Amazon Technologies, Inc.

Snap Inc.

David Andrew Killian

Samuel John Young

Peter F Hill

Brett Richard Taylor

Ameet Nirmal Vaswani

Ranganath Atreya

Ivan King Yu Sham

James Daniel Baird

Saral Jain

Dhruva Lakshmana Rao Batni

Dmitry Sokolowski

Rohit Krishna Kumar

Nikhil Dinkar Joshi

Ovais Ahmad Khan

Tittu Jose

Jonathan A Jenkins

Benjamin Ralph Hollis

David Craig Yanacek

Matthew James Wren

Bradley Baron

Eric D Crahen

Justin Jon Derby

James Alan Umstot

Krishnan Narayanan

Arunachalam Thirupathi

Sajeeva Lakmal Bandara Pallemulle

Zubair Ahmed Khan

Amit Saurav

Spoorthi Ravi

Systems, methods, and computer readable media for a live data migration system where the methods include accessing a command to perform a live data migration of a source datastore to a target datastore, accessing a delete operation, the delete operation indicating first data to be deleted from a datastore, and determining the datastore is stored in a source datastore. The methods further includes performing a two-phase commit delete of the first data from the source datastore and the target datastore, accessing a write operation, the write operation indicating second data, and performing a two-phase commit write of the second data to the source datastore and to the target datastore. The methods further include copying the source datastore to the target datastore, accessing a read operation, the read operation indicating third data to be read from the datastore, and performing the read operation of the third data from the target datastore.

Live data migration system

A decentralized two-phase commit in a datastore or database includes determining pre-validating a transaction, writing first metadata to a leader partition, the first metadata including a state indicating preparing and mutations of the transaction, and writing second metadata of a follower partition to indicate the follower partition is locked by the leader partition. The decentralized two-phase commit further includes updating the state of the first metadata to indicate committed and performing the transaction on the leader partition, performing the transaction on the follower partition and unlocking the follower partition, and updating the first metadata to remove the mutations of the transaction and updating the state of the first metadata to indicate the transaction is completed. The decentralized two-phase commit may further include selecting a leader partition key and updating the state of the first metadata to indicate committed unapplied.

Decentralized two-phase commit

Systems and methods are provided for data synchronization for client application generated data. Content may be generated by a client application executed by a creator client device and sent to a server system that may distribute the content to one or more additional client devices of recipients of the content. The server system can push content to the recipient client devices without receiving a request for the content from the recipient client devices. The content generated by the creator client device may be stored by each of the creator client device, by the server system, and by the one or more recipient client devices. The server system may also generate version data related to respective versions of content that may be consumed by a client application. The version data may be used by client devices to determine versions of content that may be missing from the client devices.

Data synchronization for content consumed via a client application

Systems and methods are provided for synchronizing data. The systems and methods include operations for: storing a synchronization entry for a messaging application feature, the synchronization entry comprising a last update timestamp associated with a first update to content of the messaging application feature received from a first source; receiving a second update to the content of the messaging application feature from the first source; determining that the second update was received within a write window of the last update timestamp; in response to determining that the second update was received within the write window of the last update timestamp, preventing updating the last update timestamp; and sending the first update and the second update to a client device in response to receiving a synchronization request from the client device based on the last update timestamp.

Windowed writes

The load level on a server system is regulated by determining time-to-live (TTL) values to provide to requesting devices that request a content resource from the server system, thereby affecting the frequency of subsequent requests. This dynamic determination of TTL values may provide resilience to system load, for example by using longer TTL values when the system is under greater load to reduce the rate at which subsequent requests are received. A dynamic TTL service may calculate a TTL value based on one or more factors, such as overall system load, resource load, hardware load, and/or software load.

Load regulation using dynamically determined time to live values

Features are disclosed for sharing network connections among client devices. A client device may connect to an intermediary system and request content from content provider. The intermediary system may, in addition to opening a connection with the content provider and receiving the content on behalf of the client, create a request profile for the client device regarding likely future content requests. The request profile may be used to assign the client device to network comparing components that have available open network connections to one or more content providers from which the client device is likely to request content. In some cases, a gateway network computing component may route content requests to other network computing components which have available open network connections to content providers hosting requested content. Additionally, the number of active network computing components may be adjusted to increase connection sharing and reuse opportunities.

