[ad_1]
Ahead-looking: It is no secret that Nvidia has been the dominant GPU provider to information facilities, however now there’s a very actual risk that AMD may grow to be a severe contender on this market as demand grows. AMD was lately approached by a shopper asking to create an AI coaching cluster consisting of a staggering 1.2 million GPUs. That might probably make it 30x extra highly effective than Frontier, the present fastest supercomputer. AMD provided lower than 2% of information middle GPUs in 2023.
In an interview with The Next Platform, Forrest Norrod, AMD’s GM of Datacenter Options revealed that they’d acquired real inquiries from purchasers to construct AI coaching clusters utilizing 1.2 million GPUs. To place that into perspective, present AI coaching clusters are usually created utilizing just a few thousand GPUs related by way of high-speed interconnect throughout a number of native server racks.
The dimensions being thought-about for AI improvement now’s unprecedented. “Among the coaching clusters which might be being contemplated are actually mind-boggling,” Norrod mentioned. In reality, the biggest recognized supercomputer used to coach AI fashions is Frontier, which has 37,888 Radeon GPUs, making AMD’s potential supercomputer 30 occasions extra highly effective than Frontier.
In fact, it is not that easy. Even at present energy ranges, there is a plethora of pitfalls to think about when creating AI coaching clusters. AI coaching requires low latency to ship immediate outcomes, makes use of important quantities of energy, and {hardware} failures have to be considered – even with just some thousand GPUs.
Most servers run at round 20% utilization and deal with hundreds of small, asynchronous jobs in distant machines. Nonetheless, the rise of AI coaching is resulting in a major change in server construction. To maintain up with machine studying fashions and algorithms, an AI information middle have to be geared up with huge quantities of computing energy specifically designed for the job. AI coaching is actually one massive, synchronous job that requires each node within the cluster to move data forwards and backwards as quick as attainable.
What’s most fascinating is that these figures are coming from AMD, which accounted for less than 2% of the information middle GPU shipments in 2023. Nvidia, which made up the opposite 98%, has remained tight-lipped about what its purchasers have requested it to create. Because the market chief, we will solely think about what they’re engaged on.
Whereas the proposed 1.2 million GPU supercomputer may appear outlandish, Norrod mentioned that “very sober folks” are contemplating spending as much as 100 billion {dollars} on AI coaching clusters. This should not come as a shock, because the previous few years within the tech world have been outlined by the explosion in AI developments. Evidently firms are prepared to take a position important sums in AI and machine studying to stay aggressive.
[ad_2]
Source
