Methods and apparatus for maintaining the cooling systems of distributed compute systems
AssigneeIntel Corporation
InventorsFrancesc Guim Bernat, Amruta Misra, Arun Hodigere, Kshitij Arun Doshi, John J. Browne
Life of this patent
Estimating the term…
Abstract
Methods and apparatus for maintaining the cooling systems of distributed compute systems are disclosed. An example apparatus disclosed herein includes memory, machine-readable instructions, and processor circuitry to execute the machine-readable instructions to determine a health of a server, determine a threshold based on a workload service level agreement associated with the server, and in response to determining the health does not a satisfy the threshold, throttle a workload on the server.
Claims17
An apparatus comprising:
- memory;
- machine-readable instructions;
- and
- at least one processor circuit to execute the machine-readable instructions to:
- determine a first component health of a compute component of a server;
- determine a second component health of a cooling component of a cooling system for the server;
- determine a health of the server based on the first component health and the second component health;
- determine a threshold based on a workload service level agreement associated with the server;
- and
- in response to determining the health of the server does not a satisfy the threshold:
- throttle a workload on the server;
- determine a first maintenance window of the compute component based on the first component health;
- determine a second maintenance window of the cooling component based on the second component health;
- and
- schedule a maintenance period for the server, the maintenance period scheduled to be within the first maintenance window and the second maintenance window.
A method comprising:
- determining a first component health of a compute component of a server;
- determining a second component health of a cooling component of a cooling system for the server;
- determining a health of the server based on the first component health and the second component health;
- determining a threshold based on a workload service level agreement associated with the server;
- and
- in response to determining the health of the server does not a satisfy the threshold:
- throttling a workload on the server;
- determining a first maintenance window of the compute component based on the first component health;
- determining a second maintenance window of the cooling component based on the second component health;
- and
- scheduling, by executing instructions with at least one processor circuit, a maintenance period for the server, the maintenance period scheduled to be within the first maintenance window and the second maintenance window.
A non-transitory computer readable storage medium comprising machine-readable instructions to cause one or more processor circuits to:
- determine a first component health of a compute component of a server;
- determine a second component health of a cooling component of a cooling system for the server;
- determine a health of the server including a cooling system based on the first component health and the second component health;
- determine a threshold based on a workload service level agreement associated with the server;
- and
- in response to determining the health of the server does not a satisfy the threshold:
- throttle a workload on the server;
- determine a first maintenance window of the compute component based on the first component health;
- determine a second maintenance window of the cooling component based on the second component health;
- and
- schedule a maintenance period for the server, the maintenance period scheduled to be within the first maintenance window and the second maintenance window.
Dependencies are inferred from the claim text.