"The memory chip supply crisis is familiar to Reg readers: manufacturers shifted capacity toward high-margin server DRAM and HBM for AI infrastructure, starving the market of standard DRAM and NAND and driving prices up."
"AI servers are typically stuffed with high performance hardware including GPU accelerators and so consume substantially more power than general-purpose systems. They are prioritized by PMIC suppliers, as they need higher value, high-current-density products."
"Samsung plans to shut down its 8-inch wafer fabrication plant in Korea, further squeezing PMIC capacity for general servers, with lead times expected to run between 35 to 40 weeks."
The chip shortage is now affecting power management and controller silicon, leading to a downgrade in server shipment growth forecasts. TrendForce reduced the 2026 growth estimate from 20% to 13% due to extended lead times for server components. The shift in manufacturing capacity towards high-margin AI products has resulted in a scarcity of standard DRAM and NAND. Lead times for power management chips and Baseboard Management Controllers have also increased significantly, with demand for general-purpose servers remaining steady despite these challenges.
Read at Theregister
Unable to calculate read time
Collection
[
|
...
]