Abstract: A method for improving asynchronous data replication between a primary storage system and a secondary storage system is disclosed. In one embodiment, such a method includes monitoring, in a cache of the primary storage system, unmirrored data elements needing to be mirrored, but that have not yet been mirrored, from the primary storage system to the secondary storage system. The method maintains an LRU list designating an order in which data elements are demoted from the cache. The method determines whether a data element at an LRU end of the LRU list is an unmirrored data element. In the event the data element at the LRU end of the LRU list is an unmirrored data element, the method moves the data element to an MRU end of the LRU list. A corresponding system and computer program product are also disclosed.

Abstract: Methods for identifying data files responsive to a search query and weighted according to a stylistic preference are provided. In one aspect, a method includes receiving a search query for a collection of media files from a device associated with a user associated with a user identifier, and identifying a set of image files from the collection that is responsive to the query. Each image file in the set is associated with a relevancy score. The method also includes retrieving a stylistic preference profile associated with the user identifier to apply to the set, and weighting the score for each image file according to the profile to generate a weighted stylistic preference score for each image file. The method further includes creating a listing of the set that is sorted according to weighted stylistic preference score, and providing the listing to the device. Systems and machine-readable media are also provided.

Abstract: Systems and methods for automated data visualization are disclosed. According to one embodiment, in an information processing apparatus comprising at least one computer processor, a method for automatically analyzing data from a plurality of data files and producing a visual output may include (1) receiving a plurality of data files from at least one data file source, each data file comprising a plurality of data columns; (2) identifying a number of columns in each of the data files; (3) using column-value matching to identify two matching columns from two different data files; (4) executing one or more machine learning algorithms on data in the data files to enrich the data; and (5) generating a first visualization for the enriched data and outputting the first visualization on a display.

Abstract: Processing and storage responsibility for a data set may be split according to separately stored shards of the data set. As one or more loads associated with shards of the dataset grow a re-sharding operation may be performed to reduce loading of particular shards and nodes that host the particular shards. A re-sharding operation may cause only a sub-set of as set of shards of the dataset to be split and only cause second portions of the split shards to be stored in additional computing nodes. In some embodiments, a number of shards to be included in the sub-set of shards to be split may be selected based on an overall number of shards in the set and a largest number in the Fibonacci sequence that is less than the overall number of shards in the set.

Abstract: A system and method for backing up virtual machines are disclosed. To better use the backup resources, an optimal set of virtual machines needing backup during a time window is generated. The optimal set depends on a total time for backing up each virtual machine needing backup and a cost metric that indicates a cost of not backing up the virtual machine during the time window. The optimal set also meets various system constraints that reflect the backup resource limitations. Performing backups according to the optimal set limits the number of missed backups and the number of service level agreement violations, thereby improving the backup performance without costly additions to its infrastructure.

Abstract: A system tokenizes values stored in a field by multiple records. The system creates a trie from the tokenized values, each branch in the trie labeled with one of the tokenized values, each node storing a count indicating the number of the multiple records associated with a tokenized value sequence beginning from a root of the trie. The system tokenizes a value stored in the field by a prospective record. Beginning from the root of the trie, the system identifies each node corresponding to a token value sequence for the prospective record's tokenized value. Beginning from the most recently identified node for the prospective record's token value sequence, the system identifies each extending node which stores a count that satisfies a threshold, each identified extending node corresponding to another token value sequence. The system uses the other token value sequence to identify one of the multiple records that matches the prospective record.

Abstract: Systems, methods, and non-transitory computer readable media are provided for validating data for integration. A file to be integrated into a dataset may be determined. The file may be characterized by a file name and metadata. Whether the dataset can be identified based on the file name and the metadata may be determined. Based on the dataset not being identifiable, information needed to identify the dataset may be requested. Based on the dataset being identifiable, the file may be routed for integration into the dataset.

Abstract: Extension patterns are deployed to core deployment patterns stored in a first software repository. An event relating to registration of a component is detected. In response to detecting the event, a tool registry can be queried to identify one or more extension deployment patterns for use with one or more core deployment patterns. Based upon the query, one or more extension deployment components can be retrieved from a second software repository. The one or more extension deployment components can be deployed on top of one or more instances of the one or more core deployment patterns.

Abstract: A pattern finding device of embodiments includes a pattern generator, a pattern storage, a pattern selector, a pattern deleter. The pattern generator generates a group of patterns from data including a plurality of items. The pattern storage stores the group of patterns generated by the pattern generator. The pattern selector selects at least one pattern to be deleted from the group of patterns with reference to an evaluated value indicating a change of an item distribution condition in the group of patterns when at least one pattern which is selected as a candidate is excluded from the group of patterns stored in the pattern storage. The pattern deleter deletes the at least one pattern that is selected by the pattern selector from the pattern storage.

