However, it isn\u2019t just about raw speed; it\u2019s about control. By designing its own silicon and pairing it seamlessly with its AI software, Google continues to cut costs while significantly boosting efficiency.<\/p>\n
In a market that\u2019s essentially defined by scale and silicon, Ironwood is Google\u2019s newest weapon, and arguably its sharpest yet.<\/p>\n
Ironwood gives Google a new edge in the cloud chip fight.Image source: Radecka/NurPhoto via Getty Images <\/p>\n
Google\u2019s Ironwood TPU represents a full-on escalation in the powerful cloud arms race.<\/p>\n
The latest generation of Google\u2019s custom AI silicon delivers an incredible power leap, nearly 10 times the throughput of TPU v5p and four times the per-chip performance of last year\u2019s version.<\/p>\n
To put it plainly, it\u2019s the fastest, smartest, and most efficient processor the tech giant has ever developed.<\/p>\n
Related: Apple could reboot Siri to finally please users<\/a><\/p>\n However, speed isn\u2019t the only story.<\/p>\n Ironwood is designed with scale in mind, as thousands of chips could potentially link together into massive \u201csuperpods,\u201d pushing data across a 9.6-terabit-per-second network while sharing nearly 1.8 petabytes of memory.<\/p>\n Additionally, liquid cooling and optical networking effectively keep the setup from overheating. Early users, including the likes of Anthropic, say Ironwood lets them do a lot more, faster, and for less.<\/p>\n 10\u00d7 the throughput of Google\u2019s previous TPU generation<\/p>\n 9,216-chip \u201csuperpod\u201d network with 1.77 PB shared memory<\/p>\n Advanced liquid cooling, along with optical switching for reliability<\/p>\n Co-designed silicon and software for peak efficiency at lower costs<\/p>\n Google is hogging all the spotlight with Ironwood, but it\u2019s far from the only Big Tech play that\u2019s betting big on custom silicon.<\/p>\n Across the cloud space, rivals continue developing their own potent chips, and in some cases, entire hardware ecosystems to gain speed, cut costs, and reduce dependency on Nvidia\u2019s pricey GPUs.<\/p>\n Related: Nvidia CEO drops unsettling prediction on AI race<\/a><\/p>\n So it\u2019s a lot less about bragging rights and more about economics.<\/p>\n Though each company\u2019s approach feels a little different, the goal is the same: to dominate the entire stack, from hardware to hyperscale.<\/p>\n Amazon Web Services (AWS): Amazon\u2019s popular cloud service utilizes Trainium (for training) and Inferentia (for inference) chips, offering major cost savings. AWS claims a nearly 50% cheaper model training<\/a> than comparable GPU setups.<\/p>\n Microsoft Azure: Under Project Athena, Microsoft\u2019s Maia 100 accelerator is the tech giant\u2019s ticket to silicon independence. It layers in-house hardware along with the potent Azure AI stack, cutting costs and reducing reliance on Nvidia.<\/p>\n Meta: Not to be left behind, Meta has its own MTIA inference chip in production and is already testing out a training chip in powering its recommendation engines.<\/p>\n Apple: The quiet pioneer. Its A-series and M-series chips show the payoff of vertical integration in efficiency, performance per watt, along with total control over its ecosystem.<\/p>\n Google\u2019s raking in the moolah from its big bets on building custom chips, where it matters most.<\/p>\n Google Cloud sales are up an impressive 34% year over year last quarter<\/a>, while operating margins have climbed 23.7%, up immensely from 17% a year earlier.<\/p>\n CEO Sundar Pichai credited the success to \u201csubstantial demand for our AI infrastructure products, including TPU-based solutions\u201d as a key growth driver.<\/p>\n Also, the company isn\u2019t shy about spending to keep that momentum alive, with capital expenditures or capex soaring to $91 billion for 2025, much of it being earmarked for AI-powered data centers.<\/p>\n More Tech Stocks:<\/p>\n That investment is already paying off in high-value wins.<\/p>\n Pichai says Google has signed over $1 billion in cloud contracts this year, more than in the previous two years combined, spearheaded by major partners such as Anthropic and Meta.<\/p>\n Google Cloud\u2019s backlog hit $155 billion in Q3<\/a>, up a superb 46% quarter-over-quarter, reflecting those eye-catching commitments.<\/p>\n Even Apple is negotiating a huge billion-dollar-a-year deal to use Google\u2019s Gemini AI model to supercharge Siri.<\/p>\n So in many ways, Pichai\u2019s pitch is simple: Google is \u201cthe only hyperscaler\u201d offering a true full-stack AI platform, covering everything from custom chips like Ironwood to frontier models like Gemini.<\/p>\n Related: Jim Cramer resets AI stock \u2018buy\u2019 list for rest of 2025<\/a><\/p>\n This story was originally reported by TheStreet<\/a> on Nov 7, 2025, where it first appeared in the Technology<\/a> section. Add TheStreet as a Preferred Source by clicking here.<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"Google isn’t just joining the coveted AI chip race; it\u2019s turning up the heat. The Alphabet (GOOGL)-owned tech…\n","protected":false},"author":2,"featured_media":127838,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[220,218,219,70236,1711,8815,61,60,74981,80],"class_list":{"0":"post-127837","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-artificialintelligence","11":"tag-custom-silicon","12":"tag-google","13":"tag-google-cloud","14":"tag-ie","15":"tag-ireland","16":"tag-ironwood","17":"tag-technology"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/127837","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/comments?post=127837"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/127837\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media\/127838"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media?parent=127837"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/categories?post=127837"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/tags?post=127837"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}