Understand this story
xMEMS has begun mass production of the XMC-2400 and XMC-4800 solid-state micro-cooling chips. According to Data Center Knowledge, these two chips are based on a silicon-based piezoelectric MEMS platform that generates directed airflow, used to dissipate heat from hot spots in subsystems such as SSDs, memory, and optical modules that are difficult for liquid-cooling cold plates and system fans to reach. The XMC-2400 provides up to 28 cubic centimeters per second of airflow and 1300 Pa of back pressure, with power consumption of about 150 mW; the XMC-4800, aimed at enterprise-grade SSDs, delivers 48 cubic centimeters per second, 1100 Pa, and about 240 mW. xMEMS positions them as a supplement to, rather than a replacement for, the overall thermal architecture, and expects them to be deployed pre-installed inside subsystems such as SSDs.
AI-generated from public reports
Latest development · Report
The Role of Solid-State Micro-Cooling in Data Centers: xMEMS Chips Target SSD Local Hot SpotsDevelopments
1 confirmed developments
xMEMS Mass-Produces XMC-2400 and XMC-4800 Solid-State Micro-Cooling Chips
Source:Data Center Knowledge
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Oct 28
Data Center Knowledge
The Role of Solid-State Micro-Cooling in Data Centers: xMEMS Chips Target SSD Local Hot Spots
xMEMS has mass-produced the XMC-2400 and XMC-4800 solid-state micro-cooling chips, which use a silicon-based piezoelectric MEMS platform to generate directed airflow, cooling hot spots in subsystems such as SSDs, memory, and optical modules that liquid-cooling cold plates and system fans struggle to cover. The XMC-2400 provides up to 28 cubic centimeters per second of airflow and 1300 Pa of back pressure, with power consumption of about 150 mW; the XMC-4800, aimed at enterprise-grade SSDs, delivers 48 cubic centimeters per second, 1100 Pa, and about 240 mW. xMEMS positions it as a supplement to, rather than a replacement for, the overall cooling architecture, and expects it to be deployed pre-installed inside subsystems such as SSDs.