json\/wp\/v2\/topics\/165657\/revisions"}],"wp:attachment":[{"href":"https:\/\/innovationspace.ansys.com\/knowledge\/wp-json\/wp\/v2\/media?parent=165657"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}