ReportWire

Tag: supercomputers

  • Trump’s Genesis Mission aims to build a centralized AI platform to power scientific breakthroughs

    [ad_1]

    President Donald Trump has issued a new Executive Order that launches the “Genesis Mission,” an AI-focused initiative that will be led by the Department of Energy. It will “harness the current AI and advanced computing revolution to double the productivity and impact of American science and engineering within a decade,” the DOE explained. One of the mission’s main goals is to build a centralized platform that will house a huge collection of datasets collected “over decades of federal investments,” as well as datasets from academic institutions and partners from the private sector.

    Those datasets will then be used to train scientific foundation models and to create AI agents, automate research workflows and accelerate scientific breakthroughs, the administration said in its announcement. “The platform will connect the world’s best supercomputers, AI systems, and next-generation quantum systems with the most advanced scientific instruments in the nation,” the Energy department said.

    Based on that statement, the platform will be linked to the two sovereign AI supercomputers the agency is building at the Oak Ridge National Laboratory, its famous research and development center. The machines, to be built by Hewlett Packard Enterprises, are meant to be the Trump AI Action Plan’s flagship supercomputers. The DOE previously revealed that the machines will be powered by AMD chips and will help tackle the biggest challenges in energy, medicine, health and national security.

    “The Genesis Mission marks a defining moment for the next era of American science. We are linking the nation’s most advanced facilities, data, and computing into one closed-loop system to create a scientific instrument for the ages, an engine for discovery that doubles R&D productivity and solves challenges once thought impossible,” said Dr. Darío Gil, the Under Secretary for Science and Genesis Mission Director.

    In the next four months, the Energy department must identify its initial set of data and model assets for the Genesis platform. The department must be able to demonstrate “an initial operating capability of the platform for at least one of the national science and technology challenges” the government has identified within nine months. While the list of challenges is pretty long, the Genesis Mission will focus on addressing three key challenges overall. First, it aims to accelerate nuclear and fusion energy, as well as to modernize the energy grid using AI. It also aims to power scientific discoveries for decades to come. Finally, it aims to create advanced AI technologies for the purpose of national security, such as systems that can ensure the reliability of America’s nuclear weapons and can accelerate the development of materials for defense.

    [ad_2]

    Mariella Moon

    Source link

  • Want to Buy a Decommissioned Supercomputer? Here’s Your Chance

    Want to Buy a Decommissioned Supercomputer? Here’s Your Chance

    [ad_1]

    On Tuesday, the US General Services Administration began an auction for the decommissioned Cheyenne supercomputer, located in Cheyenne, Wyoming. The 5.34-petaflop supercomputer ranked as the 20th most powerful in the world at the time of its installation in 2016. Bidding started at $2,500, but its price is currently $270,085.

    The supercomputer, which officially operated between January 12, 2017, and December 31, 2023, at the NCAR-Wyoming Supercomputing Center, was a powerful (and once considered energy-efficient) system that significantly advanced atmospheric and Earth system sciences research.

    “In its lifetime, Cheyenne delivered over 7 billion core-hours, served over 4,400 users, and supported nearly 1,300 NSF awards,” writes the University Corporation for Atmospheric Research (UCAR) on its official Cheyenne information page. “It played a key role in education, supporting more than 80 university courses and training events. Nearly 1,000 projects were awarded for early-career graduate students and postdocs. Perhaps most tellingly, Cheyenne-powered research generated over 4,500 peer-review publications, dissertations and theses, and other works.”

    UCAR says that Cheynne was originally slated to be replaced after five years, but the COVID-19 pandemic severely disrupted supply chains, and it clocked in two extra years in its tour of duty. The auction page says that Cheyenne recently experienced maintenance limitations due to faulty quick disconnects in its cooling system. As a result, approximately 1 percent of the compute nodes have failed, primarily due to ECC errors in the DIMMs. Given the expense and downtime associated with repairs, the decision was made to auction off the components.

