Abstract: In an embodiment, a plurality of ballots (1) for a plurality of candidates, (2) generated by a plurality of search engines, and (3) for a job description are received. Each ballot from the plurality of ballots is generated by a search engine from the plurality of search engines different than remaining search engines from the plurality of search engines. A mathematical representation that indicates, for each candidate from the plurality of candidates, how many other candidates from the plurality of candidates that candidate was ranked higher than in the plurality of ballots is generated. A final ballot ranking the plurality of candidates is generated using a trained statistical model and based on the mathematical representation. A candidate from the plurality of candidates for the job description is identified based on the final ballot.

Abstract: The present disclosure provides systems for encoding transforms based on intent to be useable by developers. The system receives input datasets, generates output datasets, and identifies characteristics of the input or output datasets. Using these characteristics, the system can encode transforms associated with the datasets based on intent. The system stores and/or associates the intent-encoded transforms in data profiles, so that the system enables the transforms to be searched, recommended, and/or combined.

Abstract: A data management system may support techniques for immutable storage of snapshot data entities, which may each include data corresponding to one or more snapshots, in a cloud environment. The data management system may determine respective retention periods and respective immutability periods for the snapshot data entities. The data management system may extend the respective immutability period for a first snapshot data entity based on the respective retention period for the first snapshot data entity being greater than or equal to a threshold duration. Additionally or alternatively, the data management system may maintain (refrain from extending) the respective immutability period for a second snapshot data entity based at least in part on the respective retention period for the second snapshot data entity being less than the threshold duration.

Abstract: The meta data containing count and key fields of CKD records are reversibly decoupled from the user data of the data field so that the data can be deduplicated. Multiple CKD records may be coalesced into a larger size CKD track. The coalesced meta data is compressed and stored in a CKD hash table. The user data is hashed, and the hash is used as a hash key that is associated with the compressed meta data in the CKD hash table. When the hash of user data associated with a CKD write IO matches the hash key of an existing entry in the table, data duplication is indicated. The compressed meta data is added to the entry and the user data is deduplicated by creating storage system meta data that points to the pre-existing copy of the user data. The storage system metadata includes unique information that enables the corresponding compressed metadata to be subsequently located in the hash table to reassemble the CKD records.

Abstract: Time and value ordering may be applied for items stored in data backups. A change log that persists changes to a data set may be updated with changes and used to update an in-memory table for the data set, which describes changes to items up to a current time. An event may be detected to seal the in-memory table from subsequent updates and a persistent data object that orders the items in the in-memory according to both keys of the respective items and the respective time values of the items, as stored in the change log, may be generated and stored as part of a backup for the data set.

Abstract: Herein are techniques for dynamic aggregation of results of a database request, including concurrent grouping of result items in memory based on quasi-dense keys. Each of many computational threads concurrently performs as follows. A hash code is calculated that represents a particular natural grouping key (NGK) for an aggregate result of a database request. Based on the hash code, the thread detects that a set of distinct NGKs that are already stored in the aggregate result does not contain the particular NGK. A distinct dense grouping key for the particular NGK is statefully generated. The dense grouping key is bound to the particular NGK. Based on said binding, the particular NGK is added to the set of distinct NGKs in the aggregate result.

Abstract: A method includes parsing a query to determine a plurality of data processing operations associated with the query and including an AS OF JOIN operation between first time series data in a first table and second time series data in a second table. A query plan of the query is generated. The query plan includes a plurality of nodes corresponding to the plurality of data processing operations. At least one of the plurality of nodes corresponding to the AS OF JOIN operation is modified to generate a modified query plan of the query. The modifying is based on applying a UNION operation on at least a first portion of column data in the first table and the second table to obtain a combined table. Execution of the query by at least one of a plurality of computing nodes is scheduled based on the modified query plan.

Abstract: A facility for distributing a search index for a corpus of documents is described. The facility accesses multiple search index segments collectively making up the search index. Each of the segments is executable to traverse an index subtree embedded in the segment to find in the index subtree a node representing a query term specified in an argument. Each of the segments corresponds to a particular indexed document field that is indexed by the search index. For each of the segments, the index subtree embedded the segment has nodes representing query terms that all produce a particular hash result. The facility calls a programmatic publication interface for a content delivery network to publish the plurality of search index segments on the content delivery network.

Abstract: Lease-based consistency may be implemented for databases to handle failovers. A database node may obtain a consistency lease that describes a point in time determined from a time-to-live amount added to a consistent point in time for database data. While the consistency lease is valid, Multi-version Concurrency Control (MVCC) snapshots assigned by the database node can be used to handle requests to access the database data. Once expired, the database node may have to renew the consistency lease in order to continue to handle write and read requests.

Abstract: Conversion that includes multiple targets or sources of a data store migration may be performed. A request that specifies a mapping between multiple sources and one or more targets, one or more sources and multiple targets, or multiple sources and multiple targets may be created for data store migration. Metadata may be obtained for the selected sources and targets for the mapping. Conversion scripts may then be generated for the mapping based on an analysis using the metadata. The conversion scripts may then be stored for later execution.

