Liquid immersion cooling platform and components thereof
An immersion cooling system and methods for operating the system are described. The system can comprise a vessel configured to hold thermally conductive, condensable dielectric fluid; a pressure controller to reduce or increase an interior pressure of the vessel; a computer component configured to be at least partially submerged within the dielectric fluid; and a fluid circulation system configured to draw the dielectric fluid from a sump area of the vessel, pass the dielectric fluid through a filter and deliver the dielectric fluid to a bath area of the vessel.
1. A system comprising:
a vessel configured to hold thermally conductive, condensable dielectric fluid;
a computer component configured to be at least partially submerged within the dielectric fluid;
a chassis configured to hold the computer component and the dielectric fluid; and
a fluid circulation system configured to draw the dielectric fluid from a sump area of the vessel, pass the dielectric fluid through a filter and deliver the dielectric fluid to the chassis wherein the sump area comprises a side wall adjacent the chassis and a side portion of the vessel.
2. The system of claim 1 , further comprising a fluid level sensor for determining a level of the dielectric fluid.
3. The system of claim 2 , wherein the fluid level sensor is located in the chassis.
4. The system of claim 2 , further comprising an inlet for receiving the dielectric fluid from a source outside of the vessel.
5. The system of claim 1 , further comprising a plurality of sensors and a management system configured to:
receive sensor data relating to the computer component; and
determine a failure of the filter based on the sensor data.
6. The system of claim 5 , wherein the sensor data includes: a temperature of the computer component, a power consumption at the vessel, an outside temperature, a dielectric fluid temperature, a temperature of incoming cooling medium, a temperature of outgoing cooling medium, a flow rate of a cooling medium, a temperature of an area above the bath area, a number of computer components present in the vessel, or a location of each computer component within the vessel.
7. The system of claim 5 , wherein the management system is configured to determine whether the computer component is overheating using a machine learning model.
8. The system of claim 7 , wherein the machine learning model is trained using the sensor data received from the vessel.
9. The system of claim 5 , wherein the management system is further configured to receive temperature data.
10. The system of claim 9 , wherein the management system is further configured to determine a state of operation of the vessel.
11. The system of claim 10 , wherein, the state of operation is: 1) activation or deactivation of a heat exchanger; 2) dielectric fluid burning; and 3) dielectric fluid leaking.
12. The system of claim 10 , wherein the management system is further configured to determine the state of operation of vessel using a machine learning model based on data received from the sensor and the temperature data.
13. The system of claim 1 , wherein the vessel is protected by a secondary layer.
14. The system of claim 13 , wherein the secondary layer is parallel to an internal layer and configured to receive leaked fluid from the internal layer.
15. The system of claim 14 , wherein a fluid sensor is provided between the secondary layer and the internal layer to detect a level of the leaked fluids.