Microsoft Technology Licensing, LLC

Harish Srinivasan

Benjamin M Schultz

Robert C Combs

James Christopher Gray

Javier Nisim Flores Assad

Daniel Frederick Cuomo, Jr

Hari R Pulapaka

Omar Cardona

Margarit Simeonov Chenchev

Jean Khawand

Alexander Malysh

Khoa Anh To

Kevin A Chin

Milan Dasgupta

Nishad Mulye

Rao Salapaka

Keith Edgar Horton

Graham Wong

Yogesh Sreenivasan

Sandeep Rangarajan

Alvin K Tan

Kiran Vadde

Gopikrishna Kannan

Vu Phan Hoang Nguyen

Tian Tan

Denys Gurzhii

David Richard Powell, Jr

Paul Anirban

Chris C Gray

Tom Xanthos

Chenyan Liu

Satish Goswami

Arup Roy

David Powell

Omar M Maabreh

Ryan Gregory Wood

Yonghong Yang

Javier N Flores Assad

James C Gray

David Richard Powell

A method is presented for processing directional data packet traffic. A main data packet, characterized by a plurality of main data packet characteristics including at least a first IP address representing a first source, and a second IP address representing a first destination, is received at a rule processing engine. Based on the main data packet characteristics, a set of rules for processing the main data packet is retrieved. A main direction unified flow for the main data packet, and a speculative reverse direction unified flow for a reverse direction data packet are generated based on the main data packet characteristics and the retrieved set of rules. The main data packet is processed based on the main direction unified flow. Responsive to receiving the reverse direction data packet, the reverse direction data packet is processed based on the speculative reverse direction unified flow.

Data packet traffic using speculative unified flow

Techniques for aggregating execution metrics during virtualization are provided. In some embodiments, aggregated execution metrics (e.g., average execution time) are generated and stored for different types of supported virtualization service operations executed by a virtualization service provider (VSP) in a virtualization stack handling requests from a virtualization service client (VSC) running in a computer system emulator. For example, execution calls to the VSP are intercepted, and execution metrics for a triggered virtualization service operation are generated and aggregated into an aggregation entry that represents aggregated performance (e.g., average execution time) of all instances of the virtualization service operation that were completed during an interval (e.g., 1 hour). Aggregated execution metrics may be stored for any number of historical intervals. Since aggregated execution metrics are effectively compressed, diagnostic information about a variety of different types of virtualization service operations may be stored and used to diagnose and repair underperforming components.

Generating, aggregating, and querying virtualization service execution metrics for cloud diagnostics at scale

Techniques are disclosed for configuring network device resources and determining locality of virtual machines in a virtual network of a cloud computing environment. Usage data for computing resources in the computing network is collected. Based on the collected usage data, a time-based usage profile for the computing resources in the computing network is determined. A periodic component in the time-based usage profile is determined. Based on the periodic component, distribution of VMs is determined or the buffer size of network devices is configured.

Proactive placement of virtualized computing resources and tuning of network resources based on seasonal variations

A virtualization partition (VP) is executed by a virtualization layer. The VP contains guest software that executes in isolation within the VP. The guest software has a background task (BT) that needs to be performed in the future. The BT is virtualized by a BT service that executes outside of the VP. The guest software registers the BT, through the virtualization layer, with a BT virtualization service. An event occurs outside of the VP that triggers the BT. The BT virtualization service responds to the event by assuring that the VP is available (executing), and optionally triggers (possibly indirectly) the execution of code in the application.

Virtualized background activations

A computing device is provided, including a processor configured to receive a selection of a performance profile of a plurality of performance profiles. The performance profile may indicate a performance variable goal state for a first network stack component included in a network stack of a plurality of network stack components. In response to receiving the selection of the performance profile, the processor may be further configured to control a network performance parameter of the first network stack component such that a performance variable of the first network stack component approaches the performance variable goal state. The processor may be further configured to transmit, to an additional network stack component of the plurality of network stack components, instructions to control an additional network performance parameter of the additional network stack component such that the performance variable of the first network stack component approaches the performance variable goal state.

Performance profile for network stack components

A supervisory control system provides power management in an electronic device by providing timeout periods for a hardware component to lower levels of the operating system such as a power management arbitrator and/or a hardware interface controller. The power management arbitrator and/or hardware interface controller transition at least a portion of a hardware component to a lower-power state based on monitored activity information of the hardware component. The supervisory control system may further provide wakeup periods to the power management arbitrator and/or a hardware interface controller to determine whether the hardware component should be transitioned to a higher-power state at the end of the wakeup period if the hardware component satisfies a transition condition.

