Controlling networking device temperatures by reducing data transfer rates from associated nodes
View Patent ↗A method includes a compute node transmitting data to a port of a first switch at a first data transfer rate, monitoring the temperature of the port, and a management node providing an instruction to the compute node in response to the port temperature exceeding a temperature limit, wherein the instruction instructs the compute node to reduce the first data transfer rate to the port. The method further includes the compute node reducing the data transfer rate to the port in response to receiving the instruction. The method is applicable to multiple compute nodes transmitting data to multiple ports of a first switch. The data transfer rate may be reduced by throttling the compute node, renegotiating a link speed between the compute node and the port, or redirecting data to another switch. The methods facilitate thermal control of a switch without its own thermal throttling capability.
1. A method, comprising:
a compute node transmitting data to a port of a first switch at a first data transfer rate;
monitoring the temperature of the port;
a management node providing an instruction to the compute node in response to the port temperature exceeding a temperature limit, wherein the instruction instructs the compute node to reduce the first data transfer rate to the port; and
the compute node reducing the data transfer rate to the port in response to receiving the instruction, wherein the compute node reduces the data transfer rate to the port by redirecting data transmission away from the first switch to a second switch.
2. The method of claim 1 , further comprising:
determining, prior to reducing the data transfer rate to the port, that fans within the first switch are operating at a maximum fan speed setpoint.
3. The method of claim 1 , wherein the port temperature is the temperature of a transceiver case for the port.
4. The method of claim 1 , wherein the compute node reducing the data transfer rate to the port further includes renegotiating a link speed between the compute node and the port of the first switch, wherein the link speed limits the data transfer rate between the compute node and the port, and wherein the renegotiated link speed is less than a previous link speed that was used while the temperature of the port exceeded the temperature limit.
5. The method of claim 4 , wherein renegotiating the link speed between the compute node and the port of the first switch includes the compute node advertising a maximum supported data transfer rate that is reduced from a previously advertised maximum supported data transfer rate, then forcing the link down, and then bringing the link back up.
6. The method of claim 1 , wherein redirecting the data transmission away from the first switch to the second switch includes the compute node redirecting the data from a first network interface controller coupled with the first switch to a second network interface controller coupled with the second switch.
7. The method of claim 6 , further comprising:
determining, prior to redirecting the data transmission to the second network interface controller coupled with the second switch, that the second switch has a temperature that does not exceed a temperature limit of the second switch.
8. A method, comprising:
a plurality of compute nodes transmitting data to a plurality of ports of a first switch;
monitoring a temperature of each of the ports;
a management node providing an instruction to a first compute node in response to a temperature of a first port exceeding a temperature limit, wherein the first compute node is transmitting data to the first port and the instruction instructs the first compute node to reduce a data transfer rate to the first port; and
the first compute node reducing the data transfer rate to the first port in response to receiving the instruction, wherein the first compute node reduces the data transfer rate to the first port by redirecting data transmission away from the first switch to a second switch.
9. The method of claim 8 , wherein the first compute node reducing the data transfer rate to the first port further includes renegotiating a link speed between the first compute node and the first port of the first switch, wherein the link speed limits the data transfer rate between the first compute node and the first port, and wherein the renegotiated link speed is less than a previous link speed that was used while the temperature of the first port exceeded the temperature limit.
10. The method of claim 9 , wherein renegotiating the link speed between the first compute node and the first port of the first switch includes the first compute node advertising a maximum supported data transfer rate that is reduced from a previously advertised supported data transfer rate, then forcing the link down, and then bringing the link back up.
11. The method of claim 8 , wherein redirecting data transmission away from the first switch to the second switch includes the compute node redirecting data transmission from a first network interface controller coupled with the first switch to a second network interface controller coupled with the second switch.
12. The method of claim 8 , wherein the first and second switches are included in a multi-node chassis along with the plurality of compute nodes.
13. A computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by one or more processors to:
transmit data from a compute node to a port of a first switch at a first data transfer rate;
monitor the temperature of the port;
provide an instruction to the compute node in response to the port temperature exceeding a temperature limit, wherein the instruction instructs the compute node to reduce the first data transfer rate to the port; and
reduce the data transfer rate from the computer node to the port in response to receiving the instruction, wherein the compute node reduces the data transfer rate to the port by redirecting data transmission away from the first switch to a second switch.
14. The computer program product of claim 13 , wherein redirecting the data transmission away from the first switch to the second switch includes redirecting the data from a first network interface controller coupled with the first switch to a second network interface controller coupled with the second switch.
15. The computer program product of claim 13 , wherein the first and second switches are included in a multi-node chassis along with the plurality of compute nodes.
16. The computer program product of claim 13 , wherein the program instructions are further executable by one or more processors to:
determine, prior to reducing the data transfer rate to the port, that fans within the first switch are operating at a maximum fan speed setpoint.
17. The computer program product of claim 13 , wherein the compute node reducing the data transfer rate to the port further includes renegotiating a link speed between the compute node and the port of the first switch, wherein the link speed limits the data transfer rate between the compute node and the port, and wherein the renegotiated link speed is less than a previous link speed that was used while the temperature of the port exceeded the temperature limit.
18. The method of claim 1 , wherein the compute node, the management node, the first switch and the second switch are included in a multi-node chassis.
19. The method of claim 8 , wherein the plurality of compute nodes, the management node, the first switch and the second switch are included in a multi-node chassis.