Amazon Technologies, Inc.

Michael James McInerny

Brandon William Porter

Harsha Ramalingam

Blair Livingstone Hotchkies

Kyle Bradley Peterson

Mark Evans Brighton

Sevag Demirjian

Anton Vladilenovich Goldberg

Daniel Dwight Longley

Almann T Goo

Robert Stanley Bailes

Daniel Wade Hitchcock

Luan Khai Nguyen

Vishal Parakh

Leon Robert Warman

Bhavnish H Lathia

Phivos Costas Aristides

Todd Vaughn Jonker

Paul Derek DeMarco

Aaron Ben Fernandes

Karthik Tamilmani

Jane Mooney

Darryl Havens

Akash Kayal

Donald Arthur Parsons

Robert Karl Beckmeyer

Techniques described herein include systems and methods for throttling requests for content to reduce stress on a check out pipeline associated with an electronic marketplace thereby avoiding overstressing a server to the point of no longer processing requests from users. In embodiments, first information may be maintained that identifies a ranking for a plurality of items based on a score. A predicted velocity of content requests about the plurality of items may be maintained and second information about an actual velocity of content requests about the plurality of items may be received. In response to a request for content, a portion of items may be identified based on the scores associated with said portion and partition the portion into a number of groups or partitions based on the predicted velocity and the second information. A data object that comprises the portion of items associated with a partition may be generated.

Method and system for dynamic traffic shaping of deals to reduce server stress

Methods, systems, and computer-readable media for implementing automated services capacity modeling using call tracing are disclosed. A plurality of demand drivers are determined based on trace data for service interactions between services in a service-oriented system. The demand drivers are determined to drive a generation of service calls to a particular service. A total call volume is determined to the particular service based on the external demand drivers. An optimized quantity of computing resources to provide the particular service is determined based on the total call volume.

Automated services capacity modeling

A method and system are provided that provide an avatar based customer service experience with a human support agent. The methods and systems receive, from a customer computing (CC) device, a request for assistance fulfillable by one of a plurality of support agents. The methods and systems launch an avatar-based exchange that includes receiving customer issue definition (CID) information from the CC device regarding the request for assistance, defining a virtual character to be presented on the CC device; and providing pre-recorded support (PRS) content based on the CID information. The PRS content is presented in combination with animation of the virtual character. The methods and systems select a support agent, and transition a basis for the avatar-based exchange from the PRS content to support agent content such that the support agent communicates with the customer through the virtual character animated on the CC device.

Avatar based customer service interface with human support agent

Features are disclosed for creating composite content pages comprising items from various network-enabled content pages. Selection of items to include in a composite page is not limited to items formatted or exposed by content providers in a specific format, such as RSS. A network computing component can retrieve, process, and format content from any content provider, and then generate a composite page containing the selected content items in a format specified by a user or automatically determined by the network computing component. Additionally, the content items can be processed to reduce or prevent the likelihood of security issues that may arise when compiling content from different content providers into a single composite page.

Remote generation of composite content pages

Current CAPTCHA tests are designed to be difficult for a bot and simple for a human-user to answer; however, as artificial intelligence improves, bots are more capable of using techniques such as optical character recognition to resolve current CAPTCHAs in similar manners as human-users. By providing a CAPTCHA challenge from a library or set of challenges that are designed in a manner that causes or likely causes a human-user to trivially get the answer to the challenge wrong, helps to confirm that a user is a human-user, as a bot would answer the challenge correctly.

Turing test via failure

Methods, systems, and computer-readable media for implementing global optimization of a service-oriented system are disclosed. Trace data is collected for a plurality of service interactions between services in a service-oriented system. Respective costs are determined for a plurality of configuration options in the service-oriented system. An optimized configuration for the service-oriented system is determined based on the respective costs and the trace data. The optimized configuration comprises a selection of one or more of the configuration options. The optimized configuration is deployed to the service-oriented system.

Global optimization of a service-oriented system

Described are techniques for determining the effect of a change in how data is handled by a first device on other devices within a system. At a first device, sending, receiving, or processing of data may be prevented for a length of time. Metric data may be determined from other devices during the length of time and during one or more time periods before or after the length of time. The metric data may indicate the effect of the handling of data by a first device on the one or more other devices.

System for modeling distributed systems

Embodiments provide techniques for identifying unique data elements in data record, removing the unique data elements from the record, storing the record, and employing the removed, unique data as a key in an index into the data storage. When the record is retrieved based on the key in the index, the removed unique data elements is restored. Removal of the unique data elements from the records may enable more efficient compression of the records using compression algorithms that employ an external reference dictionary. In some embodiments, a combination of multiple, removed data elements may be employed as a key. In some embodiments, the removal and use of the unique data as an index key proceeds in a manner that is transparent to the data consuming user, process, or device.

Indexing stored documents based on removed unique values

Methods, systems, and computer-readable media for implementing service-oriented system optimization using partial service relocation are disclosed. An optimized configuration is determined for the service-oriented system based on performance data. The optimized configuration improves a performance metric in the service-oriented system. A partial service is automatically generated based on an original service in a service-oriented system. The partial service includes a first set of program code from the original service and excludes a second set of program code from the original service. The first set of program code is included in the partial service based on its frequency of use. One or more instances of the partial service are deployed to the service-oriented system based on the optimized configuration.

Service-oriented system optimization using partial service relocation

This disclosure is directed to systems and processes of capturing a user's actions, computing actions, and/or other data, and then publishing relevant portions of the user's actions, computing actions, and/or other data to other users in a user community. The systems and processes provide techniques to minimize time and effort to create publications for the user community, while allowing the user to select relevant content and possibly annotate the content. Publication may be directed to other users that are likely to perform a similar computing action and/or to shared repositories.

System and processes to capture, edit, and publish problem solving techniques

Mechanisms are disclosed for mount and non-mount based access to filesystems contained in virtual disks in virtual machine images. Access can be had to the filesystems even when the virtual machine images are not executing. Access to the filesystems can also be provided to local and remote client applications. Access can also be provided to snapshots of the virtual machine images at various points in time.

Local and remote access to virtual machine image filesystems

Systems and methods are provided for efficiently indexing archived objects in an archive data store to allow for efficient and quick access to the archived objects. The systems and methods provided also allow alleviate the strain on a live data store and enable a requester wishing to retrieve updates to do so without knowledge of the location of the stored update on a live or archive data store.

Indexing data updates associated with an electronic catalog system

Various examples are directed to systems and methods that measure the gaze direction of a user. For example, gaze direction may be utilized to generate an eye-equivalent representation for the user, where the eye-equivalent representation depicts eyes directed according to the gaze direction.

Gaze direction tracking

Current CAPTCHA tests are designed to be difficult for a bot and simple for a human-user to answer; however, as artificial intelligence improves, bots are more capable of using techniques such as optical character recognition to resolve current CAPTCHAs in similar manners as human-users. By maintaining a library of security tests and/or questions based on products purchased by the human-user, and using those questions as a CAPTCHA challenge in order to determine if the user is a human-user on how the user responds, CAPTCHA challenges may be directed toward determining a human-user based on responses exhibiting humanity.

Growing community of inventors

Seattle, WA, United States of America

Michael James McInerny