Abstract: One embodiment of the present invention sets forth a technique for extracting data from a document. The technique includes determining, via execution of one or more machine learning models, a first set of bounding boxes for a first set of values associated with a first key within the document and a second set of bounding boxes for a second set of values associated with a second key within the document. The technique also includes generating a first set of mappings between a list of items in the document and the first set of bounding boxes and a second set of mappings between the first and second sets of bounding boxes based on locations of the bounding boxes. The technique further includes determining, for a given item, one or more associated bounding boxes in the first and second sets of bounding boxes based on the mappings.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for utilizing asynchronous double-writes to migrate data from a primary database to a secondary database. In particular, in one or more embodiments, the disclosed systems, upon receiving a work unit, write data to a primary database and generate a trigger to asynchronously write, by a sync record service, the data to a secondary database. In some embodiments, the disclosed systems reverse the direction of the asynchronous double-writes so that, upon receiving additional work units, the disclosed systems write, by the sync record service, additional data to the secondary database and generate a trigger to asynchronously write additional data to the primary database.

Abstract: A system and method for determining whether an entity is acquiring duplicate data using a machine learning model. The system includes a repository having a plurality of databases that store data and information in a format accessible to users, and a back-end server operatively coupled to the repository and being responsive to the data and information from all of the databases. The back-end server includes a processor for processing the data and information, a communications interface communicatively coupled to the processor, and a memory device storing data and executable code. The code causes the processor to collect data and information from the databases, store the collected data and information in the memory device, process the stored data and information through the machine learning model to determine whether the entity is acquiring duplicate data, and transmit a communication on the interface identifying that the entity is acquiring duplicate data.

Abstract: Time-series columnar-based information is received and indexed in a compute infrastructure for cloud-based object storage. The approach leverages a file format that enables highly-performant search and retrieval of the data stored in the cloud. In operation, an indexer receives the time-series information, indexes that information according to the file format, and forwards the indexed information for storage to the object store, where it is stored as a set of time-based partitions. A partition comprises a set of files, namely, a manifest file, a data file, and an index file. These files are structured as a compact instance of a set of raw unstructured data that comprises the partition. Highly-performant information retrieval is enabled in response to a time-bounded query, because operations at a query peer (with respect to one or more partitions) are carried out in real-time during query processing and without requiring retrieval of the data file as a whole.

Abstract: Some embodiments provide a program that receives a selection of a data object relationship definition. The data object relationship definition specifies a plurality of data objects managed by a plurality of applications and a set of relationships between data objects in the plurality of data objects. The program sends each application in a first subset of the plurality of applications a request for instances of data objects in the plurality of data objects managed by the application. The program receives, from each application in the first subset of the plurality of applications, a list of a set of candidate instances of data objects. Based on the list of the sets of candidate instances of data objects, the program further determines a set of groups of instances of data objects. The program deletes a group of instances of data objects in the set of groups of instances of data objects.

Abstract: Output responses corresponding to a plurality of matching user requests are combined in a transaction response combiner (TRC). Each output response includes response details. The TRC receives an output response for each of the user requests. Each output response is tagged with an identifier that corresponds with the respective user request. It is detected when an output response received at the TRC has one or more corresponding output responses by searching in the memory for output responses that have the same tag. Rules of a rules engine are applied to combine the output responses received at the TRC for each of the matching user requests into a final output response. The rules of the rules engine are applied to the corresponding output responses with the same tag, and to the response details for the output response that shares the same detected tag. The final output response is outputted by the TRC.

Abstract: Systems and methods are described herein relating to ingesting data objects from various sources and associating them in a hierarchy with a profile. In some aspects, a data object, including a set of fields and indicating a data object type, may be obtained for associating with a profile maintained by a computing resource service provider. The data object may be converted into a standardized format based on the data object type indicated in the data object using an object mapping associated with the data object. A key associated with the data object and indicated in the object mapping may be selected and used to search an index for a related profile. Upon finding a profile with the matching key, the profile may be updated to include an association of the standardized data object, such as in a hierarchy.

Abstract: Schema based data buffer and processing on a client device. A client device may obtain a schema for a set of data. The set of data is stored on a server device. The client device may also obtain a copy of the set of data from a server database. The copy of the set of data is stored on a client device. The client device may further obtain a set of mutations to one or more portions of the set of data. The client device may further modify the set of mutations based on the schema for the set of data. The client device may further apply the modified set of mutations to the copy of the set of data that is stored on the client device.

Abstract: Disclosed are an electronic device, a contents searching system, and a searching method. The method for searching contents of an electronic device includes: creating a keyword vector by extracting a keyword for searching; obtaining a searching contents candidate group based on a first similarity between contents indexing data created by a preset natural language processing-based search model and the created keyword vector; obtaining a second similarity between the created keyword vector and vector data included in the searching contents candidate group based on a trained artificial intelligence model; and aligning and providing contents included in the searching contents candidate group based on the obtained first similarity and the second similarity.