Abstract: The present invention extends to methods, systems, and computer program products for developing an item data model for an item. Aspects of the invention can automate the process of data collection of “facts” for “items” that information is needed about. Facts can be organized and normalized to eliminate redundant facts, and interpret what is found. Data requirements extraction and automated modeling using a combination of data virtualization, data analytics, extract, transform, and load (ETL), web crawlers, and reverse engineering systems, can be used along with other technologies to develop an item model. A model owner feeds a curating module with the information for locating the facts to be used, and initiating the modeling process. Existing data structures, websites, vendor input, etc. can be described to the import process, and an item model is produced. The model can be imported into existing modeling tools for viewing, or viewed as XML.

Abstract: Systems and methods are described herein for efficiently providing personalized search results to a user. Personalized search allows a user to receive search results that are most relevant to that user. For example, when a user with a strong interest in rap searches for “50 cent” they may be presented with top search results corresponding to 50 Cent's rap albums. Conversely, when a coin collector searches for “50 cent” they may be presented with top search results containing antique coins.

Abstract: Two different representations of content are maintained, at least one of which is a composite representation of the content. The composite representation is multiple different components, each component stored as a separate file and each component including some of the data of the content. The composite representation also includes a manifest that identifies the various components. Different devices can edit the content, and access to the content is coordinated via a synchronization system. The synchronization system manages generation of one or more representations of the content. The synchronization system also manages synchronization of the different representations of the content, allowing changes made on various different computing devices to be reflected in the different representations of the content, and managing situations in which different computing devices may attempt to change the content concurrently.

Abstract: A method, article of manufacture, and apparatus for managing data. In some embodiments, this includes, receiving a query from a client, based on the received query, analyzing a catalog for location information, based on the analysis, determining a first storage system, an associated first file system and an associated first protocol, using the associated first protocol to communicate with the first storage system, and performing at least a portion of the query on the first storage system.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for managing data for regulated environments. One of the methods includes collecting data from a plurality of sources; analyzing the data; and providing one or more outputs based on the analysis.

Abstract: Systems and methods for automating the retrieval of partitionable search results from a database and reducing the latency of delivering the partitionable search results are disclosed. A query for searching a database is constructed. Multiple processing tasks are assigned and executed to retrieve partitioned search results from a database in parallel. Each partition of search results is returned to the user as response messages. Response messages may be continuously received until all search results responsive to the query have been retrieved from the database. When multiple processing tasks are used to retrieve and deliver partitioned search results in parallel, server, internet and client resources are optimized, and the throughput of search results and metadata are increased, and the process of narrowing or filtering search results is an interactive and informed process for the user.

Abstract: A method comprising receiving subject data indicative of a subject entity. First entity data is obtained from a knowledge database using the subject data, the first entity data representative of a first related entity related to the subject entity. First text data is obtained from the knowledge database using the first entity data, the first text data representative of first text associated with the first related entity. A subject entity data set is generated. The subject entity data set comprises first related entity data based on the first entity data, and the first text data.

Abstract: A statement parsing method for a database statement comprises: conducting lexical analysis on a database statement inputted into a database, to obtain an inputted word sequence; looking up a statement similarity table according to the inputted word sequence to determine whether there is an existing word sequence similar to the inputted word sequence in the statement similarity table; if yes, obtaining the parsed data corresponding to the existing word sequence from the statement similarity table; otherwise, parsing the inputted word sequence to obtain parsed data corresponding thereto, and storing the inputted word sequence and the corresponding parsed data in the statement similarity table; and executing the database statement inputted to the database based on the parsed data corresponding to the existing or inputted word sequence. The method can quickly parse a database statement and is favorable for improving the response speed and the working efficiency of a database.

Abstract: De-normalized data structure files generation systems and methods are provided. The system obtains files from sources wherein each file include records, parses files to validate records and attributes in the records, identifies a set of similar files from the validated files, and append two or more files from the set of similar files to obtain one or more consolidated files. Each of the one or more consolidate files corresponds to a specific category. The system further a predefined logic validation on each of the one or more consolidated files to obtain a logic validated file for each of the one or more consolidated files. Each logic validated file obtained for the one or more consolidated files include validated records. The system further generates a de-normalized data structure file including de-normalized records by merging each of the logic validated files to be used for generating intelligence reports.

Abstract: Provided are a method, an apparatus, and a computer program stored in a computer readable medium for conducting an arithmetic operation efficiently in a database management server.

Abstract: A device receives a knowledge graph and an ontology for the knowledge graph, and receives a query for information associated with the knowledge graph. The device generates candidate responses to the query, and assigns scores to the candidate responses based on the knowledge graph. The device identifies a particular candidate response, of the candidate responses, based on the scores for the candidate responses, and determines, based on the knowledge graph, a neighborhood of the particular candidate response. The device generates knowledge graph embeddings for the neighborhood of the particular candidate response, and determines a particular neighborhood, with a smallest loss of quality, based on the knowledge graph embeddings. The device generates a reasoning graph based on the ontology and the particular neighborhood, and generates an explanation of the particular candidate response based on the reasoning graph. The device performs an action based the explanation of the particular candidate response.