Applied Digital Corporation (Nasdaq: APLD) (“Applied Digital” or the “Company”), a company that designs, builds and operates next-generation digital infrastructure designed for high-performance computing (“HPC”) applications, today announced the Launch of an artificial intelligence (AI) cloud service that provides high-performance computing power for AI applications, including training sophisticated language models, graphics rendering, and more. AI Cloud services are offered through Sai Computing, a wholly owned subsidiary of Applied Digital, which acts as the brand that sells the services.
The service will initially be delivered from our new 9MW HPC Jamestown facility, which currently hosts clusters of Nvidia A100, A40 and A6000 GPUs. We are committed to our customers to make the most of the initial capacity that will come online this week.
“According to a recent industry report1, the global artificial intelligence market is expected to grow at a CAGR of around 23% over the forecast period of 2023-2028, reaching a size of around US$6 trillion by 2026. .” Applied Digital CEO and Chairman Wes Cummins said:
“Applied Digital’s differentiated data center infrastructure is uniquely positioned to handle the advanced and demanding workloads that enterprises need to take advantage of AI and other HPC applications.” Cummins added. “The launch of our AI cloud service will allow us to expand our offerings and further capitalize on the unprecedented demand we are seeing from our customers for our services. We aim to be at the forefront of delivering the next generation of essential services that will make it possible.”
Construction of the dedicated HPC facility began late last year at Applied Digital’s dedicated 9MW stand-alone processing center adjacent to Applied Digital’s existing 100MW Jamestown facility. The company expects initial capacity at its new HPC facility to go live this week, with full production later this year.
Applied Digital’s next-generation data centers are ideal hosting sites for HPC applications, typically costly and traditional data centers where most of today’s facilities are not capable of providing the power density needed for AI. can provide a low-cost, high-compute solution compared to /ML workload.
