{"id":38117,"date":"2023-06-13T14:46:36","date_gmt":"2023-06-13T13:46:36","guid":{"rendered":"https:\/\/mcomputers.cz\/2023\/06\/13\/hpc-accelerator-comparison-2023\/"},"modified":"2025-01-03T15:08:42","modified_gmt":"2025-01-03T14:08:42","slug":"hpc-accelerator-comparison-2023","status":"publish","type":"post","link":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/","title":{"rendered":"HPC Accelerator Comparison 2023"},"content":{"rendered":"<div id='av_section_1'  class='avia-section av-6oeo6ab-0e5c1210bf9f541e4fa62d4279baf11c main_color avia-section-default avia-no-border-styling  avia-builder-el-0  el_before_av_section  avia-builder-el-first  avia-bg-style-scroll container_wrap fullsize'  ><div class='container av-section-cont-open' ><main  role=\"main\" itemprop=\"mainContentOfPage\"  class='template-page content  av-content-full alpha units'><div class='post-entry post-entry-type-page post-entry-38117'><div class='entry-content-wrapper clearfix'>\n<div  class='flex_column av-632vkk3-431739796aaaebb5d2691474b74247ac av_one_half  avia-builder-el-1  el_before_av_one_half  avia-builder-el-first  first flex_column_div  '     ><section  class='av_textblock_section av-m5gtj1ub-400921b0063936e827a168052f491bc5 '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock h-decor'  itemprop=\"text\" ><h2>HPC Accelerator Comparison for 2023<\/h2>\n<p><strong>13.6.2023<\/strong><\/p>\n<p>Every year, companies are coming up with new and unique accelerators and solutions for HPC that bring major leaps in the development, research and production of AI, ML and DL applications. That&#8217;s why we&#8217;ve summarized the most prominent ones and compared their individual parameters, performance and summarized their features.<\/p>\n<\/div><\/section><br \/>\n\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-5ty40b7-4530fbfedbbeea4cb30da17f9abd348a\">\n#top .hr.hr-invisible.av-5ty40b7-4530fbfedbbeea4cb30da17f9abd348a{\nheight:50px;\n}\n<\/style>\n<div  class='hr av-5ty40b7-4530fbfedbbeea4cb30da17f9abd348a hr-invisible  avia-builder-el-3  el_after_av_textblock  el_before_av_social_share '><span class='hr-inner '><span class=\"hr-inner-style\"><\/span><\/span><\/div><br \/>\n<div  class='av-social-sharing-box av-54472k3-e7390fa12778012da7410a4dd7109bd8 av-social-sharing-box-circle  avia-builder-el-4  el_after_av_hr  avia-builder-el-last  social-icons av-social-sharing-box-color-bg av-social-sharing-box-same-width'><div class=\"av-share-box\"><h5 class='av-share-link-description av-no-toc '>Share this post<\/h5><ul class=\"av-share-box-list noLightbox\"><li class='av-share-link av-social-link-facebook' ><a target=\"_blank\" aria-label=\"Share on Facebook\" href='https:\/\/www.facebook.com\/sharer.php?u=https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/&#038;t=HPC%20Accelerator%20Comparison%202023' aria-hidden='false' data-av_icon='\ue8f3' data-av_iconfont='entypo-fontello' title='' data-avia-related-tooltip='Share on Facebook' rel=\"noopener\"><span class='avia_hidden_link_text'>Share on Facebook<\/span><\/a><\/li><li class='av-share-link av-social-link-linkedin' ><a target=\"_blank\" aria-label=\"Share on LinkedIn\" href='https:\/\/linkedin.com\/shareArticle?mini=true&#038;title=HPC%20Accelerator%20Comparison%202023&#038;url=https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/' aria-hidden='false' data-av_icon='\ue8fc' data-av_iconfont='entypo-fontello' title='' data-avia-related-tooltip='Share on LinkedIn' rel=\"noopener\"><span class='avia_hidden_link_text'>Share on LinkedIn<\/span><\/a><\/li><li class='av-share-link av-social-link-mail' ><a  aria-label=\"Share by Mail\" href='mailto:?subject=HPC%20Accelerator%20Comparison%202023&#038;body=https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/' aria-hidden='false' data-av_icon='\ue805' data-av_iconfont='entypo-fontello' title='' data-avia-related-tooltip='Share by Mail'><span class='avia_hidden_link_text'>Share by Mail<\/span><\/a><\/li><li class='av-share-link av-social-link-instagram' ><a target=\"_blank\" aria-label=\"Link to Instagram\" href='https:\/\/www.instagram.com\/mcomputers_lenovoshop\/' aria-hidden='false' data-av_icon='\uf16d' data-av_iconfont='entypo-fontello' title='' data-avia-related-tooltip='Link to Instagram' rel=\"noopener\"><span class='avia_hidden_link_text'>Link to Instagram<\/span><\/a><\/li><\/ul><\/div><\/div><\/p><\/div><div  class='flex_column av-4vbtboz-daaec48c469f5584ba295a08755e7e7e av_one_half  avia-builder-el-5  el_after_av_one_half  el_before_av_hr  flex_column_div  '     ><style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-m5gtld6c-01cb7376e8b2da57169e10ab3f2bb4a0\">\n.avia-image-container.av-m5gtld6c-01cb7376e8b2da57169e10ab3f2bb4a0 img.avia_image{\nbox-shadow:none;\n}\n.avia-image-container.av-m5gtld6c-01cb7376e8b2da57169e10ab3f2bb4a0 .av-image-caption-overlay-center{\ncolor:#ffffff;\n}\n<\/style>\n<div  class='avia-image-container av-m5gtld6c-01cb7376e8b2da57169e10ab3f2bb4a0 av-styling- avia-align-center  avia-builder-el-6  avia-builder-el-no-sibling '   itemprop=\"image\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/ImageObject\" ><div class=\"avia-image-container-inner\"><div class=\"avia-image-overlay-wrap\"><img decoding=\"async\" fetchpriority=\"high\" class='wp-image-26364 avia-img-lazy-loading-not-26364 avia_image ' src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/221694014_AMD-EPYC-Genoa-Beauty-Shot_05-1.png\" alt='AMD EPYC' title='221694014_AMD EPYC (Genoa) Beauty Shot_05 (1)'  height=\"423\" width=\"1200\"  itemprop=\"thumbnailUrl\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/221694014_AMD-EPYC-Genoa-Beauty-Shot_05-1.png 1200w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/221694014_AMD-EPYC-Genoa-Beauty-Shot_05-1-300x106.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/221694014_AMD-EPYC-Genoa-Beauty-Shot_05-1-1030x363.png 1030w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/221694014_AMD-EPYC-Genoa-Beauty-Shot_05-1-768x271.png 768w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/221694014_AMD-EPYC-Genoa-Beauty-Shot_05-1-705x249.png 705w\" sizes=\"(max-width: 1200px) 100vw, 1200px\" \/><\/div><\/div><\/div><\/div>\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-4al7r3n-d6a87e4bfc5a9299ddc00245b0e009a5\">\n#top .hr.av-4al7r3n-d6a87e4bfc5a9299ddc00245b0e009a5{\nmargin-bottom:30px;\n}\n.hr.av-4al7r3n-d6a87e4bfc5a9299ddc00245b0e009a5 .hr-inner{\nwidth:100vw;\nborder-color:#e5f4ff;\n}\n<\/style>\n<div  class='hr av-4al7r3n-d6a87e4bfc5a9299ddc00245b0e009a5 hr-custom  avia-builder-el-7  el_after_av_one_half  avia-builder-el-last  hr-center hr-icon-no'><span class='hr-inner inner-border-av-border-thin'><span class=\"hr-inner-style\"><\/span><\/span><\/div>\n\n<\/div><\/div><\/main><!-- close content main element --><\/div><\/div><div id='av_section_2'  class='avia-section av-dyziyb-4e31b68ce23df69e8b6c99f85254f3f3 main_color avia-section-default avia-no-border-styling  avia-builder-el-8  el_after_av_section  el_before_av_one_full  avia-bg-style-scroll container_wrap fullsize'  ><div class='container av-section-cont-open' ><div class='template-page content  av-content-full alpha units'><div class='post-entry post-entry-type-page post-entry-38117'><div class='entry-content-wrapper clearfix'>\n<div  class='flex_column av-3a0ymwj-f56e448e43d58a5f548cfc2193f42f53 av_one_half  avia-builder-el-9  el_before_av_one_half  avia-builder-el-first  first flex_column_div  '     ><section  class='av_textblock_section av-m5gtjs0y-2e823c06c80d74f25c4391176e2f16e7 '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock h-decor'  itemprop=\"text\" ><h3>Overview of accelerators<\/h3>\n<\/div><\/section><br \/>\n<section  class='av_textblock_section av-lbkrkld2-caba850676e52989ab0ed098b1c96c10 '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock'  itemprop=\"text\" ><p>Graphics or other computing accelerators play a huge role in HPC today. Instead of all the work being processed by traditional processors, calculations are distributed to computing units that have precisely tailored instruction sets, special cores or fast memory to process problems many times faster. Unlike processors, which currently top models with around 100 cores, graphics and other accelerators have on the order of hundreds to thousands of cores. While their cores are weaker compared to processor ones, they are precisely made for the specific operations they are supposed to perform, and they excel at it. Companies are developing better and better accelerators, better algorithms and instructions, better architectures&#8230; To help us get a better handle on the current trends and to get a good overview of the options we have right now, we&#8217;ve written this comparison of the current innovations on the market.<\/p>\n<\/div><\/section><br \/>\n\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-lbkrjfry-985b7547c91ee03193198fbc499fa583\">\n.avia-image-container.av-lbkrjfry-985b7547c91ee03193198fbc499fa583 img.avia_image{\nbox-shadow:none;\n}\n.avia-image-container.av-lbkrjfry-985b7547c91ee03193198fbc499fa583 .av-image-caption-overlay-center{\ncolor:#ffffff;\n}\n<\/style>\n<div  class='avia-image-container av-lbkrjfry-985b7547c91ee03193198fbc499fa583 av-styling- avia-align-center  avia-builder-el-12  el_after_av_textblock  avia-builder-el-last '   itemprop=\"image\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/ImageObject\" ><div class=\"avia-image-container-inner\"><div class=\"avia-image-overlay-wrap\"><img decoding=\"async\" fetchpriority=\"high\" class='wp-image-24768 avia-img-lazy-loading-not-24768 avia_image ' src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-12_131925808.png\" alt='HPC Akceleratory ilustracni foto' title='image_2022-12-12_131925808'  height=\"600\" width=\"800\"  itemprop=\"thumbnailUrl\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-12_131925808.png 800w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-12_131925808-300x225.