Abstract: Methods and apparatuses for coordination of power state management in and electronic system.

Abstract: Systems and methods of managing power provide for receiving notification of a pending power state transition and using coordination hardware to determine whether the power state transition in a primary device is permitted by a set of secondary devices. In one embodiment, the primary device shares a resource with the set of secondary devices.

Abstract: Power consumption by a computer system may vary all the time based upon software and the workload. Facility such as data centers host multiple of computer systems. With continuous growing demand for power and cooling of computer systems, data centers face limitations on their ability to provide the power and cooling capability. These limitations are occasionally exasperated by problems in either power or cooling systems. The computer systems may have a method to maintain total power consumption below a set target level. An enterprise power and thermal manager, EPTM, may change this setting dynamically to improve efficiency of supporting power and cooling infrastructure. In addition, the EPTM may use this ability to improve performance, availability and to provide ability to implement various administrative policies.

Abstract: Provided is a technique for power and performance management of one or more storage devices. With a power and performance management agent, a power change notification identifying a power set point is received and a power state of at least one storage device is adjusted.

Abstract: A processor that includes a digital throttle to monitor the activity of the execution pipeline and to change a frequency of a first or second PLL clock within a single clock cycle based on a power state.

Abstract: Power consumption in a system is monitored by a controller. The controller uses a power consumption policy to determine when the power consumption of the system needs to be adjusted. Adjustment is made by gradually reducing power consumptions of one or more components in the system or allowing the one or more components to gradually increase the power consumption.

Abstract: A method is described for computing the power consumption of a device. The method comprises (1) determining the number and type of each of a plurality of components comprising a system, (2) determining a component power consumption for each of the plurality of components, and (3) determining a device power consumption by summing the component power consumptions of all of the plurality of components. A rack holds multiple systems, provides power to the systems and cool air to remove thermal energy from the systems. Power and cooling requirements for a rack are determined by (1) determining the number and type of components comprising each of the plurality of systems, (2) determining component power consumption for each of the plurality of components, (3) determining a system power consumption for each system by summing the component power consumptions of all components in each system, and (4) determining a rack power consumption by summing the system power consumptions for all systems.

Abstract: A method and apparatus for a compiler, or similar computer language translating device, to translate a computer language into a sequence of electronic instructions to be executed at run-time by at least one functional unit in a computer. At compile time, the compiler creates and analyzes the sequential order of the electronic instructions to determine exact moments when the functional unit will begin to, or complete, executing the electronic instructions. Consequently, the compiler can predict time intervals when the functional unit will be in use, or idling between instructions. In addition, the compiler knows the delay time, or latency, involved in powering up and powering down the functional unit. The compiler compares the use times, or idle times, to the latency, and creates power-controlling instructions to be embedded into the sequence of electronic instructions. The power-controlling instructions are to control power to the functional units at run-time.

Abstract: Power consumption by a computer system may vary all the time based upon software and the workload. Facility such as data centers host multiple of computer systems. With continuous growing demand for power and cooling of computer systems, data centers face limitations on their ability to provide the power and cooling capability. These limitations are occasionally exasperated by problems in either power or cooling systems. The computer systems may have a method to maintain total power consumption below a set target level. An enterprise power and thermal manager, EPTM, may change this setting dynamically to improve efficiency of supporting power and cooling infrastructure. In addition, the EPTM may use this ability to improve performance, availability and to provide ability to implement various administrative policies.

Abstract: Pads are arranged as an integrated circuit (IC) footprint, and are formed in a stackup that includes an insulating layer and multiple signal routing layers. The footprint has a polygonal shape. There is an inner pad region, a middle pad region that surrounds the inner pad region, and an outer pad region that surrounds the middle pad region. Some of the pads of each pad region are connected to a respective group of signal lines. Some of the signals that are connected to pads of the outer region which are located in a corner of the polygonal shape are routed out of the footprint in a different layer than the one used to route signal lines that are connected to pads of the outer region which are located between two adjacent corners of the polygonal shape.

Abstract: A method and apparatus for a compiler, or similar computer language translating device, to translate a computer language into a sequence of electronic instructions to be executed at run-time by at least one functional unit in a computer. At compile time, the compiler creates and analyzes the sequential order of the electronic instructions to determine exact moments when the functional unit will begin to, or complete, executing the electronic instructions. Consequently, the compiler can predict time intervals when the functional unit will be in use, or idling between instructions. In addition, the compiler knows the delay time, or latency, involved in powering up and powering down the functional unit. The compiler compares the use times, or idle times, to the latency, and creates power-controlling instructions to be embedded into the sequence of electronic instructions. The power-controlling instructions are to control power to the functional units at run-time.

Abstract: Pads are arranged as an integrated circuit (IC) footprint, and are formed in a stackup that includes an insulating layer and multiple signal routing layers. The footprint has a polygonal shape. There is an inner pad region, a middle pad region that surrounds the inner pad region, and an outer pad region that surrounds the middle pad region. Some of the pads of each pad region are connected to a respective group of signal lines. Some of the signals that are connected to pads of the outer region which are located in a corner of the polygonal shape are routed out of the footprint in a different layer than the one used to route signal lines that are connected to pads of the outer region which are located between two adjacent corners of the polygonal shape.

Abstract: Power consumption in a system is monitored by a controller. The controller uses a power consumption policy to determine when the power consumption of the system needs to be adjusted. Adjustment is made by gradually reducing power consumptions of one or more components in the system or allowing the one or more components to gradually increase the power consumption.

Abstract: A method is described for computing the power consumption of a device. The method comprises (1) determining the number and type of each of a plurality of components comprising a system, (2) determining a component power consumption for each of the plurality of components, and (3) determining a device power consumption by summing the component power consumptions of all of the plurality of components. A rack holds multiple systems, provides power to the systems and cool air to remove thermal energy from the systems. Power and cooling requirements for a rack are determined by (1) determining the number and type of components comprising each of the plurality of systems, (2) determining component power consumption for each of the plurality of components, (3) determining a system power consumption for each system by summing the component power consumptions of all components in each system, and (4) determining a rack power consumption by summing the system power consumptions for all systems.

Abstract: A system is disclosed for the regulation of electrical component temperature and power consumption rate through the reconfiguration between different interconnect bus widths.