Abstract: Power supply efficiency may be provided. First, a total power supply capacity may be determined comprising a sum of a plurality of supply capacities respectively corresponding to a plurality of power supplies serving a plurality of components. Next, a load value corresponding to the plurality of components may be determined. A number of the plurality of power supplies may then be powered down. The number of power supplies powered down may comprise a value that may cause a remaining number of the plurality of power supplies serving the plurality of components to operate within an efficiency range.

Abstract: Power supply efficiency may be provided. First, a total power supply capacity may be determined comprising a sum of a plurality of supply capacities respectively corresponding to a plurality of power supplies serving a plurality of components. Next, a load value corresponding to the plurality of components may be determined. A number of the plurality of power supplies may then be powered down. The number of power supplies powered down may comprise a value that may cause a remaining number of the plurality of power supplies serving the plurality of components to operate within an efficiency range.

Abstract: Power supply efficiency may be provided. First, a total power supply capacity may be determined comprising a sum of a plurality of supply capacities respectively corresponding to a plurality of power supplies serving a plurality of components. Next, a load value corresponding to the plurality of components may be determined. A number of the plurality of power supplies may then be powered down. The number of power supplies powered down may comprise a value that may cause a remaining number of the plurality of power supplies serving the plurality of components to operate within an efficiency range.

Abstract: In one embodiment, a machine learning server in a computer network determines a plurality of computing features shared across a given set of computing products, and collects, from each computing product of the given set, problem-solution data for each computing feature of the plurality of computing features. Problem-solution data is indicative of problems related to a respective computing feature, attempted solution actions for the problems, and outcomes of the attempted solutions on the problem. The machine learning server updates a machine learning model of suggested solutions for computing-feature-specific problems based on the collected problem-solution data, and provides, based on the machine learning model, a particular suggested solution for a particular computing-feature-specific problem to a particular computing product.

Abstract: Embodiments described herein improve availability of a power plane in a network device by using a PoE manager that is separate from an operating system in the network device. For example, when the operating system (or a PoE application executing in the operating system) becomes unavailable, either because of failure or system upgrade, the PoE manager continues managing a power plane in the network device such that connected PDs continue to receive DC power. Stated differently, by using a PoE manager that is separate from the operating system, there is no fate sharing between the PoE manager and the operating system. If the operating system is unavailable, the PoE manager continues to provide the same power allotment to the PDs. As such, updates and failures which previously made the power plane unavailable no longer affect the power supplied to the PDs.

Abstract: Power supply efficiency may be provided. First, a total power supply capacity may be determined comprising a sum of a plurality of supply capacities respectively corresponding to a plurality of power supplies serving a plurality of components. Next, a load value corresponding to the plurality of components may be determined. A number of the plurality of power supplies may then be powered down. The number of power supplies powered down may comprise a value that may cause a remaining number of the plurality of power supplies serving the plurality of components to operate within an efficiency range.

Abstract: Power supply efficiency may be provided. First, a total power supply capacity may be determined comprising a sum of a plurality of supply capacities respectively corresponding to a plurality of power supplies serving a plurality of components. Next, a load value corresponding to the plurality of components may be determined. A number of the plurality of power supplies may then be powered down. The number of power supplies powered down may comprise a value that may cause a remaining number of the plurality of power supplies serving the plurality of components to operate within an efficiency range.

Abstract: Embodiments described herein improve availability of a power plane in a network device by using a PoE manager that is separate from an operating system in the network device. For example, when the operating system (or a PoE application executing in the operating system) becomes unavailable, either because of failure or system upgrade, the PoE manager continues managing a power plane in the network device such that connected PDs continue to receive DC power. Stated differently, by using a PoE manager that is separate from the operating system, there is no fate sharing between the PoE manager and the operating system. If the operating system is unavailable, the PoE manager continues to provide the same power allotment to the PDs. As such, updates and failures which previously made the power plane unavailable no longer affect the power supplied to the PDs.

Abstract: Power supply efficiency may be provided. First, a total power supply capacity may be determined comprising a sum of a plurality of supply capacities respectively corresponding to a plurality of power supplies serving a plurality of components. Next, a load value corresponding to the plurality of components may be determined. A number of the plurality of power supplies may then be powered down. The number of power supplies powered down may comprise a value that may cause a remaining number of the plurality of power supplies serving the plurality of components to operate within an efficiency range.

Abstract: Power supply efficiency may be provided. First, a total power supply capacity may be determined comprising a sum of a plurality of supply capacities respectively corresponding to a plurality of power supplies serving a plurality of components. Next, a load value corresponding to the plurality of components may be determined. A number of the plurality of power supplies may then be powered down. The number of power supplies powered down may comprise a value that may cause a remaining number of the plurality of power supplies serving the plurality of components to operate within an efficiency range.