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-12_131925808-768x576.png 768w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-12_131925808-705x529.png 705w\" sizes=\"(max-width: 800px) 100vw, 800px\" \/><\/div><\/div><\/div><\/p><\/div><div  class='flex_column av-2uvnpxv-d733aebb851a9f360d62048f42a996ef av_one_half  avia-builder-el-13  el_after_av_one_half  avia-builder-el-last  flex_column_div  '     ><section  class='av_textblock_section av-lbqfvluv-25bc3f43fdab513a0d2f2591c35852aa '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock h-decor'  itemprop=\"text\" ><h3>Description of individual accelerators<\/h3>\n<\/div><\/section><br \/>\n<div  class='avia-buttonrow-wrap av-lbqfh6on-36989f87383559ce8a17e514bdce4e93 avia-buttonrow-left  avia-builder-el-15  el_after_av_textblock  avia-builder-el-last '>\n\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-lbqfdypc-cae3e508edc1e62e618c4faddb217e92\">\n#top #wrap_all .avia-button.av-lbqfdypc-cae3e508edc1e62e618c4faddb217e92{\nmargin-bottom:5px;\nmargin-right:5px;\n}\n<\/style>\n<a href='#intel--flex'  class='avia-button av-lbqfdypc-cae3e508edc1e62e618c4faddb217e92 avia-icon_select-no avia-size-small avia-color-theme-color'   aria-label=\"Intel Flex 170\"><span class='avia_iconbox_title' >Intel Flex 170<\/span><\/a>\n\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-lbqfdypc-8-2a0a8f2c36f06d03564eb6ef253e3d25\">\n#top #wrap_all .avia-button.av-lbqfdypc-8-2a0a8f2c36f06d03564eb6ef253e3d25{\nmargin-bottom:5px;\nmargin-right:5px;\n}\n<\/style>\n<a href=''  class='avia-button av-lbqfdypc-8-2a0a8f2c36f06d03564eb6ef253e3d25 avia-icon_select-no avia-size-small avia-color-theme-color'   aria-label=\"Intel Gaudi2\"><span class='avia_iconbox_title' >Intel Gaudi2<\/span><\/a>\n\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-lbqfdypc-7-b3a8f27a66ca9efa67fcf8e48499b8ff\">\n#top #wrap_all .avia-button.av-lbqfdypc-7-b3a8f27a66ca9efa67fcf8e48499b8ff{\nmargin-bottom:5px;\nmargin-right:5px;\n}\n<\/style>\n<a href='#amd--mi250x'  class='avia-button av-lbqfdypc-7-b3a8f27a66ca9efa67fcf8e48499b8ff avia-icon_select-no avia-size-small avia-color-theme-color'   aria-label=\"AMD MI 300X \/ 250X\"><span class='avia_iconbox_title' >AMD MI 300X \/ 250X<\/span><\/a>\n\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-lbqfdypc-6-abe98820ba4ba95512a1e7d08092cf7a\">\n#top #wrap_all .avia-button.av-lbqfdypc-6-abe98820ba4ba95512a1e7d08092cf7a{\nmargin-bottom:5px;\nmargin-right:5px;\n}\n<\/style>\n<a href='#nvidia--a100--h100'  class='avia-button av-lbqfdypc-6-abe98820ba4ba95512a1e7d08092cf7a avia-icon_select-no avia-size-small avia-color-theme-color'   aria-label=\"NVIDIA A100 \/ H100\"><span class='avia_iconbox_title' >NVIDIA A100 \/ H100<\/span><\/a>\n\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-lbqfdypc-5-3a19518cbda1403fa9111e32e6f0faf5\">\n#top #wrap_all .avia-button.av-lbqfdypc-5-3a19518cbda1403fa9111e32e6f0faf5{\nmargin-bottom:5px;\nmargin-right:5px;\n}\n<\/style>\n<a href='#graphcore--c600'  class='avia-button av-lbqfdypc-5-3a19518cbda1403fa9111e32e6f0faf5 avia-icon_select-no avia-size-small avia-color-theme-color'   aria-label=\"Graphcore C600\"><span class='avia_iconbox_title' >Graphcore C600<\/span><\/a>\n\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-lbqfdypc-4-c348a27169e2c8f886e8a7faca835148\">\n#top #wrap_all .avia-button.av-lbqfdypc-4-c348a27169e2c8f886e8a7faca835148{\nmargin-bottom:5px;\nmargin-right:5px;\n}\n<\/style>\n<a href='#cerebras--wse2'  class='avia-button av-lbqfdypc-4-c348a27169e2c8f886e8a7faca835148 avia-icon_select-no avia-size-small avia-color-theme-color'   aria-label=\"Cerebras WSE-2\"><span class='avia_iconbox_title' >Cerebras WSE-2<\/span><\/a>\n\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-lbqfdypc-3-6cea992c83124737944307e1fd3a20a8\">\n#top #wrap_all .avia-button.av-lbqfdypc-3-6cea992c83124737944307e1fd3a20a8{\nmargin-bottom:5px;\nmargin-right:5px;\n}\n<\/style>\n<a href='#groq--groqcard'  class='avia-button av-lbqfdypc-3-6cea992c83124737944307e1fd3a20a8 avia-icon_select-no avia-size-small avia-color-theme-color'   aria-label=\"Groq GroqCard\"><span class='avia_iconbox_title' >Groq GroqCard<\/span><\/a>\n<\/div><\/p><\/div>\n<\/div><\/div><\/div><!-- close content main div --><\/div><\/div><div id='after_section_2'  class='main_color av_default_container_wrap container_wrap fullsize'  ><div class='container av-section-cont-open' ><div class='template-page content  av-content-full alpha units'><div class='post-entry post-entry-type-page post-entry-38117'><div class='entry-content-wrapper clearfix'>\n<div  class='flex_column av-gi8gpar-d8fd612824715b520f1806d971279c0b av_one_full  avia-builder-el-16  el_after_av_section  el_before_av_one_full  avia-builder-el-first  first flex_column_div  '     ><section  class='av_textblock_section av-lbkrt1cy-0724f4844e589144dce0123f73b6dd4f '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock h-decor'  itemprop=\"text\" ><h2>Accelerator parameters<\/h2>\n<\/div><\/section><br \/>\n<div class='avia-data-table-wrap av-1ljk5bn-fa21ccd5a72f664d0650b9356208e8d2 avia_responsive_table avia-table-1'><table  class='avia-table avia-data-table avia_pricing_default  avia-builder-el-18  el_after_av_textblock  el_before_av_textblock '  itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/Table\" ><tbody><tr class=''><td class=''>Company<\/td><td class='avia-center-col'>Intel<\/td><td class='avia-center-col'>Intel<\/td><td class='avia-center-col'>AMD<\/td><td class='avia-center-col'>AMD<\/td><td class='avia-center-col'>NVIDIA<\/td><td class='avia-center-col'>NVIDIA<\/td><td class='avia-center-col'>Graphcore<\/td><td class='avia-center-col'>Cerebras<\/td><\/tr><tr class=''><td class=''>Accelerator<\/td><td class='avia-center-col'>Flex 170<\/td><td class='avia-center-col'>Gaudi2<\/td><td class='avia-center-col'>MI 250X<\/td><td class='avia-center-col'>MI 300X<\/td><td class='avia-center-col'>A100<\/td><td class='avia-center-col'>H100<\/td><td class='avia-center-col'>C600<\/td><td class='avia-center-col'>WSE-2<\/td><\/tr><tr class=''><td class=''>Number of cores<\/td><td class='avia-center-col'>32 Xe<\/td><td class='avia-center-col'>24 TPC<\/td><td class='avia-center-col'>14,080<\/td><td class='avia-center-col'>12 Chiplets<br \/>\n<\/td><td class='avia-center-col'>6,912<\/td><td class='avia-center-col'>7,296<\/td><td class='avia-center-col'>1,472 IPU<\/td><td class='avia-center-col'>850,000<br \/>\n<\/td><\/tr><tr class=''><td class=''>Number of transistors<\/td><td class='avia-center-col'>21,700 Mil.<\/td><td class='avia-center-col'>TB\/A<\/td><td class='avia-center-col'>58,000 Mil.<\/td><td class='avia-center-col'>153,000 Mil.<br \/>\n<\/td><td class='avia-center-col'>54,200 Mil.<\/td><td class='avia-center-col'>80,000 Mil.<\/td><td class='avia-center-col'>59,400 Mil.<\/td><td class='avia-center-col'>2.600 Billion.<\/td><\/tr><tr class=''><td class=''>Production process<\/td><td class='avia-center-col'>Intel 7 (10nm)<\/td><td class='avia-center-col'>7nm<\/td><td class='avia-center-col'>6nm<\/td><td class='avia-center-col'>5nm*<\/td><td class='avia-center-col'>7nm Ampere<br \/>\n<\/td><td class='avia-center-col'>4nm Hopper<\/td><td class='avia-center-col'>7nm MK2 <\/td><td class='avia-center-col'>7nm TSMC N7<br \/>\n<\/td><\/tr><tr class=''><td class=''>Capacity<\/td><td class='avia-center-col'>16 GB<\/td><td class='avia-center-col'>96 GB<\/td><td class='avia-center-col'>128 GB<\/td><td class='avia-center-col'>192 GB<\/td><td class='avia-center-col'>80 GB<\/td><td class='avia-center-col'>80 GB<\/td><td class='avia-center-col'>900 MB<\/td><td class='avia-center-col'>40 GB<br \/>\n<\/td><\/tr><tr class=''><td class=''>Memory type<\/td><td class='avia-center-col'>GDDR6<\/td><td class='avia-center-col'>HBM2e<\/td><td class='avia-center-col'>HBM2e<\/td><td class='avia-center-col'>HBM3<\/td><td class='avia-center-col'>HBM2e<\/td><td class='avia-center-col'>HBM2e<\/td><td class='avia-center-col'>IPM<\/td><td class='avia-center-col'>on-chip SRAM<\/td><\/tr><tr class=''><td class=''>Bus<br \/>\n<\/td><td class='avia-center-col'>PCIe gen. 4<\/td><td class='avia-center-col'>SXM5<\/td><td class='avia-center-col'>OAM<\/td><td class='avia-center-col'>SXM5*<\/td><td class='avia-center-col'>SXM4<\/td><td class='avia-center-col'>SXM5<\/td><td class='avia-center-col'>PCIe gen. 4<\/td><td class='avia-center-col'>prop.<\/td><\/tr><tr class=''><td class=''>Consumption<\/td><td class='avia-center-col'>150 W<\/td><td class='avia-center-col'>600 W<\/td><td class='avia-center-col'>560 W<\/td><td class='avia-center-col'>TB\/A<\/td><td class='avia-center-col'>500 W<\/td><td class='avia-center-col'>700 W<\/td><td class='avia-center-col'>185 W<\/td><td class='avia-center-col'>15 kW<\/td><\/tr><tr class=''><td class=''>Price<\/td><td class='avia-center-col'>6,400 USD (est.)<\/td><td class='avia-center-col'>TB\/A<\/td><td class='avia-center-col'>8,000 USD<\/td><td class='avia-center-col'>TB\/A<\/td><td class='avia-center-col'>32,097 USD<\/td><td class='avia-center-col'>36,405 USD<\/td><td class='avia-center-col'>N\/A<br \/>\n<\/td><td class='avia-center-col'>N\/A<\/td><\/tr><\/tbody><\/table><\/div><br \/>\n<section  class='av_textblock_section av-livrmiem-0cab25b5c27251eb133eb08234d4ef3a '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock'  itemprop=\"text\" ><p style=\"font-size: 11;\">*Parameters are estimates and not verified, subject to change over time<\/p>\n<\/div><\/section><\/p><\/div>\n<div  class='flex_column av-10qp12c3-18c51841157f7c77d7ab778156667ef9 av_one_full  avia-builder-el-20  el_after_av_one_full  el_before_av_one_full  first flex_column_div  column-top-margin'     ><section  class='av_textblock_section av-lbks6vkb-4790e205ad3489c04f90acb1269e0881 '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock h-decor'  itemprop=\"text\" ><h2>Comparison of accelerator performance<\/h2>\n<\/div><\/section><br \/>\n<div class='avia-data-table-wrap av-3l9ztqr-01c0975c9921573b661f042dca2943a0 avia_responsive_table avia-table-2'><table  class='avia-table avia-data-table avia_pricing_default  avia-builder-el-22  el_after_av_textblock  avia-builder-el-last '  itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/Table\" ><tbody><tr class=''><td class=''>Company<\/td><td class='avia-center-col'>Intel<\/td><td class='avia-center-col'>Intel<\/td><td class='avia-center-col'>AMD<\/td><td