Nvidia will soon begin full-scale shipments of Blackwell-based hardware for artificial intelligence systems. At the SIGGRAPH conference in Denver, USA on July 29, the company presented several updates to its software solutions and announced that it had begun shipping samples of Blackwell-based AI hardware, reports Commercial Times.
Image source: Nvidia
Nvidia CEO Jensen Huang believes that the Blackwell series of accelerators will be the most successful product in the company’s history, with cloud service providers expected to start building new data centers with AI servers. These chips will also impact other segments of the tech industry: demand for TSMC’s 4nm process technology will increase; water cooling technologies will begin to spread rapidly, with their penetration reaching 10% and benefiting companies such as Asia Vital Components, Auras Technology, Delta Electronics and Cool IT.
The new AI chips are expected to ship to customers in Q4, with full-scale production planned for 2025. This will benefit assembly plants Wistron and Ingrasys (part of Foxconn), which manufacture wafers, compute boards, and communications boards. Orders for rack-mount systems from Wiwynn, Quanta, Gigabyte, Asus, and ASRock will increase, with Quanta, Wiwynn, and Inventec already announcing related products with shipments in Q4 and further increases in volumes in the first half of next year.
In 2025, 60,000 Nvidia GB200 NVL36 rackmount systems will ship, and Blackwell GPU penetration will be between 2.1 million and 2.2 million units, making Blackwell the primary platform present in 80% of Nvidia’s high-end systems, analysts predict. TrendForce. The GB200 NVL36 systems will use a combination of air and liquid cooling solutions, while the NVL72 will be primarily liquid-cooled.
If you notice an error, select it with your mouse and press CTRL+ENTER.
