Microsoft Technology Licensing, LLC

James Wiliamson Mickens

Jeremy Eric Elson

Edmund Bernard Nightingale

Amar Phanishayee

Vijaychidambaram Velayudhan Pillai

John R Douceur

Thomas Moscibroda

David Chiyuan Chu

Li Zhuang

Debmalya Panigrahi

Kaisen Lin

Bin Fan

Asim Kadav

Richard F Rashid

Jonathan Ryan Howell

Jacob Rubin Lorch

Meredith June Morris

Jaime Brooks Teevan

Saleema Amin Amershi

Ali Raza Farasat

Jon Howell

Osama Khan

While connected to cloud storage, a computing device writes data and metadata to the cloud storage, indicates success of the write to an application of the computing device, and, after indicating success to the application, writes the data and metadata to local storage of the computing device. The data and metadata may be written to different areas of the local storage. The computing device may also determine that it has recovered from a crash or has connected to the cloud storage after operating disconnected and reconcile the local storage with the cloud storage. The reconciliation may be based at least on a comparison of the metadata stored in the area of the local storage with metadata received from the cloud storage. The cloud storage may store each item of data contiguously with its metadata as an expanded block.

Disconnected operation for systems utilizing cloud storage

Described herein are various technologies pertaining to a computing system architecture that facilitates construction of virtual motherboards and virtual storage devices. A computing system includes a plurality of computing nodes and a plurality of storage nodes, where the computing nodes are disaggregated from the storage nodes. The computing nodes include respective system on chip modules. The computing nodes and the storage nodes are communicatively coupled by way of a full bisection bandwidth network, where each storage node is allocated network bandwidth that at least matches the maximum input/output speed of the storage node. Responsive to receipt of a client request, computing nodes are allocated to the client and exposed to an application of the client as a motherboard. Likewise, storage nodes are allocated to the client and exposed to the application of the client as a larger storage device.

Constructing virtual motherboards and virtual storage devices

This document relates to data storage techniques. One example can buffer write commands and cause the write commands to be committed to storage in flush epoch order. Another example can maintain a persistent log of write commands that are arranged in the persistent log in flush epoch order. Both examples may provide a prefix consistent state in the event of a crash.

Block storage by decoupling ordering from durability

In the new architecture, a client that desires access to a piece of hardware sends HTTP requests to a device server. The server receives a HTTP requests, accesses a device on behalf of the client, and send the device's response back to the client in the form of an HTTP response. This architecture has three primary advantages. First, it offers a clean interface for clients where all devices are accessed as if they are web servers. Second, it helps make devices more secure whereby clients can be prevented from accessing hardware directly, and all device access is forced through the narrow HTTP access protocol and checked for compliance with a desired security policy. Third, since HTTP allows local and remote servers to be contacted, the proposed architecture makes it easy for clients to communicate with devices that are not physically co-resident with the client but which are accessible via a network connection.

Accessing hardware devices using web server abstractions

The subject disclosure is directed towards one or more parallel storage components for parallelizing block-level input/output associated with remote file data. Based upon a mapping scheme, the file data is partitioned into a plurality of blocks in which each may be equal in size. A translator component of the parallel storage may determine a mapping between the plurality of blocks and a plurality of storage nodes such that at least a portion of the plurality of blocks is accessible in parallel. Such a mapping, for example, may place each block in a different storage node allowing the plurality of blocks to be retrieved simultaneously and in its entirety.

Block-level access to parallel storage

A state migration module (SMM) is described herein which seamlessly, efficiently, and correctly transfers application state between user computing devices, or between different interactions with the application on the same device that occur at different respective times. The SMM operates by capturing the current state of an application relative to the operation of a high-level execution platform (such as a browser) on which the application runs, rather than capturing the overall low-level state of the computing that runs the application. The SMM can include additional provisions designed to improve the efficiency of state migration, such as asynchronous migration, incremental delta-based migration, etc.

Efficient migration of client-side web state

A baseline set of graph information associated with a predefined execution environment of a program may be obtained, the baseline set associated with a baseline time interval. The predefined execution environment includes predefined environment objects, at least a portion of which are dynamically mutatable. A second set of graph information associated with the predefined execution environment may be obtained, the second set associated with a second time interval that is later in time than the baseline time interval. The baseline set and the second set may be compared to detect virtualization activity associated with a dynamic runtime.

Virtualization detection

Web page loading time may be decreased with the use of a page-maintained storage cache on a client device. In various implementations, a server may receive a request for a web page from a web browser that is on the client device. The server may provide a chunk reassembler to the client device that determines whether any chunks in the web page are already stored on the client device. Subsequently, the server may send one or more chunk of the web page to the client device when the chunk reassembler determines that the one or more chunks of the web page are missing from the client device. The chunk reassembler may then instruct the web browser to assemble the one or more received chunks with the one or more chunks that are stored on the client device into the web page for presentation on the client device.

Page loading optimization using page-maintained cache

This patent application pertains to capturing web-based scenarios. One example detects execution of a web application. This example also automatically captures non-deterministic events of the execution in a manner that is transparent to a user of the web application.

Capturing web-based scenarios

The described implementations relate to collaborative speed determination in distributed systems. One method estimates upload speeds of a plurality of hosts. The method sends a first bandwidth probe from a first host to a second host. The method receives, at the first host, a second bandwidth probe sent from the second host. The method determines which of the first and second bandwidth probes was relatively slower. The method designates either of the first or second hosts that sent the relatively slower bandwidth probe as a loser and takes a speed of the bandwidth probe of the loser as a first estimate of the loser's upload speed.

Collaborative speed determination in distributed systems

The claimed subject matter provides a system and/or a method that facilitates generating a personalized query result for a specific user. An interface can receive at least one of a portion of a text query to be searched or a portion of personalized content related to a user that submits the portion of the text query. A personalization component can combine the portion of personalized content related to the user with a portion of personalized content related to one or more disparate users to create group personalized content, wherein the group personalized content is compared with the portion of the text query to identify a relationship there between to generate a personalized query result in accordance with the relationship.

Using related users data to enhance web search

A speculative web browser engine may enable providing transmission of content between a server and a client prior to a user-initiated request for the content hidden in imperative code (event handlers), which may reduce user-perceived latency when the user initiates the imperative code. In some aspects, a speculative browser state may be created from an actual browser state and used to run the event handlers. The event handlers may be modified to direct actions of the event handler to update the speculative browser state. Speculative content may be transmitted between the server and the client in response to an execution of the modified code. The speculative content may be stored in a cache and made readily available for use when the user initiates the event handler and finds that the desired content has already been fetched.

Growing community of inventors

Bellevue, WA, United States of America

James Wiliamson Mickens