class='avia-center-col'>AMD<\/td><td class='avia-center-col'>NVIDIA<\/td><td class='avia-center-col'>NVIDIA<\/td><td class='avia-center-col'>Graphcore<\/td><td class='avia-center-col'>Cerebras<\/td><\/tr><tr class=''><td class=''>Accelerator<\/td><td class='avia-center-col'>Xe DC GPU<\/td><td class='avia-center-col'>Gaudi<\/td><td class='avia-center-col'>MI 250X<\/td><td class='avia-center-col'>MI 300X<\/td><td class='avia-center-col'>A100<\/td><td class='avia-center-col'>H100<\/td><td class='avia-center-col'>C600<\/td><td class='avia-center-col'>WSE-2<\/td><\/tr><tr class=''><td class=''>FP16<\/td><td class='avia-center-col'>256 GFLOPs<\/td><td class='avia-center-col'>839 TFLOPs<\/td><td class='avia-center-col'>383 TFLOPs<\/td><td class='avia-center-col'>N\/A<\/td><td class='avia-center-col'>624 TFLOPs<\/td><td class='avia-center-col'>1 979 TFLOPs<\/td><td class='avia-center-col'>280 TFLOPs<\/td><td class='avia-center-col'>7 500 TFLOPs<\/td><\/tr><tr class=''><td class=''>TF32 <\/td><td class='avia-center-col'>N\/A<\/td><td class='avia-center-col'>N\/A<\/td><td class='avia-center-col'>N\/A<\/td><td class='avia-center-col'>N\/A<\/td><td class='avia-center-col'>312 TFLOPs<\/td><td class='avia-center-col'>989 TFLOPs<\/td><td class='avia-center-col'>N\/A<\/td><td class='avia-center-col'>N\/A<\/td><\/tr><tr class=''><td class=''>BFLOAT16<\/td><td class='avia-center-col'>256 GFLOPs<br \/>\n<\/td><td class='avia-center-col'>~ 800 TFLOPs *<\/td><td class='avia-center-col'>383 TFLOPs<\/td><td class='avia-center-col'>N\/A<\/td><td class='avia-center-col'>624 TFLOPs<\/td><td class='avia-center-col'>1 513 TFLOPs<\/td><td class='avia-center-col'>123.13 TFLOPs<\/td><td class='avia-center-col'>~ 7000 TFLOPs *<br \/>\n<\/td><\/tr><tr class=''><td class=''>INT8<\/td><td class='avia-center-col'>512 GOPs<\/td><td class='avia-center-col'>1678 TOPs<\/td><td class='avia-center-col'>383 TOPs<br \/>\n<\/td><td class='avia-center-col'>N\/A<\/td><td class='avia-center-col'>1 248 TOPs<\/td><td class='avia-center-col'>3 026 TOPs<\/td><td class='avia-center-col'>123 TOPs<\/td><td class='avia-center-col'>N\/A<\/td><\/tr><\/tbody><\/table><\/div><\/p><\/div>\n<div  class='flex_column av-vbpo0j-cdb1b0c0abf762d2ddecf2f65fb75584 av_one_full  avia-builder-el-23  el_after_av_one_full  el_before_av_section  avia-builder-el-last  first flex_column_div  column-top-margin'     ><section  id=\"intel--flex\"  class='av_textblock_section av-lbm3sp9k-6619d24ae0b0772839ee44b08978e3ea '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock h-decor'  itemprop=\"text\" ><h2>Intel Flex 170 DC GPU<\/h2>\n<p><img loading=\"lazy\" decoding=\"async\" class=\" wp-image-24791 alignright\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-13_131621599-1030x579.png\" alt=\"Intel Flex 140 and 170\" width=\"459\" height=\"258\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-13_131621599-1030x579.png 1030w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-13_131621599-300x169.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-13_131621599-768x432.png 768w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-13_131621599-1536x864.png 1536w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-13_131621599-1500x844.png 1500w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-13_131621599-705x397.png 705w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-13_131621599.png 1920w\" sizes=\"auto, (max-width: 459px) 100vw, 459px\" \/><\/p>\n<p>Intel Flex graphics accelerators are a new line of datacenter cards from Intel. While they&#8217;re aimed mostly at cloud gaming and computing, Intel also supplies libraries for AI applications, where it doesn&#8217;t perform too badly for its price. Of course, that&#8217;s not their primary strength, as Intel Flex cards are off primarily in video operations like encoding, scaling, compression, and more.  <\/p>\n<h3>Code AV1<\/h3>\n<p>Intel is the first company to bring a graphics accelerator to market with dedicated <a href=\"https:\/\/www.intel.com\/content\/dam\/www\/central-libraries\/us\/en\/documents\/2022-04\/handbrake-av1-solution-brief-final-version.pdf\">AV1<\/a> encoding. This is quite a breakthrough, as the AV1 video codec has many times better compression ratios than AMD and NVIDIA&#8217;s H.264. Unlike H.265 and HEVC, it is also freely available without licensing.  <\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone wp-image-24796\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-14_095818150.png\" alt=\"Intel x NVIDIA video compression\" width=\"601\" height=\"451\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-14_095818150.png 800w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-14_095818150-300x225.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-14_095818150-768x576.png 768w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-14_095818150-705x529.png 705w\" sizes=\"auto, (max-width: 601px) 100vw, 601px\" \/><\/p>\n<h3>Card for cloud gaming<\/h3>\n<p>Intel has decided to take a slightly unconventional step and is also pushing its Intel Flex DC GPU cards in the &#8220;cloud gaming&#8221; direction in marketing. Since this business sector (streaming games to the player&#8217;s device, but all computation and rendering is done remotely on the server) has grown quite a bit in the last couple of years, it&#8217;s possible that Intel will succeed here with its cards. The main advantage for this use case is mostly the already mentioned video encoding, as this is where the biggest loss in performance and quality is, even with faster transfers. Furthermore, Intel Flex cards can offer <\/p>\n<p> <img loading=\"lazy\" decoding=\"async\" class=\" wp-image-24798 alignright\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-14_101136907.png\" alt=\"Intel Flex game streaming\" width=\"271\" height=\"108\"><\/p>\n<p>up to 32 Ray Tracing cores, relatively high clock speeds (2000+ MHz) per core, and plenty of GDDR6 graphics memory; 16GB for the Flex 170 and 12GB for the weaker Flex 140. Intel boasts in its specs that with this performance, a single Flex 170 card can stream up to 68 720p30 games at once. <\/p>\n<h3>AI computing<\/h3>\n<p>Although Intel Flex DC GPU cards are mainly designed for cloud computing and media processing, Intel has created an environment for AI applications with several libraries and instruction sets. The Flex 140 and 170 cards support most of the most widely used AI libraries such as <strong>TensorFlow<\/strong>, <strong>PyTorch<\/strong>, and <strong>OpenVINO<\/strong>. Intel also mentions that AI applications should be able to scale seamlessly from Xeon processors directly to Flex cards, which could result in quite a performance lift for applications, and full Intel solutions designed for less compute-intensive computing could be a very interesting option.<\/p>\n<\/div><\/section><\/div>\n<\/div><\/div><\/div><!-- close content main div --><\/div><\/div><div id='av_section_3'  class='avia-section av-115tjeb-aa3d60dd4c8b36e03af40b1c468ff33c main_color avia-section-default avia-no-border-styling  avia-builder-el-25  el_after_av_one_full  el_before_av_one_full  avia-bg-style-scroll container_wrap fullsize'  ><div class='container av-section-cont-open' ><div class='template-page content  av-content-full alpha units'><div class='post-entry post-entry-type-page post-entry-38117'><div class='entry-content-wrapper clearfix'>\n<div  class='flex_column av-14krjtf-bac881f6dec75f92d6a2fa068497bf59 av_one_full  avia-builder-el-26  el_before_av_one_half  avia-builder-el-first  first flex_column_div  '     ><section  id=\"intel--gaudi\"  class='av_textblock_section av-vt9ycz-f78f25300a8eb45107c1c1038a231bb8 '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock h-decor'  itemprop=\"text\" ><h2>Intel Habana Labs Gaudi2<img loading=\"lazy\" decoding=\"async\" class=\" wp-image-24845 alignright\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_143047179.png\" alt=\"Intel Gaudi2\" width=\"376\" height=\"262\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_143047179.png 900w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_143047179-300x209.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_143047179-768x534.png 768w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_143047179-705x490.png 705w\" sizes=\"auto, (max-width: 376px) 100vw, 376px\" \/><\/h2>\n<p>Intel&#8217;s new division, Intel <a href=\"https:\/\/habana.ai\/\">Habana Labs<\/a>, which it bought in 2019, focuses purely on SXM cards for AI acceleration. For the latest Gaudi2, which is built on 7nm technology compared to the previous 16nm Gaudi, Intel says the card should have up to twice the throughput and performance of the NVIDIA A100. <\/p>\n<h3>The most effective accelerator for AI deep learning<\/h3>\n<p>The use of artificial intelligence has expanded into almost every industry in recent years. Habana Labs wants to make this more accessible and accelerate the process, and has therefore developed a second generation of Accelerators for a variety of AI industries, but basically for Image recognition or NLP, for example. In the mentioned areas we have a comparison of the theoretical TensorFlow performance against NVIDIA accelerators:<\/p>\n<\/div><\/section><\/div><div  class='flex_column av-15vk7yr-7bc99c2b68b94a7c343d38c003fd9349 av_one_half  avia-builder-el-28  el_after_av_one_full  el_before_av_one_half  first flex_column_div  column-top-margin'     ><style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-lbnn85bw-489014e73256498050dfadd2f7a53a10\">\n.avia-image-container.av-lbnn85bw-489014e73256498050dfadd2f7a53a10 img.avia_image{\nbox-shadow:none;\n}\n.avia-image-container.av-lbnn85bw-489014e73256498050dfadd2f7a53a10 .av-image-caption-overlay-center{\ncolor:#ffffff;\n}\n<\/style>\n<div  class='avia-image-container av-lbnn85bw-489014e73256498050dfadd2f7a53a10 av-styling- avia-align-center  avia-builder-el-29  avia-builder-el-no-sibling '   itemprop=\"image\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/ImageObject\" ><div class=\"avia-image-container-inner\"><div class=\"avia-image-overlay-wrap\"><img decoding=\"async\" fetchpriority=\"high\" class='wp-image-24810 avia-img-lazy-loading-not-24810 avia_image ' src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-14_130612539.png\" alt='Gaudi2 pomer vykonu 1' title='obr\u00e1zek_2022-12-14_130612539'  height=\"439\" width=\"741\"  itemprop=\"thumbnailUrl\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-14_130612539.png 741w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-14_130612539-300x178.