Microsoft Technology Licensing, LLC

Paresh J Maisuria

M Nashaat Soliman

James Anthony Schwartz, Jr

Arun Upadhyaya Kishan

Matthew H Holle

Ranveer Chandra

Anirudh Badam

Shira Weinberg

Emily Nicole Wilson

Gurpreet Virdi

Jared Adam Brown

Sandeep Prabhakar

Jihad Tafas

Murtuza S Naguthanawala

Mohammed Nashaat Soliman

Aniruddha Jayant Jahagirdar

Tapan Ansel

Hari R Pulapaka

Benjamin M Schultz

Rahul Nair

Randall Edward Aull

Doron J Holan

James George Cavalaris

Narayanan Ganapathy

Iulian Doroftei Calinov

Margarit Simeonov Chenchev

Ankit Srivastava

Samarth Harish Shah

Graham Wong

Abhishek Sagar

Aacer Hatem Daken

Taofiq Ezaz

Zachary Sandburg Holmes

Gabriel Knezek

Eric Kratzer

Julian Doroftei Calinov

James Kennyth Truckle

A computing device has an energy storage device system with multiple energy storage devices. Various different criteria are used to determine which one or more of the multiple energy storage devices to charge or discharge at any given time to provide power to the computing device. The criteria can include characteristics of the energy storage devices as well as hardware and/or physical characteristics of the computing device, characteristics of the energy storage devices and/or the computing device that change while the computing device operates, and predicted behavior or usage of the computing device. These criteria are evaluated during operation of the computing device, and the appropriate energy storage device(s) from which to draw power or to charge at any given time based on these criteria are determined.

Dynamic power source selection, charging, and discharging

Pooling computing resources based on inferences about a plurality of hardware devices. The method includes identifying inference information about the plurality of devices. The method further includes based on the inference information optimizing resource usage of the plurality of hardware devices.

User presence prediction driven device management

A computing device has an energy storage device system with one or more energy storage devices. A target run-time is obtained, which refers to how long the computing device is to run given the current amount of energy in the energy storage device(s). A predicted power usage over the target run-time is determined, and what, if any, power management actions to take in order to achieve the target run-time are determined. The power management actions are then taken. A target charge-time is also obtained, which refers to how long the computing device is to take to charge the energy storage device(s) to a threshold level (e.g., 100% or fully charged). A predicted power gain over the target charge-time is determined, and what, if any, power management actions to take in order to achieve the target charge-time are determined. The power management actions are then taken.

Target based power management

Deploying containers constrained by power profiles on a host system. A method includes identifying a container template, a set of reusable stored characteristics, to be used for deploying a container instance. The method further includes obtaining a power profile, defining at least one power consumption threshold, for the container instance based on at least one of the set of reusable stored characteristics of the container template or other information about the container instance. The method further includes deploying the container instance on the host system by applying the set of reusable stored characteristics and the power profile by applying one or more configuration layers which causes power to the container instance to be at least one of regulated or monitored based on information in the power profile.

Power and energy profiling for efficient virtual environments

Computing resource management of computing devices

A computing device has an energy storage device system with multiple energy storage devices. Various different criteria are used to determine which one or more of the multiple energy storage devices to discharge at any given time to provide power to the computing device. The criteria can include characteristics of the energy storage devices as well as hardware and/or physical characteristics of the computing device, characteristics of the energy storage devices and/or the computing device that change while the computing device operates, and predicted behavior or usage of the computing device. These criteria are evaluated during operation of the computing device, and the appropriate energy storage device(s) from which to draw power at any given time based on these criteria are determined.

Dynamic energy storage device discharging

Hibernating a computing system. The method includes detecting at least one condition which indicates the availability of time to perform hibernation optimization steps on the computing system. The hibernation optimization steps optimize resume time from hibernate. The method further includes detecting a trigger indicating that the computing system is to be put into a hibernate state. As a result of detecting the at least one condition and the trigger, the method further includes performing hibernation steps to hibernate the computing system in a way that optimizes resume time from hibernate of the computing system.

Optimized power transitions based on user or platform conditions

Consumption of power by device components is profiled on a per-session basis of user interaction. A session engine determines start of a user interaction session with the device (e.g., by detecting screen turn on). The engine generates a session identifier (SID) and broadcasts same to device component(s). In response to the SID, the component(s) record power consumption data. Upon receiving a signal indicating end of the user interaction session (e.g., screen turn off), the session engine broadcasts a notice allowing component(s) to stop recording power consumption data. The components communicate aggregated data to the session engine for storage in a centralized location (e.g., trace buffer). In response to a query posed within a command line prompt, stored data may be parsed on a per-session basis for inclusion in a detailed report of power consumption. Power consumption inefficiencies endemic to components and/or user behaviors may be thus be identified and remediated.

Per-screen on session power consumption profiling

An operating system of a computing device determines an importance of the threads running on the computing device, such as assigning the importance of the threads as critical or non-critical. The operating system determines when there are no threads having at least a threshold importance (e.g., no critical threads), and forces one or more components of the computing device into a forced idle state in response to determining that there are no threads having at least the threshold importance. The forced idle state of a device component is a low power state, such as a state in which program instructions are not executed, so the computing device is forced into a forced idle state that reduces power usage in the computing device despite there being threads (e.g., non-critical threads) to be executed.

Power conservation of computing components using forced idle state

Adaptive doze to hibernate scheme techniques are described for power management of a computing device. Rather than relying upon a fixed timer to control device power states, the adaptive doze to hibernate scheme monitors various hibernate parameters and adapts the hibernation experience in dependence upon the parameters. The hibernate parameters may include but are not limited to a standby budget, minimum standby time, reserve screen on time, and indications of user presence. In operation, a power manager monitors battery drain rate and adaptively determines when to change the device power states of the computing device based on the observed drain rate and the hibernate parameters. The power manager may selectively switch between various states (e.g., high performance, active, wake, standby, hibernate, off, etc.) accordingly.

Adaptive doze to hibernate

A computer system that accesses multi-transport devices and associates device instances of the multi-transport device with each other. Devices provide the same device identifier such as a device unique identifier regardless of the transport over which the device is accessed. A visual representation for the multi-transport device may be provided to a user using the device identifier. Also, a single functional representation for the multi-transport device may be provided based on an identifier provided by each function of the device. The functional representation may be independent of the addition and removal of transports and allows abstracting device access via multiple transports from a user and an application. Transports may be selected at an application and/or a system level. Operating characteristics of available transports may be used to select a desired transport, based on transport requirements specified by the application, the user or the system.

Generalized architecture to support representation of multi-transport devices

Growing community of inventors

Bellevue, WA, United States of America

Paresh J Maisuria