{"id":87210,"date":"2025-08-23T18:00:50","date_gmt":"2025-08-24T01:00:50","guid":{"rendered":"https:\/\/phisonblog.com\/?p=87210"},"modified":"2025-08-25T17:27:13","modified_gmt":"2025-08-26T00:27:13","slug":"phison-ai-dominates-2025-fms-conference","status":"publish","type":"post","link":"https:\/\/phisonblog.com\/ja\/phison-ai-dominates-2025-fms-conference\/","title":{"rendered":"PHISON \u2013 2025\u5e74\u306eFMS\u30ab\u30f3\u30d5\u30a1\u30ec\u30f3\u30b9\u306fAI\u304c\u5e2d\u5dfb"},"content":{"rendered":"<p>[et_pb_section fb_built=&#8221;1&#8243; _builder_version=&#8221;4.16&#8243; global_colors_info=&#8221;{}&#8221;][et_pb_row _builder_version=&#8221;4.16&#8243; background_size=&#8221;initial&#8221; background_position=&#8221;top_left&#8221; background_repeat=&#8221;repeat&#8221; global_colors_info=&#8221;{}&#8221;][et_pb_column type=&#8221;4_4&#8243; _builder_version=&#8221;4.16&#8243; custom_padding=&#8221;|||&#8221; global_colors_info=&#8221;{}&#8221; custom_padding__hover=&#8221;|||&#8221;][et_pb_text _builder_version=&#8221;4.16&#8243; background_size=&#8221;initial&#8221; background_position=&#8221;top_left&#8221; background_repeat=&#8221;repeat&#8221; global_colors_info=&#8221;{}&#8221;]&#8221;Phison&#8217;s engineers reasoned that GPU performance scales with memory size, meaning high-bandwidth memory (HBM) DRAM, and proper management of this memory along with the use of cheaper NAND flash, can support large models with a lower number of GPUs than normal. Phison&#8217;s combined hardware and software approach to this problem,\u00a0<a href=\"https:\/\/www.phison.com\/images\/products_datasheet\/aiDAPTIV-Plus1-PageFlyer.pdf\" target=\"_blank\" rel=\"noopener\">dubbed<\/a> &#8220;aiDAPTIV+,&#8221; allows a single GPU to support LLMs at a somewhat slower performance than a multiple-GPU system at a significantly lower cost.&#8221;- techtarget.com<\/p>\n<p>Source: <a href=\"https:\/\/www.techtarget.com\/searchstorage\/opinion\/AI-dominates-2025-Future-of-Memory-and-Storage-conference\" target=\"_blank\" rel=\"noopener\">techtarget.com<\/a><br \/>\n[\/et_pb_text][\/et_pb_column][\/et_pb_row][\/et_pb_section]<\/p>\n","protected":false},"excerpt":{"rendered":"<p>&#8220;Phison&#8217;s engineers reasoned that GPU performance scales with memory size, meaning high-bandwidth memory (HBM) DRAM, and proper management of this memory along with the use of cheaper NAND flash, can support large models with a lower number of GPUs than normal. Phison&#8217;s combined hardware and software approach to this problem,\u00a0dubbed &#8220;aiDAPTIV+,&#8221; allows a single GPU [&hellip;]<\/p>\n","protected":false},"author":10,"featured_media":87211,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_et_pb_use_builder":"on","_et_pb_old_content":"<p>\"Phison's engineers reasoned that GPU performance scales with memory size, meaning high-bandwidth memory (HBM) DRAM, and proper management of this memory along with the use of cheaper NAND flash, can support large models with a lower number of GPUs than normal. Phison's combined hardware and software approach to this problem,\u00a0<a href=\"https:\/\/www.phison.com\/images\/products_datasheet\/aiDAPTIV-Plus1-PageFlyer.pdf\" target=\"_blank\" rel=\"noopener\">dubbed<\/a> \"aiDAPTIV+,\" allows a single GPU to support LLMs at a somewhat slower performance than a multiple-GPU system at a significantly lower cost.\"- techtarget.com<\/p><p>Source: <a href=\"https:\/\/www.techtarget.com\/searchstorage\/opinion\/AI-dominates-2025-Future-of-Memory-and-Storage-conference\">techtarget.com<\/a><\/p>","_et_gb_content_width":"","inline_featured_image":false,"footnotes":""},"categories":[114],"tags":[22],"class_list":["post-87210","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-phison-in-the-news","tag-long-content"],"acf":[],"_links":{"self":[{"href":"https:\/\/phisonblog.com\/ja\/wp-json\/wp\/v2\/posts\/87210","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/phisonblog.com\/ja\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/phisonblog.com\/ja\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/phisonblog.com\/ja\/wp-json\/wp\/v2\/users\/10"}],"replies":[{"embeddable":true,"href":"https:\/\/phisonblog.com\/ja\/wp-json\/wp\/v2\/comments?post=87210"}],"version-history":[{"count":3,"href":"https:\/\/phisonblog.com\/ja\/wp-json\/wp\/v2\/posts\/87210\/revisions"}],"predecessor-version":[{"id":87216,"href":"https:\/\/phisonblog.com\/ja\/wp-json\/wp\/v2\/posts\/87210\/revisions\/87216"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/phisonblog.com\/ja\/wp-json\/wp\/v2\/media\/87211"}],"wp:attachment":[{"href":"https:\/\/phisonblog.com\/ja\/wp-json\/wp\/v2\/media?parent=87210"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/phisonblog.com\/ja\/wp-json\/wp\/v2\/categories?post=87210"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/phisonblog.com\/ja\/wp-json\/wp\/v2\/tags?post=87210"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}