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-14_130612539-705x418.png 705w\" sizes=\"(max-width: 741px) 100vw, 741px\" \/><\/div><\/div><\/div><\/div><div  class='flex_column av-vnsrlf-7915c728db78865c6dc784c4a3aebc02 av_one_half  avia-builder-el-30  el_after_av_one_half  el_before_av_one_full  flex_column_div  column-top-margin'     ><style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-lbnn8apz-692dcc260ddb3c2250fd0e0b949fc7a1\">\n.avia-image-container.av-lbnn8apz-692dcc260ddb3c2250fd0e0b949fc7a1 img.avia_image{\nbox-shadow:none;\n}\n.avia-image-container.av-lbnn8apz-692dcc260ddb3c2250fd0e0b949fc7a1 .av-image-caption-overlay-center{\ncolor:#ffffff;\n}\n<\/style>\n<div  class='avia-image-container av-lbnn8apz-692dcc260ddb3c2250fd0e0b949fc7a1 av-styling- avia-align-center  avia-builder-el-31  avia-builder-el-no-sibling '   itemprop=\"image\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/ImageObject\" ><div class=\"avia-image-container-inner\"><div class=\"avia-image-overlay-wrap\"><img decoding=\"async\" fetchpriority=\"high\" class='wp-image-24812 avia-img-lazy-loading-not-24812 avia_image ' src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-14_130651194.png\" alt='Gaudi2 pomer vykonu 2' title='obr\u00e1zek_2022-12-14_130651194'  height=\"450\" width=\"703\"  itemprop=\"thumbnailUrl\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-14_130651194.png 703w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-14_130651194-300x192.png 300w\" sizes=\"(max-width: 703px) 100vw, 703px\" \/><\/div><\/div><\/div><\/div><div  class='flex_column av-m28s37-c27af93a5422852d16e7e9e9e7b3e62b av_one_full  avia-builder-el-32  el_after_av_one_half  avia-builder-el-last  first flex_column_div  column-top-margin'     ><section  class='av_textblock_section av-lbnpnqnv-baf6361137efbf8252ee7c64b114bb99 '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock h-decor'  itemprop=\"text\" ><h3>RoCE2 scaling<\/h3>\n<p>Habana products have a simple and effective solution for scaling training and compute capacity on Gaudi2 by integrating 24 100-Gigabit RDMA over Converged Ethernet (RoCE2) ports on the chip card, more than double the number of the first generation. 21 ports on each Gaudi2 processor are dedicated to interconnects between up to seven other processors in an all-to-all star topology across the server. The remaining three ports are for scalable output that can reach up to 2.4 Tbps for an entire eight-card system (HLS-Gaudi2). To facilitate the design of more complex solutions, Habana offers a complete eight-chip solution directly on a single board. Through the industry standard RoCE, end customers can create Gaudi2 systems exactly to their needs for a DL cluster and scale from one to thousands of Gaudi2.<\/p>\n<\/div><\/section><\/div>\n<\/div><\/div><\/div><!-- close content main div --><\/div><\/div><div id='after_section_3'  class='main_color av_default_container_wrap container_wrap fullsize'  ><div class='container av-section-cont-open' ><div class='template-page content  av-content-full alpha units'><div class='post-entry post-entry-type-page post-entry-38117'><div class='entry-content-wrapper clearfix'>\n<div  class='flex_column av-o7vpdf-3b189c27a356058b978029b84095fef7 av_one_full  avia-builder-el-34  el_after_av_section  el_before_av_one_full  avia-builder-el-first  first flex_column_div  '     ><section  id=\"amd--mi250x\"  class='av_textblock_section av-lbngq4jx-e7e1045080892cadb208db93bf0560e2 '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock h-decor'  itemprop=\"text\" ><h2>AMD MI 300X<\/h2>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-26596 alignright\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/obrazek_2023-06-15_115136351-1030x551.png\" alt=\"AMD MI 300X announcement\" width=\"424\" height=\"227\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/obrazek_2023-06-15_115136351-1030x551.png 1030w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/obrazek_2023-06-15_115136351-300x160.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/obrazek_2023-06-15_115136351-768x410.png 768w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/obrazek_2023-06-15_115136351-705x377.png 705w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/obrazek_2023-06-15_115136351.png 1044w\" sizes=\"auto, (max-width: 424px) 100vw, 424px\" \/><\/p>\n<p>This newly announced accelerator is being developed by AMD for the purpose of training artificial intelligence, as the market is currently <em>booming<\/em> in this sector. The MI 300X accelerators are intended to be a direct competitor to NVIDIA&#8217;s H100 accelerators. The MI 300X cards are composed of 12 chipsets that are built on a 5nm\/6nm manufacturing process. Overall, we can notice that the MI 300X cards are essentially MI 300A APUs without the &#8220;APU&#8221; &#8211; that is, they are a standalone card. The MI 300X is also intended to compete with the H100 cards with its huge memory capacity of 192GB of HBM3. The card is expected to be launched in Q3 2023 and commercially deployed in the planned exa-scale <strong>El Capitan<\/strong> supercomputer. <\/p>\n<h2>AMD MI 250X<img loading=\"lazy\" decoding=\"async\" class=\" wp-image-24842 alignright\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_142623406.png\" alt=\"AMD Instinct MI250X\" width=\"360\" height=\"215\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_142623406.png 620w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_142623406-300x180.png 300w\" sizes=\"auto, (max-width: 360px) 100vw, 360px\" \/><\/h2>\n<p>AMD&#8217;s Instinct MI 2xx accelerators are positioned as a direct competitor to NVIDIA&#8217;s A100 SXM cards and are built on the new CDNA2 architecture, which is an enhancement to the previous, proven CDNA architecture. Compared to it, it has improved efficiency and power consumption relative to performance, but the architecture itself has also received minor and major modifications. One of them, for example, is the &#8220;Elevated Fanout Bridge 2.5D&#8221;, which is an alternative type of SMD interconnect between the accelerator core, memory and other components on the card. First of all, it saves manufacturing costs as it is made from more affordable substrates, but it is also much easier to manufacture. Furthermore, EFB2.5D makes it much easier to scale the hardware.    <\/p>\n<h3>Frontier &#8211; AMD \u00d7 HPE<\/h3>\n<p>The world&#8217;s most powerful supercomputer, <a href=\"https:\/\/www.olcf.ornl.gov\/frontier\/\">Frontier<\/a> is built on 3rd generation AMD EPYC processors and AMD Instinct MI250 accelerators in an HP Enterprise chassis. With a configuration of four of these accelerators per compute node and AMD Infinity Fabric interconnects, Frontier achieves a peak performance of 1.6 ExaFLOP, making it the world&#8217;s first Exa-scale supercomputer.  <\/p>\n<h3>AMD Infinity architecture<\/h3>\n<p>With the help of several new features and technologies from AMD, the technology solutions with 3rd generation AMD EPYC processors and AMD Instinct accelerators can get the most out of the x86 architecture. 7nm SoC systems provide up to double the performance of previous generations, resulting in up to four times the TFLOPs in AI applications, so that MI250X accelerators have a theoretical point-to-point throughput of around 800 GB\/s on their I\/O interface.<\/p>\n<\/div><\/section><\/div><div  class='flex_column av-iqvdar-5ea8c3ceaf2a1f77cb2d22cf47a31e53 av_one_full  avia-builder-el-36  el_after_av_one_full  el_before_av_section  avia-builder-el-last  first flex_column_div  column-top-margin'     ><section  id=\"nvidia--a100--h100\"  class='av_textblock_section av-lbp9r0m4-56c431bf7f579286747baf5e55374fdb '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock h-decor'  itemprop=\"text\" ><h2>NVIDIA A100 \/ H100<img loading=\"lazy\" decoding=\"async\" class=\" wp-image-24847 alignright\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_170523241-1030x579.png\" alt=\"NVIDIA A100 \/ H100\" width=\"496\" height=\"279\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_170523241-1030x579.png 1030w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_170523241-300x169.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_170523241-768x432.png 768w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_170523241-705x397.png 705w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_170523241.png 1200w\" sizes=\"auto, (max-width: 496px) 100vw, 496px\" \/><\/h2>\n<p>The NVIDIA A100 TC GPU (Tensor Core GPU) accelerator provides the acceleration needed on any scaled system for flexible AI, data analytics, and HPC development and training solutions. The A100 card is built on the Ampere architecture, which delivers up to 20x more performance over the previous generation in some applications. With 80GB of HBM2e memory and a bitrate of over 2TBps, it is the ideal choice for running the largest models and datasets. In contrast, NVIDIA&#8217;s H100 is not a replacement, but a complement to NVIDIA&#8217;s version; it is built on the newer 4nm Hopper architecture versus 7nm Ampere. Performance has more than doubled, plus it has much higher efficiency thanks to a smaller manufacturing process. With the new NVIDIA NVLink Switch system, up to 256 H100 cards can also be clustered, which can bring exponential speedups in NLP or any AI application, for example.  <\/p>\n<h3>AI, Deep Learning, Machine Learning, Data Processing, NLP<\/h3>\n<p><img loading=\"lazy\" decoding=\"async\" class=\" wp-image-24851 alignright\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_175247855-1030x508.png\" alt=\"AI Training speed difference between NVIDIA A100 and H100\" width=\"466\" height=\"230\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_175247855-1030x508.png 1030w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_175247855-300x148.