Abstract: A process includes an application layer of a host of a computer platform using a smart network interface card (NIC) of the computer platform to provide an input/output (I/O) service for the application layer. The I/O service includes a service that is associated with a cloud operator domain; the smart NIC is installed in a connector; and the application layers associated with a cloud tenant domain. The process includes a baseboard management controller of the smart NIC managing the computer platform. Managing the computer platform includes the baseboard management controller managing the host; the baseboard management controller managing components of the smart NIC other than the baseboard management controller; and managing the host includes the baseboard management controller communicating with the host via the connector to control a system power state of the computer platform.

Abstract: A process includes a port of a bridge providing a reference clock signal to a first end of an interconnect extending between the first port and a network interface controller. The reference clock signal propagates over the interconnect to provide, at a second end of the interconnect, a delayed reference clock signal at the network interface controller. Pursuant to the process, the bridge senses a timing of the delayed reference clock signal. The process includes communicating management traffic between a network interface of a baseboard management controller and the network interface controller via the interconnect. The communication of the management traffic includes the port, responsive to the sensing of the timing of the delayed reference clock signal, synchronizing communication of data with the first end of the interconnect to the delayed reference clock signal.

Abstract: A process includes a port of a bridge providing a reference clock signal to a first end of an interconnect extending between the first port and a network interface controller. The reference clock signal propagates over the interconnect to provide, at a second end of the interconnect, a delayed reference clock signal at the network interface controller. Pursuant to the process, the bridge senses a timing of the delayed reference clock signal. The process includes communicating management traffic between a network interface of a baseboard management controller and the network interface controller via the interconnect. The communication of the management traffic includes the port, responsive to the sensing of the timing of the delayed reference clock signal, synchronizing communication of data with the first end of the interconnect to the delayed reference clock signal.

Abstract: A supervisory service of a node that includes a smart input/output (I/O) peripheral is extended into a cloud operator domain that is associated with the smart I/O peripheral. The supervisory service determines a state of a ready state indicator that is provided by the smart I/O peripheral. Based on the state, the supervisory service performs at least one of regulating an availability of an instance of an application operating environment of the node or determining whether the smart I/O peripheral is ready to be configured by the supervisory service.

Abstract: A process includes an application layer of a host of a computer platform using a smart network interface card (NIC) of the computer platform to provide an input/output (I/O) service for the application layer. The I/O service includes a service that is associated with a cloud operator domain; the smart NIC is installed in a connector; and the application layers associated with a cloud tenant domain. The process includes a baseboard management controller of the smart NIC managing the computer platform. Managing the computer platform includes the baseboard management controller managing the host; the baseboard management controller managing components of the smart NIC other than the baseboard management controller; and managing the host includes the baseboard management controller communicating with the host via the connector to control a system power state of the computer platform.

Abstract: A supervisory service of a node that includes a smart input/output (I/O) peripheral is extended into a cloud operator domain that is associated with the smart I/O peripheral. The supervisory service determines a state of a ready state indicator that is provided by the smart I/O peripheral. Based on the state, the supervisory service performs at least one of regulating an availability of an instance of an application operating environment of the node or determining whether the smart I/O peripheral is ready to be configured by the supervisory service.

Abstract: Example implementations relate to a self-tune controller. For example, the self-tune controller may poll, via an out-of-band data stream, low-level operation information about a processor or a bus of a computing system under a present workload. At least some of the low-level operation information may be descriptive of a nature of traffic on the bus. The self-tune controller may program, via an out-of-band control signal, a setting of the computing system for the present workload based on the low-level operation information.

Abstract: In one example in accordance with the present disclosure, a method for determining operating frequencies includes receiving a warranty period for a computer component. The method includes determining an operating frequency that will cause a predicted life cycle of the computer component operating at the operating frequency to fall within the warranty period. The method includes setting the computer component to operate at the operating frequency.

Abstract: In one example in accordance with the present disclosure, a method includes mapping, using post-package repair, an address associated with a first memory row of a computing device to a spare memory row of the computing device, wherein the spare memory row has a memory failure pattern, and reading data from the spare memory row.

Abstract: In one example in accordance with the present disclosure, a method includes mapping, using post-package repair, an address associated with a first memory row of a computing device to a spare memory row of the computing device, wherein the spare memory row has a memory failure pattern, and reading data from the spare memory row.

Abstract: In one example in accordance with the present disclosure, a method for determining operating frequencies includes receiving a warranty period for a computer component. The method includes determining an operating frequency that will cause a predicted life cycle of the computer component operating at the operating frequency to fall within the warranty period. The method includes setting the computer component to operate at the operating frequency.

Abstract: Example implementations relate to a self-tune controller. For example, the self-tune controller may poll, via an out-of-band data stream, low-level operation information about a processor or a bus of a computing system under a present workload. At least some of the low-level operation information may be descriptive of a nature of traffic on the bus. The self-tune controller may program, via an out-of-band control signal, a setting of the computing system for the present workload based on the low-level operation information.

Abstract: In a system, temperature measurements are logged. From the logged measurements, duration of operation of the system in each of a plurality of temperature bands is determined.

Abstract: A processing circuit independent of a processor determines a current utilization of the processor, based on events of an execution pipeline of the processor. According to the determined utilization, the processing circuit causes the processor to transition from a first of the plurality of performance states to a second of the plurality of performance states.

Abstract: In a system, temperature measurements are logged. From the logged measurements, duration of operation of the system in each of a plurality of temperature bands is determined.

Abstract: A processing circuit independent of a processor determines a current utilization of the processor, based on events of an execution pipeline of the processor. According to the determined utilization, the processing circuit causes the processor to transition from a first of the plurality of performance states to a second of the plurality of performance states.

Abstract: A method and apparatus for controlling the power usage of a processor is disclosed. The power usage of the processor is monitored. When the power usage of the processor exceeds a threshold power usage value, the power used by the processor is reduced or limited. A processor utilization value is also monitored. When the processor utilization value is above a threshold utilization value, the processor is ramped to a higher performance state. When power to the processor is being limited, a first rate is used to ramp the processor to the higher performance state. When power to the processor is not being limited, then a second rate, different from the first rate, is used to ramp the processor to the higher performance state.

Abstract: A method and apparatus for controlling the power usage of a processor is disclosed. The power usage of the processor is monitored. When the power usage of the processor exceeds a threshold power usage value, the power used by the processor is reduced or limited. A processor utilization value is also monitored. When the processor utilization value is above a threshold utilization value, the processor is ramped to a higher performance state. When power to the processor is being limited, a first rate is used to ramp the processor to the higher performance state. When power to the processor is not being limited, then a second rate, different from the first rate, is used to ramp the processor to the higher performance state.

Abstract: A method and system manages revisions of software code for one or more servers in an infrastructure. An indication of the version of the code currently residing with one of the servers is transmitted to the infrastructure. An indication of the version of the code currently residing with the infrastructure for the server's type is also transmitted to the one of the servers. A determination is made as to whether the server's version should be updated to the infrastructure's, whether the infrastructure's version should be updated to the server's, or neither based on the transmitted indications in view of a set of preconfigured update rules. If the infrastructure's version should be updated to the server's, the image of the code currently residing with the infrastructure is replaced with an image of the code currently residing with the server.

Abstract: A system includes hardware, a software layer, a platform layer, and a management communication channel between the software layer and the platform layer. The management communication channel provides an interface to enable the software layer to issue a hardware management command to the platform layer, where the hardware management command is to specify a change of a setting of the hardware, and where the management communication channel allows a hardware management engine of the platform layer to collaborate with the software layer to perform the change of the setting of the hardware.