NVIDIA ConnectX-8 SuperNIC User Manual
NVIDIA Docs Hub  NVIDIA Networking  Networking Adapters  NVIDIA ConnectX-8 SuperNIC User Manual  Monitoring

On This Page

Monitoring

Thermal Sensors

The adapter card incorporates the ConnectX IC, which operates in the range of temperatures between 0°C and 105°C.

Three thermal threshold definitions impact the overall system operation state and are designed to ensure timely application of server-level corrective actions, such as increased cooling, to keep the device temperature in the working range:

  • Warning – 97°C (default) When the device crosses this threshold, an Over-Temperature Warning message is issued by the management SW. This may be used to increase the cooling of the system (e.g. - by increased Fans RPM). Note that this temperature threshold does not require nor lead to any action by hardware (such as adapter card shutdown).

  • Critical – 102°C (default): When the device crosses this threshold, an Over-Temperature Critical message is issued by the management SW.

  • Fatal – 110°C: Upon reaching this temperature, an Over-temperature Fatal message is issued. If the the device continues to heat-up, the FW will automatically shut down upon reaching 115°C threshold.

The card's thermal sensors can be read through the system’s SMBus. The user can read these thermal sensors and adapt the system airflow following the readouts and the needs of the above-mentioned IC thermal requirements.

SuperNIC Heatsink

The heatsink is attached to the SuperNIC IC to dissipate the heat from the IC. It is attached by spring-loaded screws.

The SuperNIC IC has a thermal shutdown safety mechanism that automatically shuts down the card in cases of high-temperature events, improper thermal coupling, or heatsink removal.

For the required airflow (LFM) per OPN, please refer to the Specifications chapter.

Recently Updated

Show More

wait.gif

© Copyright 2025, NVIDIA. Last updated on Mar 20, 2025.
content here