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_175247855-768x379.png 768w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_175247855-1536x758.png 1536w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_175247855-1500x740.png 1500w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_175247855-705x348.png 705w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_175247855.png 1877w\" sizes=\"auto, (max-width: 466px) 100vw, 466px\" \/><\/p>\n<p>Both cards are a great choice for neural network training solutions for any sector of AI, data analytics, data processing or other applications that work with huge data sets, for which the fastest possible interconnection between cards and high memory capacity is essential. In addition, the NVIDIA H100 features fourth-generation Tensor cores and the Transformer Engine with FP8 precision, which can achieve up to nine times the training speed on MoE models. These parameters, plus fourth-generation NVLink, which has a data throughput of up to 900 GBps for inter-GPU connections, the NVLINK Switch system that accelerates communication between each individual card in a node, PCIe 5th generation or SXM5, and NVIDIA MagnumIO software, can provide efficient scaling from smaller industries to massive, unified GPU clusters.<\/p>\n<\/div><\/section><\/div><\/p>\n<\/div><\/div><\/div><!-- close content main div --><\/div><\/div>\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-livtl6ye-6e8a77dd37d5ff66fb86d5caf9a1eb49\">\n.avia-section.av-livtl6ye-6e8a77dd37d5ff66fb86d5caf9a1eb49{\nbackground-color:#f7f7f7;\nbackground-image:unset;\n}\n<\/style>\n<div id='av_section_4'  class='avia-section av-livtl6ye-6e8a77dd37d5ff66fb86d5caf9a1eb49 main_color avia-section-default avia-no-border-styling  avia-builder-el-38  el_after_av_one_full  el_before_av_one_full  avia-bg-style-scroll container_wrap fullsize'  ><div class='container av-section-cont-open' ><div class='template-page content  av-content-full alpha units'><div class='post-entry post-entry-type-page post-entry-38117'><div class='entry-content-wrapper clearfix'>\n\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-livtdpzm-07c2473ce83f3104b19df765510db1df\">\n#top .av-special-heading.av-livtdpzm-07c2473ce83f3104b19df765510db1df{\npadding-bottom:10px;\n}\nbody .av-special-heading.av-livtdpzm-07c2473ce83f3104b19df765510db1df .av-special-heading-tag .heading-char{\nfont-size:25px;\n}\n.av-special-heading.av-livtdpzm-07c2473ce83f3104b19df765510db1df .av-subheading{\nfont-size:15px;\n}\n<\/style>\n<div  class='av-special-heading av-livtdpzm-07c2473ce83f3104b19df765510db1df av-special-heading-h3  avia-builder-el-39  el_before_av_textblock  avia-builder-el-first  h-decor'><h3 class='av-special-heading-tag '  itemprop=\"headline\"  >Z\u010cU + TechSoft: testing NVIDIA accelerators in Ansys Fluent<\/h3><div class=\"special-heading-border\"><div class=\"special-heading-inner-border\"><\/div><\/div><\/div>\n<section  class='av_textblock_section av-livtx4yf-ac27b0d618cb7f2ca65675fd9fda3f87 '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock'  itemprop=\"text\" ><p><em>The Technology Research Centre of the University of West Bohemia in Pilsen<\/em> together with <strong>TechSoft<\/strong>, supported by M Computers, carried out extensive testing of NVIDIA graphics accelerators in the <strong>Ansys Fluent R1<\/strong> simulation program. Some of the comparisons and results are shown below:<\/p>\n<\/div><\/section>\n<div  class='flex_column av-120f7w3-0303b5aa13c6b4333c0e4edcaa449629 av_one_half  avia-builder-el-41  el_after_av_textblock  el_before_av_one_half  first flex_column_div  column-top-margin'     ><style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-livu1bsg-eb1119384051eea0ca3e37447929a7a8\">\n.avia-image-container.av-livu1bsg-eb1119384051eea0ca3e37447929a7a8 img.avia_image{\nbox-shadow:none;\n}\n.avia-image-container.av-livu1bsg-eb1119384051eea0ca3e37447929a7a8 .av-image-caption-overlay-center{\ncolor:#ffffff;\n}\n<\/style>\n<div  class='avia-image-container av-livu1bsg-eb1119384051eea0ca3e37447929a7a8 av-styling- avia-align-center  avia-builder-el-42  avia-builder-el-no-sibling '   itemprop=\"image\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/ImageObject\" ><div class=\"avia-image-container-inner\"><div class=\"avia-image-overlay-wrap\"><img decoding=\"async\" fetchpriority=\"high\" class='wp-image-26425 avia-img-lazy-loading-not-26425 avia_image ' src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165305933-1030x577.png\" alt='' title='image_2023-06-14_165305933'  height=\"577\" width=\"1030\"  itemprop=\"thumbnailUrl\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165305933-1030x577.png 1030w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165305933-300x168.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165305933-768x430.png 768w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165305933-1536x861.png 1536w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165305933-1500x840.png 1500w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165305933-705x395.png 705w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165305933.png 1608w\" sizes=\"(max-width: 1030px) 100vw, 1030px\" \/><\/div><\/div><\/div><\/div><div  class='flex_column av-i9lsr7-dabdf2bf51476623e8415dba778343a3 av_one_half  avia-builder-el-43  el_after_av_one_half  el_before_av_hr  flex_column_div  column-top-margin'     ><style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-livu2n6x-1deb5fed8634a5b396a2135d0375a12d\">\n.avia-image-container.av-livu2n6x-1deb5fed8634a5b396a2135d0375a12d img.avia_image{\nbox-shadow:none;\n}\n.avia-image-container.av-livu2n6x-1deb5fed8634a5b396a2135d0375a12d .av-image-caption-overlay-center{\ncolor:#ffffff;\n}\n<\/style>\n<div  class='avia-image-container av-livu2n6x-1deb5fed8634a5b396a2135d0375a12d av-styling- avia-align-center  avia-builder-el-44  avia-builder-el-no-sibling '   itemprop=\"image\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/ImageObject\" ><div class=\"avia-image-container-inner\"><div class=\"avia-image-overlay-wrap\"><img decoding=\"async\" fetchpriority=\"high\" class='wp-image-26429 avia-img-lazy-loading-not-26429 avia_image ' src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165539148-1030x579.png\" alt='' title='image_2023-06-14_165539148'  height=\"579\" width=\"1030\"  itemprop=\"thumbnailUrl\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165539148-1030x579.png 1030w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165539148-300x169.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165539148-768x431.png 768w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165539148-1536x863.png 1536w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165539148-1500x843.png 1500w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165539148-705x396.png 705w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2023\/06\/image_2023-06-14_165539148.png 1602w\" sizes=\"(max-width: 1030px) 100vw, 1030px\" \/><\/div><\/div><\/div><\/div>\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-livulff0-f3e8cb73cf7b355a87b01500109a924a\">\n#top .hr.hr-invisible.av-livulff0-f3e8cb73cf7b355a87b01500109a924a{\nheight:50px;\n}\n<\/style>\n<div  class='hr av-livulff0-f3e8cb73cf7b355a87b01500109a924a hr-invisible  avia-builder-el-45  el_after_av_one_half  el_before_av_textblock '><span class='hr-inner '><span class=\"hr-inner-style\"><\/span><\/span><\/div>\n<section  class='av_textblock_section av-livuka9f-0e23101173f420eba10f0e5a999d18cf '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock'  itemprop=\"text\" ><p>We would like to thank our colleagues Jind\u0159ich K\u0148ourek from Z\u010cU and Richard Paulas from TechSoft engineering s.r.o. for excellent processing and presentation of the data obtained with the help of solutions supplied by M Computers s.r.o.!<\/p>\n<\/div><\/section>\n<\/div><\/div><\/div><!-- close content main div --><\/div><\/div><div id='after_section_4'  class='main_color av_default_container_wrap container_wrap fullsize'  ><div class='container av-section-cont-open' ><div class='template-page content  av-content-full alpha units'><div class='post-entry post-entry-type-page post-entry-38117'><div class='entry-content-wrapper clearfix'>\n<div  class='flex_column av-oi02nn-2693a82210c50028d9e127f81f280077 av_one_full  avia-builder-el-47  el_after_av_section  el_before_av_one_full  avia-builder-el-first  first flex_column_div  '     ><section  id=\"graphcore--c600\"  class='av_textblock_section av-lbpe6emz-95c29a542a76c038120a455cd33352c3 '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock h-decor'  itemprop=\"text\" ><h2>Graphcore C600<img loading=\"lazy\" decoding=\"async\" class=\" wp-image-24855 alignright\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/Graphcore_C600.png\" alt=\"Graphcore C600 IPU\" width=\"429\" height=\"253\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/Graphcore_C600.png 957w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/Graphcore_C600-300x177.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/Graphcore_C600-768x453.png 768w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/Graphcore_C600-705x415.png 705w\" sizes=\"auto, (max-width: 429px) 100vw, 429px\" \/><\/h2>\n<p>The Graphcore C600 accelerator is an IPU processor PCIe card with high performance for machine learning. The card is built on the new Mk2 architecture, which adds support for FP8 precision, for example, which can further accelerate some AI calculations. The card generally serves as the aforementioned IPU, Infrastructure Processing Unit, which is a card that mainly takes care of lightening the load of the CPU and preprocessing some data that would otherwise take the CPU many times longer. &#8220;Fun fact&#8221;; if you think Graphcore&#8217;s PCIe accelerators look rather unconventional, you&#8217;re absolutely right; their design was engineered by an AI algorithm The card is made in a dual-slot, full-height format and can be deployed in standard server chassis, typically two to 2U servers and up to eight to 4U.  <\/p>\n<h3>IPU-Link<\/h3>\n<p>The IPU-Link technology was developed by Graphcore primarily for interconnecting Accelerators to each other, it is a similar principle to e.g. NVLink. However, IPU-Link has the ability to connect up to eight cards, improving the performance and scaling of solutions based on Graphcore accelerators.  <\/p>\n<h3>In &#8211; Processor Memory (IPM)<\/h3>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-24878 alignright\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-16_132103501.png\" alt=\"Graphcore IPU memory management\" width=\"458\" height=\"305\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-16_132103501.png 960w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-16_132103501-300x200.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-16_132103501-768x512.png 768w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-16_132103501-705x470.png 705w\" sizes=\"auto, (max-width: 458px) 100vw, 458px\" \/><\/p>\n<p>Graphcore&#8217;s IPU architecture has another unique feature, which is memory integrated directly into the core of the computing unit. It can be compared to the Lx cache and also has similar advantages &#8211; great response, data flow and speed. In the Mk2 architecture, each IPU core has 900 MB of its own memory, + up to 448 GB of streaming\/system memory that acts as a buffer.<\/p>\n<\/div><\/section><\/div>\n<div  class='flex_column av-w2qqvn-142a16617181ae5acd37cc1ba1537719 av_one_full  avia-builder-el-49  el_after_av_one_full  el_before_av_section  avia-builder-el-last  first flex_column_div  column-top-margin'     ><section  id=\"cerebras--wse2\"  class='av_textblock_section av-lbuphdn3-b934389c2cfeddc2d3824bd4fc7b46c7 '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock h-decor'  itemprop=\"text\" ><h2>Cerebras WSE-2<img loading=\"lazy\" decoding=\"async\" class=\" wp-image-24920 alignright\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-19_123616784.png\" alt=\"Cerebras WSE-2 wafer scale system\" width=\"310\" height=\"308\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-19_123616784.png 943w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-19_123616784-300x297.png 300w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-19_123616784-80x80.png 80w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-19_123616784-768x761.png 768w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-19_123616784-36x36.png 36w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-19_123616784-180x180.png 180w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/image_2022-12-19_123616784-705x699.png 705w\" sizes=\"auto, (max-width: 310px) 100vw, 310px\" \/><\/h2>\n<p>In the case of Cerebras, this is completely unique &#8211; a chip with an area of 46.225 <sup>mm2<\/sup>, which is more than 50 times larger than the largest graphics chip. In such a large wafer, 850,000 cores have been deployed with a 7nm process, totalling over 2.6 trillion transistors. The <strong>WSE<\/strong> (Wafer Scale Engine) concept, more or less the entire compute unit of a supercomputer on one massive chip, has countless advantages over clustered systems, the main one being latency and data flow between cores. Since they are all printed on a single surface, the communication paths between units are completely minimal, compared to Fabric interconnects in traditional systems, which restrict the flow and slow it down slightly.<\/p>\n<\/div><\/section><\/div>\n<\/div><\/div><\/div><!-- close content main div --><\/div><\/div><div id='av_section_5'  class='avia-section av-2fx9flf-330b20e50bafb3a93aff5a64ea183558 main_color avia-section-default avia-no-border-styling  avia-builder-el-51  el_after_av_one_full  avia-builder-el-last  avia-bg-style-scroll container_wrap fullsize'  ><div class='container av-section-cont-open' ><div class='template-page content  av-content-full alpha units'><div class='post-entry post-entry-type-page post-entry-38117'><div class='entry-content-wrapper clearfix'>\n<section  class='av_textblock_section av-2ai9yz7-41829768f8693af36bb37f3595a9dc3c '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock h-decor'  itemprop=\"text\" ><h2>News<\/h2>\n<\/div><\/section>\n<div  class='flex_column av-1gketeb-290cb893e3afc580b80d626a6903173c av_one_full  avia-builder-el-53  el_after_av_textblock  avia-builder-el-last  post-slider-col3 first flex_column_div  column-top-margin'     ><p><div  data-slideshow-options=\"{&quot;animation&quot;:&quot;fade&quot;,&quot;autoplay&quot;:true,&quot;loop_autoplay&quot;:&quot;endless&quot;,&quot;interval&quot;:&quot;5&quot;,&quot;loop_manual&quot;:&quot;manual-endless&quot;,&quot;autoplay_stopper&quot;:false,&quot;noNavigation&quot;:false,&quot;show_slide_delay&quot;:90}\" class='avia-content-slider avia-content-slider-active avia-content-slider1 avia-content-slider-even  avia-builder-el-54  el_before_av_textblock  avia-builder-el-first  new-post-grid av-slideshow-ui av-control-default   av-no-slider-navigation av-slideshow-autoplay av-loop-endless av-loop-manual-endless '  itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/Blog\" ><div class=\"avia-content-slider-inner\"><div class=\"slide-entry-wrap\"><article class='slide-entry flex_column  post-entry post-entry-112753 slide-entry-overview slide-loop-1 slide-parity-odd  av_one_fourth first real-thumbnail posttype-post post-format-standard'  itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><a href='https:\/\/mcomputers.cz\/en\/2026\/04\/23\/we-won-three-awards-at-the-nvidia-partner-awards-emea-2026\/' data-rel='slide-1' class='slide-image' title='We won three awards at the NVIDIA Partner Awards EMEA 2026'><img decoding=\"async\" fetchpriority=\"high\" width=\"710\" height=\"375\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2026\/04\/NVIDIA-Partner-Awards-EMEA-2026-710x375.png\" class=\"wp-image-112689 avia-img-lazy-loading-not-112689 attachment-magazine size-magazine wp-post-image\" alt=\"\" \/><\/a><div class=\"slide-content\"><header class=\"entry-content-header\" aria-label=\"Slide: We won three awards at the NVIDIA Partner Awards EMEA 2026\"><h3 class='slide-entry-title entry-title '  itemprop=\"headline\" ><a href='https:\/\/mcomputers.cz\/en\/2026\/04\/23\/we-won-three-awards-at-the-nvidia-partner-awards-emea-2026\/' title='We won three awards at the NVIDIA Partner Awards EMEA 2026'>We won three awards at the NVIDIA Partner Awards EMEA 2026<\/a><\/h3><span class=\"av-vertical-delimiter\"><\/span><\/header><div class=\"slide-meta\"><time class='slide-meta-time updated'  itemprop=\"datePublished\" datetime=\"2026-04-23T14:03:57+01:00\" >23.4.2026<\/time><\/div><div class='slide-entry-excerpt entry-content'  itemprop=\"text\" >\r\n\r\n\r\n\r\n<\/div><\/div><footer class=\"entry-footer\"><\/footer><span class='hidden'>\n\t\t\t\t<span class='av-structured-data'  itemprop=\"image\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/ImageObject\" >\n\t\t\t\t\t\t<span itemprop='url'>https:\/\/mcomputers.cz\/wp-content\/uploads\/2026\/04\/NVIDIA-Partner-Awards-EMEA-2026.png<\/span>\n\t\t\t\t\t\t<span itemprop='height'>731<\/span>\n\t\t\t\t\t\t<span itemprop='width'>1030<\/span>\n\t\t\t\t<\/span>\n\t\t\t\t<span class='av-structured-data'  itemprop=\"publisher\" itemtype=\"https:\/\/schema.org\/Organization\" itemscope=\"itemscope\" >\n\t\t\t\t\t\t<span itemprop='name'>Jitka Bl\u00e1hov\u00e1<\/span>\n\t\t\t\t\t\t<span itemprop='logo' itemscope itemtype='https:\/\/schema.org\/ImageObject'>\n\t\t\t\t\t\t\t<span itemprop='url'>http:\/\/mcomputers.cz\/wp-content\/uploads\/2024\/12\/obrazek_2024-12-03_164614578-e1733305217994.png<\/span>\n\t\t\t\t\t\t<\/span>\n\t\t\t\t<\/span><span class='av-structured-data'  itemprop=\"author\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/Person\" ><span itemprop='name'>Jitka Bl\u00e1hov\u00e1<\/span><\/span><span class='av-structured-data'  itemprop=\"datePublished\" datetime=\"2023-06-13T14:46:36+01:00\" >2026-04-23 14:03:57<\/span><span class='av-structured-data'  itemprop=\"dateModified\" itemtype=\"https:\/\/schema.org\/dateModified\" >2026-04-24 16:50:44<\/span><span class='av-structured-data'  itemprop=\"mainEntityOfPage\" itemtype=\"https:\/\/schema.org\/mainEntityOfPage\" ><span itemprop='name'>We won three awards at the NVIDIA Partner Awards EMEA 2026<\/span><\/span><\/span><\/article><article class='slide-entry flex_column  post-entry post-entry-112545 slide-entry-overview slide-loop-2 slide-parity-even  av_one_fourth  real-thumbnail posttype-post post-format-standard'  itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><a href='https:\/\/mcomputers.cz\/en\/2026\/04\/01\/with-prg-ai-and-hpe-we-organized-the-nvidia-gtc-2026-watch\/' data-rel='slide-1' class='slide-image' title='With prg.ai and HPE we organized the NVIDIA GTC 2026 Watch'><img decoding=\"async\" fetchpriority=\"high\" width=\"710\" height=\"375\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2026\/04\/PXL_20260331_133027775.MP_-710x375.jpg\" class=\"wp-image-112361 avia-img-lazy-loading-not-112361 attachment-magazine size-magazine wp-post-image\" alt=\"\" \/><\/a><div class=\"slide-content\"><header class=\"entry-content-header\" aria-label=\"Slide: With prg.ai and HPE we organized the NVIDIA GTC 2026 Watch\"><h3 class='slide-entry-title entry-title '  itemprop=\"headline\" ><a href='https:\/\/mcomputers.cz\/en\/2026\/04\/01\/with-prg-ai-and-hpe-we-organized-the-nvidia-gtc-2026-watch\/' title='With prg.ai and HPE we organized the NVIDIA GTC 2026 Watch'>With prg.ai and HPE we organized the NVIDIA GTC 2026 Watch<\/a><\/h3><span class=\"av-vertical-delimiter\"><\/span><\/header><div class=\"slide-meta\"><time class='slide-meta-time updated'  itemprop=\"datePublished\" datetime=\"2026-04-01T13:57:21+01:00\" >1.4.2026<\/time><\/div><div class='slide-entry-excerpt entry-content'  itemprop=\"text\" >\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n<\/div><\/div><footer class=\"entry-footer\"><\/footer><span class='hidden'>\n\t\t\t\t<span class='av-structured-data'  itemprop=\"image\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/ImageObject\" >\n\t\t\t\t\t\t<span