Microsoft Technology Licensing, LLC

Kfir Ami-ad

Eric B Fox

Roshane Silva

Gokhan Uluderya

Tyler Furtwangler

Bijul Soni

Sanjay Rama

Arye Gittelman

Sterling John Crockett

Harish Jayanti

Srinivasa Raghavan Santhanam

Derrick Yeqiang Quan

Gaurav Bindlish

Dillon Nugent

Mitali Prakash Shenoy

Payel Hazra

A system for creating a target audience group includes receiving a request from a user interface screen, to create a target audience group from among users of an organization associated with a cloud computing environment, where the request including selected values for a plurality of categories of data associated with the users of the organization. Upon receiving the request; the method creates an audience targeting definition for the target audience group that includes the plurality of categories of data and their selected values and stores the audience targeting definition to a data store. Next, one or more data sources of the cloud computing environment are queried to identify users for which values for the categories of data corresponds with the selected values and a list of those users is stored in a different data store. The list of the users is then made available to a plurality of applications of the cloud computing environment for communicating with or presenting an experience to the members of the target audience group.

Unified audience targeting system

Loading of web application pages and dependent files are optimized both in terms of load order and selective authentication. A baseline file (also referred to as the wire frame page) may include main page framework and most dependent files. Files referenced in the baseline file may be organized such that they load in an optimized fashion. The baseline file and the dependent files referenced therein may be loaded anonymously without waiting for the completion of the authentication process. The secure content may be the only authenticated request made, thus allowing most of the files to load in parallel to the longer authenticated request.

Web page load time reduction by optimized authentication

Requests to a server farm are managed in an application intelligent manner based on server health and client information. A request management (RM) module is configured to make smart routing decisions, based on routing rules that relate the nature of requests to a dynamic topology of a server infrastructure. The RM may use its knowledge of server applications, servers, and clients to deny potentially harmful requests from entering the service, route good requests to the suitable machines, and maintain a history of the decisions and consequences automatic and/or manual optimization.

Application intelligent request management based on server health and client information

Server health and client information are exchanged through headers for request management. Headers in standardized or proprietary protocol communication between servers and a request management module and/or clients and the request management module may be used to exchange server health and client information. The exchanged server health and/or client information may be employed in throttling, routing, and/or load balancing the incoming requests. Rules specified by the client through the header exchange may also be used for throttling, routing, and/or load balancing decisions.

Exchange of server health and client information through headers for request management

Growing community of inventors

Redmond, WA, United States of America

Kfir Ami-ad