    With a peak performance of 5,340 teraflops (4,788 Linpack teraflops), this SGI ICE XA system was capable of performing over 3 billion calculations per second for every watt of energy consumed, making it three times more energy-efficient than its predecessor, Yellowstone. The system featured 4,032 dual-socket nodes, each with two 18-core, 2.3-GHz Intel Xeon E5-2697v4 processors, for a total of 145,152 CPU cores. It also included 313 terabytes of memory and 40 petabytes of storage. The entire system in operation consumed about 1.7 megawatts of power.

    Just to compare, the world’s top-rated supercomputer at the moment—Frontier at Oak Ridge National Labs in Tennessee—features a theoretical peak performance of 1,679.82 petaflops, includes 8,699,904 CPU cores, and uses 22.7 megawatts of power.

    The GSA notes that potential buyers of Cheyenne should be aware that professional movers with appropriate equipment will be required to handle the heavy racks and components. The auction includes seven E-Cell pairs (14 total), each with a cooling distribution unit (CDU). Each E-Cell weighs approximately 1,500 lbs. Additionally, the auction features two air-cooled Cheyenne Management Racks, each weighing 2,500 lbs, that contain servers, switches, and power units.

    As of this writing, 23 potential buyers have bid on this computing monster so far. The auction closes on May 3 at 6:11 pm Central Time if you’re interested in bidding. But don’t get too excited by photos of the extensive cabling: As the auction site notes, “fiber optic and CAT5/6 cabling are excluded from the resale package.”

    This story originally appeared on Ars Technica.

    [ad_2]

    Benj Edwards, Ars Technica

    Source link

  • Tesla shares jump after Morgan Stanley predicts Dojo supercomputer could add $500 billion in market value | CNN Business

    Tesla shares jump after Morgan Stanley predicts Dojo supercomputer could add $500 billion in market value | CNN Business

    [ad_1]


    New York
    CNN
     — 

    Tesla’s Dojo supercomputer could fuel a $500 billion jump in the electric vehicle maker’s market value, analysts at Morgan Stanley said in a note Monday.

    Shares of Tesla jumped more than 6% during early trading Monday morning, on the heels of the rosy prediction from Morgan Stanley’s team about the automaker’s supercomputing efforts. The Morgan Stanley team, lead by longtime Tesla analyst Adam Jonas, predicted that the massive drive in value could come from Dojo potentially unlocking new revenue streams through the wider adoption of robotaxis and software services.

    The analysts compared the potential of Dojo at Tesla to the “same forces that have driven” Amazon Web Services to propel Amazon’s profitability to new heights.

    “Investors have long debated whether Tesla is an auto company or a tech company. We believe it’s both, but see the biggest value driver from here being software and services revenue,” the note stated.

    Dojo, an in-house supercomputer that has been in the works at Tesla for some five years, is designed to train AI systems to complete complex tasks like assisting Tesla’s driver-assistance system Autopilot as well as help propel its “Full Self-Driving” efforts.

    The Morgan Stanley analysts see Dojo as being able to open up “new addressable markets that extend well beyond selling vehicles at a fixed price.”

    The analysts added that the latest version of Tesla’s full self-driving system (expected to be unveiled at the end of the year) and Tesla’s next AI day (expected in early 2024, but yet to be announced) will be “worth watching.”

    Shares of Tesla have doubled since the beginning of the year, but are still far off from the all-time intraday high of $414.50 hit in November 2021. The world’s most valuable carmaker had a market cap of some $788.74 billion as of the market close on Friday.

    [ad_2]

    Source link

  • Tesla’s AI Day is tonight. It may wow you — or end with a gaffe | CNN Business

    Tesla’s AI Day is tonight. It may wow you — or end with a gaffe | CNN Business

    [ad_1]


    Washington, DC
    CNN Business
     — 

    Tesla

    (TSLA)     will hold its second annual AI Day in Palo Alto, California, Friday evening. The six-hour event will include updates on Tesla

    (TSLA)    ’s work in artificial intelligence, “Full Self-Driving,” its supercomputer “Dojo” and maybe a humanoid robot, according to invitations posted online by Tesla

    (TSLA)     supporters. The event is expected to be live-streamed.