itemprop='url'>https:\/\/mcomputers.cz\/wp-content\/uploads\/2026\/04\/PXL_20260331_133027775.MP_-scaled.jpg<\/span>\n\t\t\t\t\t\t<span itemprop='height'>1928<\/span>\n\t\t\t\t\t\t<span itemprop='width'>2560<\/span>\n\t\t\t\t<\/span>\n\t\t\t\t<span class='av-structured-data'  itemprop=\"publisher\" itemtype=\"https:\/\/schema.org\/Organization\" itemscope=\"itemscope\" >\n\t\t\t\t\t\t<span itemprop='name'>Jitka Bl\u00e1hov\u00e1<\/span>\n\t\t\t\t\t\t<span itemprop='logo' itemscope itemtype='https:\/\/schema.org\/ImageObject'>\n\t\t\t\t\t\t\t<span itemprop='url'>http:\/\/mcomputers.cz\/wp-content\/uploads\/2024\/12\/obrazek_2024-12-03_164614578-e1733305217994.png<\/span>\n\t\t\t\t\t\t<\/span>\n\t\t\t\t<\/span><span class='av-structured-data'  itemprop=\"author\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/Person\" ><span itemprop='name'>Jitka Bl\u00e1hov\u00e1<\/span><\/span><span class='av-structured-data'  itemprop=\"datePublished\" datetime=\"2023-06-13T14:46:36+01:00\" >2026-04-01 13:57:21<\/span><span class='av-structured-data'  itemprop=\"dateModified\" itemtype=\"https:\/\/schema.org\/dateModified\" >2026-04-13 08:18:02<\/span><span class='av-structured-data'  itemprop=\"mainEntityOfPage\" itemtype=\"https:\/\/schema.org\/mainEntityOfPage\" ><span itemprop='name'>With prg.ai and HPE we organized the NVIDIA GTC 2026 Watch<\/span><\/span><\/span><\/article><article class='slide-entry flex_column  post-entry post-entry-112287 slide-entry-overview slide-loop-3 slide-parity-odd  av_one_fourth  real-thumbnail posttype-post post-format-standard'  itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><a href='https:\/\/mcomputers.cz\/en\/2026\/03\/30\/easter-visit-from-the-childrens-home-cerna-voda-2\/' data-rel='slide-1' class='slide-image' title='Easter visit from the Children&#8217;s Home \u010cern\u00e1 Voda'><img decoding=\"async\" fetchpriority=\"high\" width=\"710\" height=\"375\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2026\/03\/IMG_8485_u2-710x375.jpg\" class=\"wp-image-112245 avia-img-lazy-loading-not-112245 attachment-magazine size-magazine wp-post-image\" alt=\"\" \/><\/a><div class=\"slide-content\"><header class=\"entry-content-header\" aria-label=\"Slide: Easter visit from the Children&#039;s Home \u010cern\u00e1 Voda\"><h3 class='slide-entry-title entry-title '  itemprop=\"headline\" ><a href='https:\/\/mcomputers.cz\/en\/2026\/03\/30\/easter-visit-from-the-childrens-home-cerna-voda-2\/' title='Easter visit from the Children&#8217;s Home \u010cern\u00e1 Voda'>Easter visit from the Children&#8217;s Home \u010cern\u00e1 Voda<\/a><\/h3><span class=\"av-vertical-delimiter\"><\/span><\/header><div class=\"slide-meta\"><time class='slide-meta-time updated'  itemprop=\"datePublished\" datetime=\"2026-03-30T17:00:19+01:00\" >30.3.2026<\/time><\/div><div class='slide-entry-excerpt entry-content'  itemprop=\"text\" >\n\n\n\n\n\n\n\n<\/div><\/div><footer class=\"entry-footer\"><\/footer><span class='hidden'>\n\t\t\t\t<span class='av-structured-data'  itemprop=\"image\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/ImageObject\" >\n\t\t\t\t\t\t<span itemprop='url'>https:\/\/mcomputers.cz\/wp-content\/uploads\/2026\/03\/IMG_8485_u2.jpg<\/span>\n\t\t\t\t\t\t<span itemprop='height'>1134<\/span>\n\t\t\t\t\t\t<span itemprop='width'>1536<\/span>\n\t\t\t\t<\/span>\n\t\t\t\t<span class='av-structured-data'  itemprop=\"publisher\" itemtype=\"https:\/\/schema.org\/Organization\" itemscope=\"itemscope\" >\n\t\t\t\t\t\t<span itemprop='name'>Jitka Bl\u00e1hov\u00e1<\/span>\n\t\t\t\t\t\t<span itemprop='logo' itemscope itemtype='https:\/\/schema.org\/ImageObject'>\n\t\t\t\t\t\t\t<span itemprop='url'>http:\/\/mcomputers.cz\/wp-content\/uploads\/2024\/12\/obrazek_2024-12-03_164614578-e1733305217994.png<\/span>\n\t\t\t\t\t\t<\/span>\n\t\t\t\t<\/span><span class='av-structured-data'  itemprop=\"author\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/Person\" ><span itemprop='name'>Jitka Bl\u00e1hov\u00e1<\/span><\/span><span class='av-structured-data'  itemprop=\"datePublished\" datetime=\"2023-06-13T14:46:36+01:00\" >2026-03-30 17:00:19<\/span><span class='av-structured-data'  itemprop=\"dateModified\" itemtype=\"https:\/\/schema.org\/dateModified\" >2026-03-30 17:05:29<\/span><span class='av-structured-data'  itemprop=\"mainEntityOfPage\" itemtype=\"https:\/\/schema.org\/mainEntityOfPage\" ><span itemprop='name'>Easter visit from the Children&#8217;s Home \u010cern\u00e1 Voda<\/span><\/span><\/span><\/article><article class='slide-entry flex_column  post-entry post-entry-112221 slide-entry-overview slide-loop-4 slide-parity-even  post-entry-last  av_one_fourth  real-thumbnail posttype-post post-format-standard'  itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><a href='https:\/\/mcomputers.cz\/en\/2026\/03\/27\/openclaw-now-on-nvidia-dgx-spark\/' data-rel='slide-1' class='slide-image' title='OpenClaw now on NVIDIA DGX Spark'><img decoding=\"async\" fetchpriority=\"high\" width=\"710\" height=\"375\" src=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2026\/03\/DGX-Spark-1-710x375.png\" class=\"wp-image-112188 avia-img-lazy-loading-not-112188 attachment-magazine size-magazine wp-post-image\" alt=\"\" srcset=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2026\/03\/DGX-Spark-1-710x375.png 710w, https:\/\/mcomputers.cz\/wp-content\/uploads\/2026\/03\/DGX-Spark-1-300x157.png 300w\" sizes=\"(max-width: 710px) 100vw, 710px\" \/><\/a><div class=\"slide-content\"><header class=\"entry-content-header\" aria-label=\"Slide: OpenClaw now on NVIDIA DGX Spark\"><h3 class='slide-entry-title entry-title '  itemprop=\"headline\" ><a href='https:\/\/mcomputers.cz\/en\/2026\/03\/27\/openclaw-now-on-nvidia-dgx-spark\/' title='OpenClaw now on NVIDIA DGX Spark'>OpenClaw now on NVIDIA DGX Spark<\/a><\/h3><span class=\"av-vertical-delimiter\"><\/span><\/header><div class=\"slide-meta\"><time class='slide-meta-time updated'  itemprop=\"datePublished\" datetime=\"2026-03-27T14:42:53+01:00\" >27.3.2026<\/time><\/div><div class='slide-entry-excerpt entry-content'  itemprop=\"text\" >\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n<\/div><\/div><footer class=\"entry-footer\"><\/footer><span class='hidden'>\n\t\t\t\t<span class='av-structured-data'  itemprop=\"image\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/ImageObject\" >\n\t\t\t\t\t\t<span itemprop='url'>https:\/\/mcomputers.cz\/wp-content\/uploads\/2026\/03\/DGX-Spark-1.png<\/span>\n\t\t\t\t\t\t<span itemprop='height'>627<\/span>\n\t\t\t\t\t\t<span itemprop='width'>1200<\/span>\n\t\t\t\t<\/span>\n\t\t\t\t<span class='av-structured-data'  itemprop=\"publisher\" itemtype=\"https:\/\/schema.org\/Organization\" itemscope=\"itemscope\" >\n\t\t\t\t\t\t<span itemprop='name'>Jitka Bl\u00e1hov\u00e1<\/span>\n\t\t\t\t\t\t<span itemprop='logo' itemscope itemtype='https:\/\/schema.org\/ImageObject'>\n\t\t\t\t\t\t\t<span itemprop='url'>http:\/\/mcomputers.cz\/wp-content\/uploads\/2024\/12\/obrazek_2024-12-03_164614578-e1733305217994.png<\/span>\n\t\t\t\t\t\t<\/span>\n\t\t\t\t<\/span><span class='av-structured-data'  itemprop=\"author\" itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/Person\" ><span itemprop='name'>Jitka Bl\u00e1hov\u00e1<\/span><\/span><span class='av-structured-data'  itemprop=\"datePublished\" datetime=\"2023-06-13T14:46:36+01:00\" >2026-03-27 14:42:53<\/span><span class='av-structured-data'  itemprop=\"dateModified\" itemtype=\"https:\/\/schema.org\/dateModified\" >2026-03-30 15:36:06<\/span><span class='av-structured-data'  itemprop=\"mainEntityOfPage\" itemtype=\"https:\/\/schema.org\/mainEntityOfPage\" ><span itemprop='name'>OpenClaw now on NVIDIA DGX Spark<\/span><\/span><\/span><\/article><\/div><\/div><\/div><br \/>\n<section  class='av_textblock_section av-psqln7-55710d2cef64e934a839659da9257874 '   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock'  itemprop=\"text\" ><p style=\"text-align: right;\"><a href=\"https:\/\/mcomputers.cz\/o-nas\/novinky\/\">View more news<\/a><\/p>\n<\/div><\/section><\/p><\/div>\n\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":12,"featured_media":24843,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[75],"tags":[],"class_list":["post-38117","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news"],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v27.0 (Yoast SEO v27.4) - https:\/\/yoast.com\/product\/yoast-seo-premium-wordpress\/ -->\n<title>HPC Accelerator Comparison 2023 &#8211; M Computers s.r.o.<\/title>\n<meta name=\"description\" content=\"Every year there are new accelerators for HPC and AI, ML and DL application development. That&#039;s why we compared their performance, parameters and new features.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"HPC Accelerator Comparison 2023\" \/>\n<meta property=\"og:description\" content=\"Every year there are new accelerators for HPC and AI, ML and DL application development. That&#039;s why we compared their performance, parameters and new features.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/\" \/>\n<meta property=\"og:site_name\" content=\"M Computers s.r.o.\" \/>\n<meta property=\"article:published_time\" content=\"2023-06-13T13:46:36+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2025-01-03T14:08:42+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_142623406.png\" \/>\n\t<meta property=\"og:image:width\" content=\"620\" \/>\n\t<meta property=\"og:image:height\" content=\"371\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Mikol\u00e1\u0161 Pe\u0161ek\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@m_computers_cz\" \/>\n<meta name=\"twitter:site\" content=\"@m_computers_cz\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Mikol\u00e1\u0161 Pe\u0161ek\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"98 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/en\\\/2023\\\/06\\\/13\\\/hpc-accelerator-comparison-2023\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/en\\\/2023\\\/06\\\/13\\\/hpc-accelerator-comparison-2023\\\/\"},\"author\":{\"name\":\"Mikol\u00e1\u0161 Pe\u0161ek\",\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/#\\\/schema\\\/person\\\/148aba8a2913606468a4d7043861a4f5\"},\"headline\":\"HPC Accelerator Comparison 2023\",\"datePublished\":\"2023-06-13T13:46:36+00:00\",\"dateModified\":\"2025-01-03T14:08:42+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/en\\\/2023\\\/06\\\/13\\\/hpc-accelerator-comparison-2023\\\/\"},\"wordCount\":15338,\"publisher\":{\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/en\\\/2023\\\/06\\\/13\\\/hpc-accelerator-comparison-2023\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/mcomputers.cz\\\/wp-content\\\/uploads\\\/2022\\\/12\\\/obrazek_2022-12-15_142623406.png\",\"articleSection\":[\"News\"],\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/en\\\/2023\\\/06\\\/13\\\/hpc-accelerator-comparison-2023\\\/\",\"url\":\"https:\\\/\\\/mcomputers.cz\\\/en\\\/2023\\\/06\\\/13\\\/hpc-accelerator-comparison-2023\\\/\",\"name\":\"HPC Accelerator Comparison 2023 &#8211; M Computers s.r.o.