Abstract: A method includes obtaining a plurality of data partitions of a data set for storage in a database system. The method further includes determining a first data storage coding scheme for a first partition of the plurality of data partitions, where the first data storage coding scheme includes first encoding parameters regarding encoding the first partition into first data segments and first parity segments. The method further includes determining a second data storage coding scheme for a second partition of the plurality of data partitions. The method further includes dividing the first partition to produce a first number of first raw data segments, where the first number is based on the first data storage coding scheme. The method further includes dividing the second partition to produce a second number of second raw data segments, where the second number is based on the second data storage coding scheme.

Abstract: Techniques for data management are described. A file may be saved that includes a representation of an exported database, where the representation may be based on a live database being exported from a computing system via an export operation that spans a time period. A determination that one or more transactions were committed to the live database during the time period may be made. Based on the determination, a log indicating the one or more transactions may be obtained from the computing system. Based on the log, a determination of whether the exported database captured the one or more transactions may be made. Based on the determination, the file may be updated to represent an updated database that reflects an updated version of the live database, the updated version of the live database reflection application of the one or more transactions to the exported database.

Abstract: Techniques are disclosed relating to updating schema information for multi-tenant databases. In some embodiments, a multi-tenant database system stores a standard schema for a database object, where the standard schema is shared by multiple tenants and a tenant-specific schema for the database object. The tenant-specific schema may include one or more tenant-specific custom elements for the database object. The system may enforce a naming scheme in which tenant-specific elements include a predetermined pattern that is not included in object elements of the standard schema. The system may, in response to detecting an update of the standard schema, automatically propagate the update to the tenant schema, while the tenant is online.

Abstract: In some implementations, the current subject matter relates to deployment of database objects. In some implementations, there is provided a method that includes receiving a container including a plurality of database objects; detecting whether the plurality of database objects in the container includes a linking object pointing to an object at another container; in response to detecting the linking object, generating an external object table to store metadata about the obj ect at the other container; obtain metadata about the external object to store in the external object table; and store the obtained metadata in the external object table. Related systems, methods, and articles of manufacture are also disclosed.

Abstract: A first stack running on a processor receives the transaction data, reference data, and context data. The reference data is independent of the transaction and of a user. The context data is associated with the user but is independent of the transaction. The first stack strips the transaction of derivable data to obtain stripped data. The derivable data includes data that can be derived from the stripped data, the context data, and the reference data. The derivable data can stream the stripped data to a global database available and redundant across multiple geographical regions. After the first stack fails, a second stack can resume the transaction by retrieving the stripped data from the global database, and retrieving the context data, and the reference data. The second stack can recreate the transaction data based on the stripped data, the context data, and the reference data, and can resume the transaction.

Abstract: A method includes creating, by a first provider, a first listing referencing first shared data and comprising first access controls, wherein access to the first shared data by a second provider is filtered based on the first access controls, creating, by the second provider, a second listing referencing second shared data and the first shared data filtered based on the first access controls, and adding the second listing to a catalog in a data exchange, the catalog comprising metadata describing the second shared data.

Abstract: Disclosed in some examples, are methods, systems, and machine readable mediums which provide an event processing system and indicator engine that provide indicators to one or more of the channel systems that pertain to a status of an issue experienced by a customer. The indicators are created in part based upon updates to the issue entered into a customer issue system and when displayed in a channel system, the indicators provide an indication that there is an update on the particular issue affecting the customer.

Abstract: An apparatus comprises a processing device configured to monitor logs of a database system, to analyze the logs to collect ongoing transaction information for one or more applications utilizing the database system, and to maintain the ongoing transaction information for each of the one or more applications in a corresponding one of a set of one or more queues of a data store external to the database system. The processing device is also configured to detect one or more designated events affecting operation of the database system and, responsive to detecting at least one of the one or more designated events, to process one or more unsaved transactions of the database system utilizing the information stored in the set of one or more queues of the data store external to the database system.

Abstract: In one aspect, a method of detecting database anomalies, includes reading historical data in a destination database at an end of a data pipeline, determining bounds including an upper bound and a lower bound based on the read historical data, reading current data for a first specified time period in the destination database, responsive to determining the upper or the lower bound is exceeded, determining database transactions that caused the exceeding, and transmitting alerts to owners of the database transactions.

Abstract: Apparatuses and methods are provided and applied to graphically represent the history of changes to a datastore maintaining a plurality of datastore records, each datastore record of the plurality of datastore records comprising at least one sector, as a directed acyclic graph to facilitate making, undoing, and redoing independent changes without regard to any explicit branching structure. Current records store content, a log directed acyclic graph relates various changes, and a current change record mapping relates each sector of each datastore record to a respective current change record. Changes by users requiring moderation are independently overlayed upon changes visible to additional users, with changes being approved and conflicting changes being resolved by administrating users.

Abstract: Systems and methods for synthesizing cloud snapshots of production data. A client system includes data and snapshots may be taken of that data. These snapshots are transferred to the cloud and hydrated to block devices that are similar to the devices on which the production data exists. Snapshots are then taken of the block devices. The cloud snapshots are the same as and correspond to the locally generated snapshots.