{"id":56352,"date":"2026-09-17T17:09:32","date_gmt":"2026-09-17T16:09:32","guid":{"rendered":"https:\/\/www.intelligentcio.com\/apac\/?p=56352"},"modified":"2026-09-17T17:09:33","modified_gmt":"2026-09-17T16:09:33","slug":"huawei-pioneers-new-computing-architecture-for-the-ai-era","status":"publish","type":"post","link":"https:\/\/www.intelligentcio.com\/apac\/2026\/09\/17\/huawei-pioneers-new-computing-architecture-for-the-ai-era\/","title":{"rendered":"Huawei pioneers new computing architecture for the AI era"},"content":{"rendered":"\n<p>Huawei has announced a new computing architecture for the AI era; the Peerium Computing Architecture. The development responds to the ever-growing demand for AI compute as it enables processors at the million scale to work as one computer.<\/p>\n\n\n\n<p>The architecture achieves strong scaling to the million-processor level through nested parallelism, unified memory addressing and peer interconnect. It breaks through the Turing paradigm with the introduction of Nested BSP (Nested Bulk Synchronous Parallel), extends the von Neumann single-machine architecture and overturns the master\u2013slave architecture that has prevailed for decades, so that a million processors truly become one larger computer.<\/p>\n\n\n\n<p>The key interconnect technology making the Peerium Computing Architecture possible is UnifiedBus (UB). Built on a single open protocol, UB is a high-speed bus that scales without limit to connect CPUs, NPUs, memory, SSDs, network interface cards (NICs) and switches. With UB, peer interconnect is achieved across compute, storage and networking.<\/p>\n\n\n\n<p>The Atlas 950 SuperPoD and SuperPoD-based SuperClusters are the first-generation product built on the Peerium Computing Architecture. An Atlas 950 SuperCluster with 256,000 cards is already being deployed and the Atlas 960 system based on near-packaged optics (NPO) is currently under testing.<\/p>\n\n\n\n<p>Eric Xu, Huawei\u2019s Rotating Chairman, outlined that in the AI era, Huawei is drawing on this new architecture to continuously build the SuperPoDs and SuperPod-based SuperClusters that meet customer needs for training and inference, making computing power available everywhere and intelligence accessible to all.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Huawei has announced a new computing architecture for the AI era; the Peerium Computing Architecture. The development responds to the ever-growing demand for AI compute as it enables processors at the million scale to work as one computer. The architecture achieves strong scaling to the million-processor level through nested parallelism, unified memory addressing and peer [&hellip;]<\/p>\n","protected":false},"author":21,"featured_media":56353,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[11969,12089,11958,11968,11960,8283,36,7349,42,44],"tags":[9786,110,12125,9787,12124,11150],"class_list":["post-56352","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai","category-data","category-hh-huawei-hub","category-huawei-hub","category-hh-top-story","category-innovation","category-networking","category-networks","category-industry-verticals-telecom","category-top-stories","tag-eric-xu","tag-huawei","tag-peerium-computing-architecture","tag-superpods","tag-ub","tag-unifiedbus"],"acf":[],"publishpress_future_workflow_manual_trigger":{"enabledWorkflows":[]},"_links":{"self":[{"href":"https:\/\/www.intelligentcio.com\/apac\/wp-json\/wp\/v2\/posts\/56352","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.intelligentcio.com\/apac\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.intelligentcio.com\/apac\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.intelligentcio.com\/apac\/wp-json\/wp\/v2\/users\/21"}],"replies":[{"embeddable":true,"href":"https:\/\/www.intelligentcio.com\/apac\/wp-json\/wp\/v2\/comments?post=56352"}],"version-history":[{"count":1,"href":"https:\/\/www.intelligentcio.com\/apac\/wp-json\/wp\/v2\/posts\/56352\/revisions"}],"predecessor-version":[{"id":56354,"href":"https:\/\/www.intelligentcio.com\/apac\/wp-json\/wp\/v2\/posts\/56352\/revisions\/56354"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.intelligentcio.com\/apac\/wp-json\/wp\/v2\/media\/56353"}],"wp:attachment":[{"href":"https:\/\/www.intelligentcio.com\/apac\/wp-json\/wp\/v2\/media?parent=56352"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.intelligentcio.com\/apac\/wp-json\/wp\/v2\/categories?post=56352"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.intelligentcio.com\/apac\/wp-json\/wp\/v2\/tags?post=56352"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}