Abstract: A computer system and computer-implemented method for duplicating an application state are provided, the method including: recording one or more point-in-time characteristics generated by prior user inputs at one or more user interface elements, the one or more point-in-time characteristics associated with a first application state of a first application instance; transferring the one or more point-in-time characteristics to the provisioned memory resources for generating the second application state; generating a second application instance based on the one or more point-in-time characteristics; configuring the second application state based on the one or more point-in-time characteristics to duplicate the first application state of the first application instance; and storing the prior user inputs in a journal, wherein the journal is configured to enable reproduction of a state of a plurality of modified states of the first application instance.

Abstract: The present disclosure provides systems, methods, and computer program products for providing efficient embedding table storage and lookup in machine-learning models.

Abstract: An indication to perform a backup of file system data is received. The file system data includes a content file having a size greater than a threshold size. A plurality of component file metadata structures for the content file are generated. File metadata of the content file split across the plurality of component file metadata structures is stored. The plurality of component file metadata structures are associated with different portions of the content file. A component file metadata structure of the plurality of component file metadata structures stores file metadata corresponding to a portion of the content file. The file metadata corresponding to the portion of the content file includes one or more references to locations of data chunks associated with the portion of the content file.

Abstract: A database management system maintains a collection of data using a log-based storage structure. In response to a request to store data items (key-value pairs), the database management system stores mapping information to a data item in an in-memory data structure while the data items are stored on the log-based storage structure. The hash of the key includes information to identify an index entry that comprises information that maps to the location of the data item stored on the log-based storage structure.

Abstract: A system and method for destroying data stored on an immutable distributed ledger utilizes technology from the following fields: encryption, digital signatures, data structures, distributed storage, distributed ledger technology, and smart contracts. Immutable distributed ledgers provide benefits for sensitive data, including availability, integrity, and data processing visibility. The system and method places sensitive data on an immutable distributed ledger and maintains these advantages of immutable distributed ledgers. The system and method also supports the efficient deletion of this sensitive data without compromising the integrity of the ledger.

Abstract: A method includes receiving data in a first format to be written into a data storage, where the first format is a multiversion concurrency control (MVCC) row format. The data is stored in the first format in the data storage and is transitioned from the first format to a second format based on a first transition parameter, where the second format is an MVCC columnar format. Data is stored in the second format in the data storage and is transitioned from the second format to a third format based on a second transition parameter, where the third format is a columnar format. The data storage stores non-overlapping data in the first format, the second format, and the third format.

Abstract: A method for facilitating automated synchronous data migration in live databases is disclosed. The method includes parsing a source database to identify characteristics, the characteristics corresponding to data tables in the source database; categorizing the data tables into groups based on the identified characteristics; validating the data tables in the groups by performing several tests; copying the groups to a target database based on a result of the validating; reviewing the copied groups to detect errors; and automatically initiating corrective actions in response to the detected errors. Additionally, the method further includes receiving new data sets from an application for persistence in the source database, the source database corresponding to a live database; generating duplicate data sets that correspond to the new data sets; and synchronously persisting, in real-time, the new data sets in the source database and the duplicate data sets in the target database.

Abstract: In a set of data analytics queries, at least a one of the queries comprising more than one operator, and each query being at least one of i) a producer of data for an other query in the set, and ii) a consumer of data from an other query in the set. In such examples, one or more computing devices identify each producer/consumer relationship between the queries. The one or more computing devices identify one or more optimizations among the queries based on the identified relationships. The one or more computing devices then apply at least one identified optimization to at least one of the queries.

Abstract: A system, method, and computer readable medium perform a method of query relaxation. A query output from a conversational system is received. At least one search term in the query is identified. Instance data is output to the conversational system in response to determining that the instance data in a data store matches the at least one search term in the query. The computer device outputs the received query to an external domain-specific knowledge source in response to determining that the at least one search term does not the match instance data in the data store. The computer device receives the relaxed data matches from the external domain-specific knowledge source being semantically-related to at least one search term in the query based on a plurality of criteria associated with the query. The computing device generates a response to the query based on contextual information and structural information.

Abstract: Updates are applied to a multi-entry bucket of a persistent multi-bucket hash table indexed by a hash index having a bucket portion and a collision portion, each entry of each bucket storing a corresponding value. The bucket is initially stored in a buffer and both a hash lookup structure and value lookup structure are generated for the bucket, the hash lookup structure usable to identify an entry of the bucket based on collision portion, the value lookup structure usable to identify an entry of the bucket based on value. For each update, a value of the update is applied to the value lookup structure to identify a corresponding entry, and the entry in the buffer is modified as required by the update. Subsequently the bucket in the buffer is persisted back to the hash table using the hash lookup structure.