    Dojo is a supercomputer being designed to train AI systems to complete complex tasks like Tesla’s driver-assistance systems Autopilot and “Full Self-Driving,” which sometimes perform some driving tasks like steering and keeping up with traffic. Tesla’s previous AI Day included detailed technical explanations of the company’s work in a bid to attract leading engineers.

    Tesla CEO Elon Musk has claimed before that in the long run people will think of Tesla as an AI company, rather than a car company or energy company. He has said that Tesla AI may play a role in computers matching general human abilities, a huge milestone many experts say is decades away and perhaps unattainable. Musk, who has a long history of predictions, has said it may be reached in 2029.

    But more limited and easier to develop forms of artificial intelligence — like identifying emergency vehicles stopped on a highway — have proven to be a significant hurdle for the company as it pursues its dreams of self-driving cars. AI powers “Full Self-Driving,” but the system has faced criticism and backlash as it still requires driver intervention to prevent collisions and Musk’s deadlines for its capabilities slip year after year.

    And this summer Tesla’s director of artificial intelligence, Andrej Karpathy, exited the company, several months after it was announced he was taking a sabbatical.

    It’s not easy to predict what may or may not show up at any event helmed by Musk. Products heralded and talked about sometimes don’t perform as designed — like when Musk showed off the Tesla Cybertruck’s supposedly “unbreakable” windows, that promptly broke — and can’t even be bought years later. (Three years after the event Tesla sells a T-shirt that memorializes the broken window, but it has yet to sell a Cybertruck.)

    Musk has unquestionably disrupted entire industries with his work at Tesla and SpaceX. But he’s also earned a reputation for missing deadlines and overpromising.

    Last year’s AI Day “surprise,” for instance, was a Tesla “robot,” which was just a human dancing in a suit.

    Musk then claimed that the automaker is building a 5-foot-8, 125-pound humanoid robot, called Optimus or Tesla Bot and a prototype would likely be unveiled this year. It’s unclear if a prototype will be revealed Friday, but Musk tweeted Thursday that the event would include “cool hardware demos.”

    Tesla is also working on wheeled robots for manufacturing and autonomous logistics, according to a Tesla job posting for a senior humanoid mechatronics robotics architect.

    Musk claimed last year that the humanoid robot would have a profound impact on the economy. It would begin by working on boring, repetitive and dangerous tasks, he said.

    Tesla and Musk are not, of course, the first to bet on robots. Robots already handle many factory jobs, and companies like Boston Dynamics have worked for years to develop humanoid, animal-like, and other robots for industrial applications.

    Humanoid robots have long fascinated the public and earned a place in pop culture as powerful but sometimes dangerous. Tesla tapped into this when it posted on Instagram in a promotion for the event that, “if you can run faster than 5 mph, you’ll be fine.” The Tesla humanoid robot is planned to have a top speed of 5 mph, the automaker has said.

    But creating a humanoid robot that rivals a human’s abilities has proved incredibly difficult for robotics experts. Artificial intelligence has seen major advances yet trails the general abilities of a human toddler. Most robots in use today are restricted to simple tasks in basic environments like vacuuming a home or moving parts in a factory.

    Tesla would not be the first automaker to build a humanoid robot, either. Honda worked on a series of robots, known as Asimo, for nearly 20 years. The Japanese company shut down development of Asimo in 2018. Korean automaker Hyundai bought Boston Dynamics in 2020.

    Musk said Thursday that AI Day would be “highly technical” as it is meant for recruiting engineers to work on artificial intelligence, robotics and computer chips.

    “Engineers who understand what problems need to be solved will like what they see,” Musk tweeted Friday.

    Tesla did not respond to a request for comment.

    [ad_2]

    Source link