\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/en\\\/2023\\\/06\\\/13\\\/hpc-accelerator-comparison-2023\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/en\\\/2023\\\/06\\\/13\\\/hpc-accelerator-comparison-2023\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/mcomputers.cz\\\/wp-content\\\/uploads\\\/2022\\\/12\\\/obrazek_2022-12-15_142623406.png\",\"datePublished\":\"2023-06-13T13:46:36+00:00\",\"dateModified\":\"2025-01-03T14:08:42+00:00\",\"description\":\"Every year there are new accelerators for HPC and AI, ML and DL application development. That's why we compared their performance, parameters and new features.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/en\\\/2023\\\/06\\\/13\\\/hpc-accelerator-comparison-2023\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/mcomputers.cz\\\/en\\\/2023\\\/06\\\/13\\\/hpc-accelerator-comparison-2023\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/en\\\/2023\\\/06\\\/13\\\/hpc-accelerator-comparison-2023\\\/#primaryimage\",\"url\":\"https:\\\/\\\/mcomputers.cz\\\/wp-content\\\/uploads\\\/2022\\\/12\\\/obrazek_2022-12-15_142623406.png\",\"contentUrl\":\"https:\\\/\\\/mcomputers.cz\\\/wp-content\\\/uploads\\\/2022\\\/12\\\/obrazek_2022-12-15_142623406.png\",\"width\":620,\"height\":371,\"caption\":\"jzuk,u.\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/en\\\/2023\\\/06\\\/13\\\/hpc-accelerator-comparison-2023\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Dom\u016f\",\"item\":\"https:\\\/\\\/mcomputers.cz\\\/en\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"HPC Accelerator Comparison 2023\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/#website\",\"url\":\"https:\\\/\\\/mcomputers.cz\\\/\",\"name\":\"M Computers s.r.o.\",\"description\":\"M Computers - V\u00e1\u0161 pr\u016fvodce sv\u011btem IT\",\"publisher\":{\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/mcomputers.cz\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/#organization\",\"name\":\"M Computers s.r.o.\",\"url\":\"https:\\\/\\\/mcomputers.cz\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/mcomputers.cz\\\/wp-content\\\/uploads\\\/2021\\\/04\\\/eM.jpg\",\"contentUrl\":\"https:\\\/\\\/mcomputers.cz\\\/wp-content\\\/uploads\\\/2021\\\/04\\\/eM.jpg\",\"width\":351,\"height\":353,\"caption\":\"M Computers s.r.o.\"},\"image\":{\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/x.com\\\/m_computers_cz\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/mcomputers\",\"https:\\\/\\\/www.youtube.com\\\/channel\\\/UCqJt59UzfU6teJL1vVHvAbw\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/mcomputers.cz\\\/#\\\/schema\\\/person\\\/148aba8a2913606468a4d7043861a4f5\",\"name\":\"Mikol\u00e1\u0161 Pe\u0161ek\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/9489de9dba97edd15f902602b60c2c835f889e11e1fa1305b3b91af9b3d307d6?s=96&d=mm&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/9489de9dba97edd15f902602b60c2c835f889e11e1fa1305b3b91af9b3d307d6?s=96&d=mm&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/9489de9dba97edd15f902602b60c2c835f889e11e1fa1305b3b91af9b3d307d6?s=96&d=mm&r=g\",\"caption\":\"Mikol\u00e1\u0161 Pe\u0161ek\"},\"url\":\"https:\\\/\\\/mcomputers.cz\\\/en\\\/author\\\/mikolas-pesekmcomputers-cz\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"HPC Accelerator Comparison 2023 &#8211; M Computers s.r.o.","description":"Every year there are new accelerators for HPC and AI, ML and DL application development. That's why we compared their performance, parameters and new features.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/","og_locale":"en_US","og_type":"article","og_title":"HPC Accelerator Comparison 2023","og_description":"Every year there are new accelerators for HPC and AI, ML and DL application development. That's why we compared their performance, parameters and new features.","og_url":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/","og_site_name":"M Computers s.r.o.","article_published_time":"2023-06-13T13:46:36+00:00","article_modified_time":"2025-01-03T14:08:42+00:00","og_image":[{"width":620,"height":371,"url":"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_142623406.png","type":"image\/png"}],"author":"Mikol\u00e1\u0161 Pe\u0161ek","twitter_card":"summary_large_image","twitter_creator":"@m_computers_cz","twitter_site":"@m_computers_cz","twitter_misc":{"Written by":"Mikol\u00e1\u0161 Pe\u0161ek","Est. reading time":"98 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/#article","isPartOf":{"@id":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/"},"author":{"name":"Mikol\u00e1\u0161 Pe\u0161ek","@id":"https:\/\/mcomputers.cz\/#\/schema\/person\/148aba8a2913606468a4d7043861a4f5"},"headline":"HPC Accelerator Comparison 2023","datePublished":"2023-06-13T13:46:36+00:00","dateModified":"2025-01-03T14:08:42+00:00","mainEntityOfPage":{"@id":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/"},"wordCount":15338,"publisher":{"@id":"https:\/\/mcomputers.cz\/#organization"},"image":{"@id":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/#primaryimage"},"thumbnailUrl":"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_142623406.png","articleSection":["News"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/","url":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/","name":"HPC Accelerator Comparison 2023 &#8211; M Computers s.r.o.","isPartOf":{"@id":"https:\/\/mcomputers.cz\/#website"},"primaryImageOfPage":{"@id":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/#primaryimage"},"image":{"@id":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/#primaryimage"},"thumbnailUrl":"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_142623406.png","datePublished":"2023-06-13T13:46:36+00:00","dateModified":"2025-01-03T14:08:42+00:00","description":"Every year there are new accelerators for HPC and AI, ML and DL application development. That's why we compared their performance, parameters and new features.","breadcrumb":{"@id":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/#primaryimage","url":"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_142623406.png","contentUrl":"https:\/\/mcomputers.cz\/wp-content\/uploads\/2022\/12\/obrazek_2022-12-15_142623406.png","width":620,"height":371,"caption":"jzuk,u."},{"@type":"BreadcrumbList","@id":"https:\/\/mcomputers.cz\/en\/2023\/06\/13\/hpc-accelerator-comparison-2023\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Dom\u016f","item":"https:\/\/mcomputers.cz\/en\/"},{"@type":"ListItem","position":2,"name":"HPC Accelerator Comparison 2023"}]},{"@type":"WebSite","@id":"https:\/\/mcomputers.cz\/#website","url":"https:\/\/mcomputers.cz\/","name":"M Computers s.r.o.","description":"M Computers - V\u00e1\u0161 pr\u016fvodce sv\u011btem IT","publisher":{"@id":"https:\/\/mcomputers.cz\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/mcomputers.cz\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/mcomputers.cz\/#organization","name":"M Computers s.r.o.","url":"https:\/\/mcomputers.cz\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/mcomputers.cz\/#\/schema\/logo\/image\/","url":"https:\/\/mcomputers.cz\/wp-content\/uploads\/2021\/04\/eM.jpg","contentUrl":"https:\/\/mcomputers.cz\/wp-content\/uploads\/2021\/04\/eM.jpg","width":351,"height":353,"caption":"M Computers s.r.o."},"image":{"@id":"https:\/\/mcomputers.cz\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/m_computers_cz","https:\/\/www.linkedin.com\/company\/mcomputers","https:\/\/www.youtube.com\/channel\/UCqJt59UzfU6teJL1vVHvAbw"]},{"@type":"Person","@id":"https:\/\/mcomputers.cz\/#\/schema\/person\/148aba8a2913606468a4d7043861a4f5","name":"Mikol\u00e1\u0161 Pe\u0161ek","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/secure.gravatar.com\/avatar\/9489de9dba97edd15f902602b60c2c835f889e11e1fa1305b3b91af9b3d307d6?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/9489de9dba97edd15f902602b60c2c835f889e11e1fa1305b3b91af9b3d307d6?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/9489de9dba97edd15f902602b60c2c835f889e11e1fa1305b3b91af9b3d307d6?s=96&d=mm&r=g","caption":"Mikol\u00e1\u0161 Pe\u0161ek"},"url":"https:\/\/mcomputers.cz\/en\/author\/mikolas-pesekmcomputers-cz\/"}]}},"_links":{"self":[{"href":"https:\/\/mcomputers.cz\/en\/wp-json\/wp\/v2\/posts\/38117","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/mcomputers.cz\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mcomputers.cz\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mcomputers.cz\/en\/wp-json\/wp\/v2\/users\/12"}],"replies":[{"embeddable":true,"href":"https:\/\/mcomputers.cz\/en\/wp-json\/wp\/v2\/comments?post=38117"}],"version-history":[{"count":2,"href":"https:\/\/mcomputers.cz\/en\/wp-json\/wp\/v2\/posts\/38117\/revisions"}],"predecessor-version":[{"id":38119,"href":"https:\/\/mcomputers.cz\/en\/wp-json\/wp\/v2\/posts\/38117\/revisions\/38119"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/mcomputers.cz\/en\/wp-json\/wp\/v2\/media\/24843"}],"wp:attachment":[{"href":"https:\/\/mcomputers.cz\/en\/wp-json\/wp\/v2\/media?parent=38117"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mcomputers.cz\/en\/wp-json\/wp\/v2\/categories?post=38117"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mcomputers.cz\/en\/wp-json\/wp\/v2\/tags?post=38117"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}