After a double-digit on-year decline in 2023, global server shipments are expected to resume growth in 2024, but the volumes will fall back to slightly above the 2019 level.
Based on DIGITIMES Research's statistics, fourth-quarter 2023 global server shipments grew 3.1% from the prior quarter, coming somewhat short of expectation.
The three primary high performance computing (HPC) chip suppliers, Intel, AMD, and Nvidia, all experienced lackluster sales in the first half of 2023 amid weak demand for HPC applications, according to a survey of DIGITIMES Research.
In response to explosive generative AI and large language model (LLM) demand, major cloud service providers and leading server brands are stepping up efforts toward AI servers, with a focus on ramping up their procurement of high-end AI servers featuring accelerators with high bandwidth memory (HBM) integrated.
Global server shipments slipped 5.7% sequentially in the second quarter and are likely to see only single-digit on-quarter growth in the third quarter due to brand vendors and CSPs both turning conservative toward the second half of the year, even though the new mainstream server CPU platforms are ready for supply.
Demand for generative AI (Gen AI) and large language model (LLM) is rising rapidly, driven by the emergence of ChatGPT, a chatbot developed by OpenAI. For their large scale as well as the massive data sets and resources required to train LLMs, cloud service providers (CSP) are generally adopting the method of combining inference and prompt engineering for their AI solutions to support clients' customization needs.
Computex 2023 was hosted at Taipei Nangang Exhibition Center from May 30 through June 2 and DIGITIMES Research paid a visit to the show to survey and outline the trends that new AI servers showcased at the event feature including the increasing popularity of models powered by Nvidia's HGX H100 among the suppliers; AI servers with PCIe support trending toward high-density and high-application-flexibility development; servers powered by Nvidia's GH200 boasting the best performance running Nvidia's Large Language Model (LLM) and Nvidia's release of MGX reference server designs.
Global server shipments performed weaker than DIGITIMES Research had anticipated, slipping 13.8% sequentially to fall below four million units. Having a lower comparison base in the same quarter a year ago and the fact that brand vendors are expected to begin volume shipments of servers with new CPU platforms, overall server shipments are expected to witness a low single-digit sequential growth in the second quarter of 2023.