Uber Technologies, Inc.

Jakob Holdgaard Thomsen

Rene Wenzel Schmidt

Alexey Pavlenko

Joakim Recht

Dron Rathore

Zura Kutsia

Jesper Lindstrøm Nielsen

Anders Bach Madsen

Timothy Smyth

Lasse Knud Damgaard

Vladimir Gavrilenko

Jeppe Welling Hansen

A method or a system for storing a file from a client device in multiple storage locations based on a placement policy with varied constraints for different copies of the file. The placement policy includes sets of policy constraints for different copies, specifying whether each copy should be uploaded synchronously or asynchronously. For each copy of the file, a specific query is generated based on its associated set of policy constraints. These queries are used to evaluate the attributes of various storage locations, selecting a candidate storage location for each copy. The file copies are then stored at their respective selected locations according to the synchronous or asynchronous upload requirement specified by their respective policy constraints.

Storage location selection according to query evaluation

An online system performs load shedding in case of system overloads. The system maps each request to a tier and a cohort. The tier is determined based on a type of request and the cohort is determined based on the user. Each tier includes multiple cohorts. The tiers and cohorts are ranked by priority. If the system determines that the system is overloaded, the system determines a threshold tier and a threshold cohort for load shedding. The threshold tier and threshold cohort indicate a threshold priority of requests that are processed. If the system determines that the unprocessed request has a priority below the threshold priority indicated by the threshold tier and the threshold cohort, the system rejecting the unprocessed request. The system executes unprocessed requests that are not rejected.

Priority-based load shedding for computing systems

A method for selecting storage locations for storage of data. According to a placement policy, a file is stored in a database comprising a plurality of storage locations. The method includes determining a query for a copy comprising policy constraints in the placement policy and storage location constraints. The method includes extracting one or more data attributes of the file. The method includes determining an attribute set for each storage location including the data attributes and the storage location attributes for each storage location. The method includes identifying a set of one or more candidate storage locations for storage of the copy of the file by evaluating the attribute set for each storage location against the query. The method includes selecting a candidate storage location from the set and providing the copy of the file to the selected candidate storage location for storage.

An online storage system receives a plurality of documents to be stored in a directory. The storage system stores document data from each document in a document database. The storage system generates an entry for each document in an entry table and indexes the documents stored in the directory. The storage system samples a subset of the plurality of documents assigned to the directory in a directory index to determine a sampled subset of the plurality of documents. The storage system indexes the sampled subset in a directory index. The storage system can receive a request, from a client device, to view the indexed documents in the directory. Responsive to the request, the storage system presents the indexed documents in the directory retrieved from the directory index.

Indexing documents in a nested hierarchy of directories

The data of an online system is distributed over a number of different physical servers and databases on those servers. The online system shards the data to increase system scalability and replicates the shards to increase data reliability. Features such as sharding and replication in turn increase data storage system complexity. Accordingly, in order to provide simplified data storage semantics to applications that use the online system, the online system provides access to data via asynchronous callback functions that are triggered once data has been written and is available for reading by client code. Thus, applications need only specify code to be used in response to data write events, and can thereafter interact with the online system data using a data-driven paradigm.

Asynchronous notifications for a datastore of a distributed system

A configuration deployment system stores configuration settings for a plurality of applications disposed at one or more computing hosts. The configuration settings are stored in a versioned repository and are organized in a hierarchical namespace structure. To deploy the configuration settings, a deployment service monitors the configuration settings at the configuration repository and identifies when a new configuration setting or a new version of an existing configuration setting is available in the configuration repository. The new configuration setting is transmitted to a computing device for use in an application. The application applies the changed configuration setting upon verifying that the change is compatible with the application.

Growing community of inventors

Risskov, Denmark

Jakob Holdgaard Thomsen