In an aspect of the disclosure, a method, a computer-readable medium, and a computer system are provided. The computer system includes a baseboard management controller (BMC). The BMC receives a first message from a first remote device on a management network. The BMC determines whether the first message is directed to a storage service or fabric service running on a host of the BMC. The host is a storage device. The BMC extracts a service management command from the first message, when the first message is directed to the storage service or fabric service. The BMC sends, through a BMC communication channel to the host, a second message containing the service management command to the host. The BMC communication channel established for communicating baseboard management commands between the BMC and the host.

A computer system includes a BMC and a host of the BMC. The BMC receives a first message from a first remote device on a management network. The BMC determines whether the first message is directed to a storage service or fabric service executed on a main processor of a storage controller of the host. The host is a storage device. The storage controller includes an RDMA controller in communication with the main processor through an internal communication channel of the storage controller. The RDMA controller is managed by the storage service. The BMC extracts a service management command from the first message, when the first message is directed to the storage service or fabric service. The BMC sends, through a BMC communication channel established for communicating baseboard management commands between the BMC and the host, a second message containing the service management command to the host.

An electronic device is provided. The electronic device may include a plurality of electronic circuits comprising a first electronic circuit and a second electronic circuit; and a processor configured to control at least one of the first electronic circuit or the second electronic circuit depending on at least one parameter indicating thermal properties of the first electronic circuit and a predetermined parameter indicating a thermal interdependency between the first electronic circuit and the second electronic circuit.

A server with a double-firmware storage space and a firmware update method therefor are provided. The server includes a baseboard management controller (BMC) with a control module and an update module, a selection circuit, a first storage circuit, and a second storage circuit. The control module executes a first firmware program from the first storage circuit. The update module stores a second firmware program in the second storage circuit according to a firmware update instruction when the control module executes the first firmware program. The BMC resets after the second firmware program is stored in the second storage circuit. After the BMC is reset, the control module obtains the second firmware program from the second storage circuit and executes the second firmware program.

A method for monitoring a server, a monitoring device and a monitoring system are provided. The method includes: setting a system event to be monitored from the server; setting a trigger recording mode and a predetermined recording amount for recording system information of the server; capturing the system information related to the system event from a baseboard management controller (BMC) managing the server into a storage circuit according to the trigger recording mode and the predetermined recording amount; and displaying the system information stored in the storage circuit on an user interface.

A network monitoring system comprising: a prediction unit configured to acquire information needed for detecting an abnormality in a virtual network function provided in a virtualization infrastructure, and predict performance information of a virtual network function that is to be newly provided and monitored; a generation unit configured to generate an abnormality determination rule to serve as a reference for determining whether or not actual performance information of a virtual network function falls within an allowed range that is set based on the predicted performance information; a determination unit configured to perform abnormality determination using the actual performance information of the virtual network function and the abnormality determination rule; and an output unit configured to output an abnormality notification if it is determined by the determination unit that there is an abnormality.

This disclosure provides a server structure, a management controller of the mainboard of a mainboard obtains a sensor record from a resource box and saves the sensor record in the server. The mainboard has an external interface, and the management controller of the mainboard of the mainboard externally provides the sensor record of the mainboard and the obtained from the resource box of the resource box management controller of the mainboard through the external interface. This disclosure further provides a method for resource aggregation for the server, which comprises: S1: the management controller of the mainboard of the mainboard loads the sensor record of the resource box to the server; S2: the management controller of the mainboard of the mainboard integrals the sensor record of the resource box and the sensor record of the mainboard into a whole set sensor record.

A computer program product may include storage media embodying program instructions executable by a baseboard management controller (BMC) within a compute node to: receive a request to install a central processing unit (CPU) in the compute node; identify a current hardware configuration of the compute node; identify a plurality of available locations within the compute node that are compatible with installation of the CPU; calculate, for each of the identified plurality of available locations, a predicted performance score for the CPU on the basis that the CPU were to be installed in the available location, wherein the predicted performance scores are calculated in response to receiving the request; select a location from among the plurality of available locations that is associated with the greatest performance score for the CPU; and generate user output indicating the selected location where the CPU should be installed.

Disclosed are various embodiments of power source redundancy in a power supply for a rack mounted computing device. The power supply includes a plurality of AC power converters configured to receive power from corresponding power sources. A first AC power converter provides DC power to a common DC bus of the power supply. A second AC power converter provides DC power to the common DC bus in response to a change in the voltage level provided by the first AC power converter.

A mechanism is described for facilitating dynamic and trusted cloud-based extension upgrades for computing systems according to one embodiment of the invention. A method of embodiments of the invention includes detecting a computing device needing an upgrade. The upgrade may relate to a hardware component at the computing device needing an upgrade element for the upgrade. The method may further include calling a first cloud server to provide the upgrade over a network. The first cloud server may have first resources including the upgrade element. The method may further include facilitating the hardware component to access the upgrade element at the first cloud server without having to upgrade or replace the hardware component.