Sharing network connections to content sources

Minimizing bandwidth consumption in rendering dynamic content, such as video streaming. Traditional streaming video applications utilize lossy protocols that often result in degraded user experiences. A dynamic content encoder/decoder allows a server-based browser application to generate vector-based rendering instructions for processing by a client-based browser application. Additionally, as the dynamic content changes the content to be displayed, a server-based browser application generates updated vector-based instructions and encoded spatial and temporal differences to minimize the data transmitted to the client-based browser.

Vector-based encoding for content rendering

Time to live ('TTL') values are determined based on one or more factors. The TTL values may be included in responses to requests for resources, thereby affecting the frequency of subsequent requests. This dynamic determination of TTL values may provide resilience to system load, for example by using longer TTL values when the system is under greater load in order to reduce the rate at which subsequent requests are received. A dynamic TTL service may calculate a TTL value based on one or more factors, such as overall system load, resource load, hardware load, and/or software load. In various embodiments, a dynamic TTL service may act natively within a service, within a system framework, as a proxy, as a cluster, and/or as a broker.

Features are disclosed for analyzing requests for network accessible content, including but not limited to web pages, to determine which topics and other characteristics are popular or are gaining in popularity ('trending'). Content items or sources may be profiled to determine characteristics that two or more content items or sources may have in common. Content requests from multiple client devices may be tracked and analyzed to determine the trending or popular characteristics. Data feeds or reports regarding the summarized content requests may be generated and distributed to content servers and other entities. The data feeds may be used to tailor content, such as by highlighting or featuring content associated with the most-requested content characteristics, or utilizing demographic data to tailor content for different users.

Determining popular and trending content characteristics

Minimizing bandwidth consumption in rendering dynamic content, such as video streaming. Traditional streaming video applications utilize lossy protocols that often result in degraded user experiences. A dynamic content encoder/decoder allows a server based browser application to generate vector based rendering instructions for processing by a client based browser application. Additionally, as the dynamic content changes the content to be displayed, a server based browser application generates updated vector based instructions and encoded spatial and temporal differences to minimize the data transmitted to the client based browser.

Technologies for performing hybrid rendering of web pages are described in which some web page resources are rendered by a server to create processed layers (e.g., composited layers) and other web page resources are left in a format that can be locally rendered at the client. For example, a server can split web page resources into a first part and a second part. The first part can be rendered down to processed layers and placeholder layers can be created representing the second part. The processed layers, the placeholder layers, and the second part of the web page resources can be sent to the client. The client can render the second part of the web page resources and merge them with the processed layers. The client can display the web page using the merged layers. The client can update the web page by locally re-rendering the second part of the web page resources.

Hybrid rendering of a web page

Features are disclosed for analyzing requests for network accessible content, including but not limited to web pages, to determine which topics and other characteristics are popular or are gaining in popularity ('trending'). Content items or sources may be profiled to determine characteristics that two or more content items or sources may have in common. Content requests from multiple client devices may be tracked and analyzed to determine the trending or popular characteristics. Data feeds or reports regarding the summarized content requests may be generated and distributed to content servers, advertisers, and other entities. The data feeds may be used to tailor content, such as by highlighting or featuring content associated with the most-requested content characteristics, or utilizing demographic data to tailor content for different users.

Features are disclosed for determining a sequence of content, including but not limited to web pages, that a user of a client device is likely to request or otherwise find interesting based on previous content requests (of that user and/or other users), and to making the sequence of content available for viewing on the client device. The identified pattern may later be used as a basis to prefetch the content, and to make the sequence of prefetched content available on the client device as a 'suggested browsing session.' The suggested browsing session may include a sequence of content pages typically requested by the user, and/or may include linked pages and recommendations that the user is otherwise likely to find interesting.

Growing community of inventors

Seattle, WA, United States of America

David Andrew Killian