Abstract: In one embodiment, a system, apparatus, and method are described for providing limited data connectivity for devices connected to a switch when the switch enters bootloader mode. The switch has a central processing unit (CPU), the CPU comprising a reload handler and an application-specific integrated circuit (ASIC), the ASIC comprising ASIC forwarding logic. The ASIC is instructed, by the reload handler, to store an ASIC database, the ASIC database for storing the ASIC forwarding logic. The reload handler maintains a physical layer (PHY) state of the switch. Use of spanning tree protocol (STP) and Transmission Control Protocol (TCP) keepalive is disabled in the switch. A state of stack hardware is retained in switch memory. New ports of the switch are prevented from becoming active, and the ASIC forwarding logic is, in response to receiving the instruction, stored in the ASIC database. Related systems, apparatuses, and methods are also described.

Abstract: Power supply efficiency may be provided. First, a total power supply capacity may be determined comprising a sum of a plurality of supply capacities respectively corresponding to a plurality of power supplies serving a plurality of components. Next, a load value corresponding to the plurality of components may be determined. A number of the plurality of power supplies may then be powered down. The number of power supplies powered down may comprise a value that may cause a remaining number of the plurality of power supplies serving the plurality of components to operate within an efficiency range.

Abstract: In one embodiment, a system, apparatus, and method are described for providing limited data connectivity for devices connected to a switch when the switch enters bootloader mode. The switch has a central processing unit (CPU), the CPU comprising a reload handler and an application-specific integrated circuit (ASIC), the ASIC comprising ASIC forwarding logic. The ASIC is instructed, by the reload handler, to store an ASIC database, the ASIC database for storing the ASIC forwarding logic. The reload handler maintains a physical layer (PHY) state of the switch. Use of spanning tree protocol (STP) and Transmission Control Protocol (TCP) keepalive is disabled in the switch. A state of stack hardware is retained in switch memory. New ports of the switch are prevented from becoming active, and the ASIC forwarding logic is, in response to receiving the instruction, stored in the ASIC database. Related systems, apparatuses, and methods are also described.

Abstract: In one embodiment, a machine learning server in a computer network determines a plurality of computing features shared across a given set of computing products, and collects, from each computing product of the given set, problem-solution data for each computing feature of the plurality of computing features. Problem-solution data is indicative of problems related to a respective computing feature, attempted solution actions for the problems, and outcomes of the attempted solutions on the problem. The machine learning server updates a machine learning model of suggested solutions for computing-feature-specific problems based on the collected problem-solution data, and provides, based on the machine learning model, a particular suggested solution for a particular computing-feature-specific problem to a particular computing product.

Abstract: Embodiments described herein improve availability of a power plane in a network device by using a PoE manager that is separate from an operating system in the network device. For example, when the operating system (or a PoE application executing in the operating system) becomes unavailable, either because of failure or system upgrade, the PoE manager continues managing a power plane in the network device such that connected PDs continue to receive DC power. Stated differently, by using a PoE manager that is separate from the operating system, there is no fate sharing between the PoE manager and the operating system. If the operating system is unavailable, the PoE manager continues to provide the same power allotment to the PDs. As such, updates and failures which previously made the power plane unavailable no longer affect the power supplied to the PDs.

Abstract: Methods and systems are disclosed which may provide MultiGig capability to a system where a physical layer device (PHY) or a network device does not have the capacity to support all available line speeds while operating in a single system-interface mode between MAC and PHY devices.

Abstract: Embodiments described herein improve availability of a power plane in a network device by using a PoE manager that is separate from an operating system in the network device. For example, when the operating system (or a PoE application executing in the operating system) becomes unavailable, either because of failure or system upgrade, the PoE manager continues managing a power plane in the network device such that connected PDs continue to receive DC power. Stated differently, by using a PoE manager that is separate from the operating system, there is no fate sharing between the PoE manager and the operating system. If the operating system is unavailable, the PoE manager continues to provide the same power allotment to the PDs. As such, updates and failures which previously made the power plane unavailable no longer affect the power supplied to the PDs.

Abstract: Methods and systems are disclosed which may provide MultiGig capability to a system where a physical layer device (PHY) or a network device does not have the capacity to support all available line speeds while operating in a single system-interface mode between MAC and PHY devices.

Abstract: Disclosed herein are systems and methods of negotiating an acceptable speed for communication between at least two Ethernet devices connected by a link. In one aspect, the method monitors the link health and status and makes a speed choice decision based on the error rate and link partner capability.

Abstract: Embodiments described herein improve availability of a power plane in a network device by using a PoE manager that is separate from an operating system in the network device. For example, when the operating system (or a PoE application executing in the operating system) becomes unavailable, either because of failure or system upgrade, the PoE manager continues managing a power plane in the network device such that connected PDs continue to receive DC power. Stated differently, by using a PoE manager that is separate from the operating system, there is no fate sharing between the PoE manager and the operating system. If the operating system is unavailable, the PoE manager continues to provide the same power allotment to the PDs. As such, updates and failures which previously made the power plane unavailable no longer affect the power supplied to the PDs.