Supervisory control of power management

This application describes client devices that control network transmission based on a power state. The client device determines a power state of the computing device and a priority of an application executing on the computing device. The client device determines whether to permit the application to communicate with a remote network accessible device via the network communication hardware based at least on the priority of the application and the power state of the computing device. Also described is a power state service that selects a client device to be provided with a notification based on power states of the client devices.

Power-aware network communication

A system for aggregated electronic device power management includes an aggregated power management arbitrator configured to collect activity information from one or more electronic devices regarding the use of a type of hardware component on the electronic devices. The aggregated power management arbitrator selects a timeout period for transmitting to one of the one or more electronic devices to define a minimum amount of time before the hardware component is transitioned to a lower-power state. The aggregated power management arbitrator may determine usage patterns for one of the electronic devices based on the collected activity information and the applications executing on the electronic device. A timeout period may be adjusted based on the determined usage pattern to improve efficiency of the power management of the electronic device.

Aggregated electronic device power management

Computing devices are configured to passively monitor network stacks and protocols for a respective computing device, transmit metadata and statistics gathered by the monitoring to a remote service, and utilize a crowd-sourced heuristic model responsively generated by the remote service to proactively predict connectivity issues and connect to a best available network media and access device for the network media. A computing device's operating system may monitor various networking protocols without the computing device engaging in constant network activities (e.g., video streaming). The statistics obtained from this passive monitoring can be utilized by the remote service using various machine learning techniques to predict when networks will subsequently fail. Profiles are developed and sorted within the model to be used by individual computing devices to seamlessly connect to access devices based on performance, as opposed to connecting to the access device previously utilized by the user.

Application of machine learning for building predictive models enabling smart fail over between different network media types

The techniques described herein monitor, store, and evaluate network information associated with an application to determine a connectivity option to use to communicate data. A connectivity option includes a network interface and a type of network connection. The determination is made based on power consumption information associated with available connectivity options. Consequently, a device on which the application is installed can better manage its power consumption associated with network communications.

Multi-interface power-aware networking

A supervisory control system provides power management in an electronic device by providing timeout periods for a hardware component to lower levels of the operating system such as a power management arbitrator and/or a hardware interface controller. The timeout periods are chosen based on usage patterns detected on the electronic device and based on user preference information. The power management arbitrator and/or hardware interface controller transition at least a portion of a hardware component to a lower-power state based on monitored activity information of the hardware component. The supervisory control system may further provide wakeup periods to the power management arbitrator and/or a hardware interface controller to determine whether the hardware component should be transitioned to a higher power state at the end of the wakeup period if the hardware component satisfies a transition condition.

Usage pattern based supervisory control of power management

Methods and systems that provide for a control path to enable at least one of the two data paths for a network device (e.g., a modem) are provided. An example method includes allowing a device to communicate data, over a network, using a shared control path, within the device, corresponding to a first bus protocol. The method may include using the shared control path, obtaining control information corresponding to a data path, within the device, for allowing the device to communicate data over the network using a network interface. The method may further include automatically determining data transfer capabilities for transmitting or receiving data using the network interface based on the control information, if the data transfer capabilities are determined to be capable of supporting a first data path. The first data path that uses a different protocol from the first bus protocol may be automatically established.

Control path to enable at least one of two data paths for a network device

Functionality is described herein for performing at least one network connectivity task on a client device with the aid of one or more assistant devices. In some implementations, a client device (such as a smartphone, desktop personal computing device, etc.) relies on an assistant device to assist it in updating its programs, including its driver programs. In other implementations, a client device relies on an assistant device in establishing a network connection with a network-accessible entity. Functionality is also described herein for performing at least one program execution task on a client device with the aid of one or more assistant devices. For instance, the client device may rely on the assistant device to assist it in executing a driver program. The driver program, in turn, enables the client device to interact with a peripheral device or some other component.

Offloading network connectivity and execution tasks to an assistant device

Systems, methods, apparatuses, and computer program products are described herein that enable a computing device to dynamically generate virtual private network (VPN) connection profiles to which policies can be mapped and then subsequently enforced. The VPN connection profiles are dynamically generated by a translation engine by obtaining parameters from payloads received from a plurality of different VPN gateways pursuant to establishing connections therewith in accordance with a respective plurality of different tunneling protocols, which may include both standard and proprietary tunneling protocols. The dynamically-generated VPN connection profiles are then provided to a connection manager, which enforces any policies associated with the VPN connection profiles when servicing requests for connections received from applications running on the computing device.

Growing community of inventors

Sammamish, WA, United States of America

Harish Srinivasan