From nuclear reactors to AI<\/strong><\/p>\nAzizian was a postdoc at MIT in 2013 when he met Bucci, who was then a research scientist. With funding from the MIT Energy Initiative, they worked on heat transfer in nuclear reactors before Azizian went into industry, where he shifted his focus to cooling chips. Azizian first worked on Microsoft\u2019s HoloLens augmented reality headset and then joined Nvidia, which produces the graphical processing units companies use to train and run the latest AI models. Meanwhile, Bucci continued conducting research at MIT, becoming an assistant professor in 2016.<\/p>\n
Azizian walked into his first data center in 2017, where he was struck by the massive, noisy fans that filled the building as they cooled.<\/p>\n
\u201cI thought, \u2018Holy crap, this is not how you cool facilities,\u2019\u201d Azizian recalls, noting air cooling can still take up 40 percent of the power going into a data center. \u201cIt was not an efficient way of doing things, but since it wasn\u2019t hurting the performance, no one cared that the cooling technology was 50 years old.\u201d<\/p>\n
Azizian began talking with Bucci about applying their knowledge around optimizing heat transfer in nuclear reactors to data centers. Scientists have spent decades finding better ways to move heat in nuclear reactors.<\/p>\n
\u201cHeat transfer determines how much energy you can extract from the reactor core, which translates directly to revenue,\u201d Azizian explains.<\/p>\n
The founders started Ferveret in 2021. A lot has changed since Azizian walked into his first data center. Chip companies have packed more and more components onto their chips as the explosion in artificial intelligence has put a premium on squeezing as much computing capacity as possible out of limited power supplies.<\/p>\n
That has driven data center operators to use liquid to cool chips \u2014 often through a technique known as immersion cooling that submerges chips in liquid. The most effective form of immersion cooling brings the liquid to a boil.<\/p>\n
\u201cLiquid is a better heat transfer medium than air. That\u2019s why when you stick your hand into room temperature water it still feels cold,\u201d Bucci explains. \u201cWhen liquid is boiling, it becomes even better at removing heat because the phase change requires a lot of energy, which is the energy you remove from the chip. That lets you transfer large quantities of heat with minimal temperature differences between the chips and the liquid.\u201d<\/p>\n
Unfortunately, boiling liquid adds complexity to the system because it forces operators to capture and reliquefy the bubbles while controlling for pressure, temperature, and fluid inventory.<\/p>\n
Ferveret\u2019s system is adapted from a process in nuclear reactors called subcooled boiling. It uses a liquid with a low boiling point and none of the toxic PFAS \u201cforever chemicals\u201d that other approaches rely on. At the surface of the chip, Ferveret\u2019s liquid produces smaller bubbles than other immersion cooling approaches. Those bubbles detach more frequently and quickly recondense in the surrounding liquid, accelerating the bubble-rewetting cycle at the surface of the chip to hasten heat transfer.<\/p>\n
Ferveret delivers its APC system in small boxes, each of which houses one server. The founders say their modular systems make it easier to deploy the system and simplify maintenance.<\/p>\n
\u201cThe physics enable us to get to form factors that weren\u2019t possible in the past,\u201d Azizian says. \u201cMost immersion cooling solutions are large tanks that people submerge the servers in. We have a smaller, modular rack-mounted solution that makes it adaptable to the current infrastructure, so it\u2019s easier for people to deploy our technology.\u201d<\/p>\n
Ferveret also offers control software that adjusts the power going to each server in real-time to further improve efficiency.<\/p>\n
\u201cWe deliver full-stack systems that include the cooling box, the rack, the cooling distribution units, and sensors that measure the temperature and pressure,\u201d Bucci says. \u201cOur software monitors those sensors and optimizes the operating condition inside each box to ensure that energy consumption is minimized in the system.\u201d<\/p>\n
AI with fewer resources<\/strong><\/p>\nIn addition to helping data centers to run more efficiently, Ferveret is also improving sustainability by making it easier to operate data centers in remote regions with more renewable energy.<\/p>\n
\u201cThe sun shines in places where you don\u2019t have much water, so the advantage of us being water-free is we allow you to build data centers where you have solar energy but nothing to cool the data center down,\u201d Bucci says. \u201cThis technology can help deploy data centers in regions where normally you wouldn\u2019t have the resources to do so, including Africa, the Middle East, and of course parts of America. It\u2019s a huge unlock.\u201d<\/p>\n
Ferveret is in talks with the large cloud computing companies known as hyperscalers, and is currently part of Nvidia\u2019s Inception program for startups. The company plans to announce expanded partnerships later this year. From there, the founders plan to quickly scale their technology to help the AI industry continue to grow without further straining the planet.<\/p>\n
\u201cThe computing industry is facing a huge challenge in the form of access to power, and they have a problem with access to water in many regions,\u201d Azizian says. \u201cThat will only become more limiting as the industry grows. The main goal for these data center operators would be to get more tokens from the power they have. We\u2019ve shown we can do that.\u201d<\/p>\n
Fuente:<\/strong> https:\/\/news.mit.edu<\/em><\/a><\/p>\n","protected":false},"excerpt":{"rendered":"El auge de la inteligencia artificial se ve impulsado por una enorme expansi\u00f3n de los centros de datos. Se\u00a0prev\u00e9\u00a0que, para finales de la d\u00e9cada, estos… <\/p>\n","protected":false},"author":1,"featured_media":18859,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[2,23],"tags":[],"_links":{"self":[{"href":"https:\/\/www.fie.undef.edu.ar\/ceptm\/index.php?rest_route=\/wp\/v2\/posts\/18858"}],"collection":[{"href":"https:\/\/www.fie.undef.edu.ar\/ceptm\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.fie.undef.edu.ar\/ceptm\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.fie.undef.edu.ar\/ceptm\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.fie.undef.edu.ar\/ceptm\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=18858"}],"version-history":[{"count":1,"href":"https:\/\/www.fie.undef.edu.ar\/ceptm\/index.php?rest_route=\/wp\/v2\/posts\/18858\/revisions"}],"predecessor-version":[{"id":18860,"href":"https:\/\/www.fie.undef.edu.ar\/ceptm\/index.php?rest_route=\/wp\/v2\/posts\/18858\/revisions\/18860"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.fie.undef.edu.ar\/ceptm\/index.php?rest_route=\/wp\/v2\/media\/18859"}],"wp:attachment":[{"href":"https:\/\/www.fie.undef.edu.ar\/ceptm\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=18858"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.fie.undef.edu.ar\/ceptm\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=18858"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.fie.undef.edu.ar\/ceptm\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=18858"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}