Microsoft Technology Licensing, LLC

Thomas Hejlsberg

Lars Hammer

Esben Nyhuus Kristoffersen

John Stairs

Gert Willem Robyns

Heinrich Hoffmann Clausen

Mostafa Balat

Stuart Glasson

Kevin Martin Racer

Jacob Winther Jespersen

Christopher Michael Rudolph

Michael Steven Hammond

Michael Riddersholm Petersen

Morten Jensen

Marc Hansen

Bogdan Sturzoiu

Chethan Thopaiah

Jesper Lau Falkebo

Cristian-Alexandru Chereches

Boaz Lev

Simon Kristiansen Ejsing

Mehmet Kerem Kiziltunc

Marko Perisic

Christian Heide Damm

Torben Wind Meyhoff

Morten Nielsen

Klaus Marius Hansen

Andreea-Bogdana Botez

Soren Grubov

Peter Sorensen

Andrei S Panko

Jens Jensen

Training prediction models and applying machine learning prediction to data is illustrated herein. A prediction instance comprising a set of data and metadata associated with the set of data identifying a prediction type is obtained. The data and metadata are used to determine an entity to train a prediction model using the prediction type. A trained prediction model is obtained from the entity. A notification system may be configured to react to monitor contextual information and apply the prediction. A workflow system may automatically perform a function in a workflow based on prediction.

Metadata-driven machine learning for systems

Workflow definitions are broken into a set of components. Each workflow component has an instance-specific workflow definition that it operates from. Different steps can be injected into different workflow instance-specific definitions, based upon a given state of each individual workflow instance.

Workflow generation and editing

Instrumenting code. One embodiment includes identifying a line of source code having an incomplete instrumentation. A unique identifier tag is obtained from an identifier source. The unique identifier tag is inserted into the instrumentation.

Tagged tracing, logging and performance measurements

Customizations, made at various levels in the customization channel, are saved as separate sets of customization deltas that are applied to a base computing system in order to generate a given solution. In a multi-tenant environment, groups of tenants access different solutions or customization deltas. The tenants are automatically grouped, based upon the solution or customization deltas that they access, and the deltas for those customizations are applied to the base system at runtime, for each group.

Multi-tenant, tenant-specific applications

An item to be processed is received and scanned. The scan identifies any functions or procedures in the item being processed, and extracts actual code representing those functions or procedures. Any invoking functions, which invoke the identified function, are, themselves, identified. Actual code for the invoking function is extracted as well. An output documentation file is generated that includes a name of the identified function or procedure, as well as actual code for the function or procedure, and actual code from at least one example of an invoking function.

Data extraction and generation tool

A cache manager application provides a data caching policy in a multiple tenant enterprise resource planning (ERP) system. The cache manager application manages multiple tenant caches in a single process. The application applies the caching policy. The caching policy optimizes system performance compared to local cache optimization. As a result, tenants with high cache consumption receive a larger portion of caching resources.

Data caching policy in multiple tenant enterprise resource planning system

Concepts and technologies are described herein for reducing locking during database transactions without using a cache. In accordance with the concepts and technologies disclosed herein, a server application generates a changeset in response to receiving an operation for modifying base data stored in a database. The changeset is stored as a set of tables that are stored at the database with the base data, and include a first table reflecting records in the base data that have been created and/or modified by the operations. The changeset further can include a second table reflecting records in the base data that have been deleted by the operations.

Reducing locking during database transactions

Modeled Types allow a developer to define attributes and/or code when defining a new type. When the Modeled Type is used, specific values may be assigned to the attributes. The consumer of the values of the attributes can be the type itself by executing the logic during application execution, or the consumer may be the runtime framework or the application using the values by querying the meta-data of the Modeled Type. Modeled Type Aliasing defines a Modeled Type based on another Modeled Type allowing both a new unique name but at the same time optionally supplying specific values to one or more of the attributes of the base Modeled Type. Context-aware Modeled Types provide a description of how to determine the value for the attribute.

Modeled types-attributes, aliases and context-awareness

A debugging tool is provided that is capable of recording code paths taken during a program's execution, and along the way capture the memory state at configurable points in time such as function exits or statements. The tool also records interactions with external systems like a database, and captures which statements resulted in interactions with these systems. An analysis of the output of the debugging tool enables a developer to understand which part of the code executed; what the value of relevant variables were at configurable points in time; and which code statements correlate to which interactions with external systems.

Execution capture and replay based debugging

Debugging operations on individual client sessions for a remotely executed shared application are enabled to be performed as the client requests are processed on the executing server without disrupting execution of other client sessions. A remote debugging client may connect to a debugging engine executed on the server allowing the debugging client to view source code, set breakpoints, view client connections, and receive callbacks or notifications when a breakpoint is hit by the client session being debugged. The debugging client may also control execution by stepping through client code enabling debugging of multiple clients simultaneously.

Client session based debugging

A checkpoint system is provided to quickly identify differences or deltas between a baseline version and a modified version of a database and to quickly restore the database to the baseline version using the deltas. The checkpoint system identifies the deltas between versions of the database by a rapid comparison of the rows of each table within a baseline version and a modified version of the database. The checkpoint system can use the rapid comparison when testing to identify sets of deltas resulting from performing an old procedure on a test database and a new procedure on the test database. The rapid comparison can be applied to these sets of deltas to identify differences resulting from the new procedure. The deltas can also be used to rapidly restore a database from a modified version to a baseline version.

Generation of database deltas and restoration

Embodiments of the present invention pertain to implementation of textual names as the actual identifiers of elements of models. A structured system provided for constructing the textual names so as to be unique in a specific context. In one embodiment, namespaces are added to a name as a mechanism for ensuring uniqueness across multiple development sources. In one embodiment, upon deployment to a runtime environment, alternate names are assigned to support efficient processing.

Name-based identifiers of model elements

A system and method directed to capturing the structure of data models using entity patterns is provided wherein an entity pattern is an abstraction, for example in the MICROSOFT BUSINESS FRAMEWORK&#xae; (MBF), that surfaces in the model editor of the framework at the root level and consists of one or more entity roles that together define a structural pattern of a model. The entity pattern may be considered as an &#xfc;ber-model, or a model template for a model. Prescriptive rules are added to the design of entities, required properties and relations to other entities (entity roles). Applications of the entity patterns are validated at design time. The runtime framework supports any applied pattern and invokes the pattern specific code. It is a general advantage that the runtime operation is model driven and interprets the model data at runtime. The application and usage of a pattern is part of the model of the application/software program.

Growing community of inventors

Horsholm, Denmark

Thomas Hejlsberg