Abstract: An approach is provided for storing a snapshot of an execution image of a virtual machine (VM) instance. Based on VM, application, criticality, and change parameters, and based on a recommendation model trained on historical data about effects that changes caused by VM execution had on a capacity of a storage system that stores execution images of VM instances in first datastores, a tier is selected from multiple tiers of the storage system as the tier recommended for storing the snapshot. A datastore is selected for storing the snapshot so that the selected datastore is associated with the selected tier and selected from second datastores dedicated to storing only snapshots. The snapshot is created in the selected datastore and in the selected tier and unused space in a portion of the storage system for storing the first datastores is reduced.

Abstract: A relational database storage method according to one embodiment generates a join graph including tables and edges based on either a database schema including referential constraints among database tables or a query workload containing a set of join predicates, classifies the type of tables based on the cost of a join operation between the tables included in the join graph, classifies the type of edges in the join graph on the basis of the type of the tables connected to each other by the edge, adds a virtual edge between two tables satisfying the indirect join condition, and generates a database partitioning scheme for the join graph based on the set of tables in the join graph and their type, the set of edges in the join graph and their type, and virtual edges.

Abstract: In one implementation, lightweight incremental backups are created by a data recovery tool, which provides an extra layer of protection against data loss, and also provides a way of moving repositories between databases without the need of copying each of the repositories. Implementations include backing up a distributed cluster of services providing a simple, safe, and convenient way of instantiating files system repositories in a database, thereby avoiding the performance and scalability issues of using a database repository itself. In other implementations, a lightweight backup object is created that includes a threshold amount of compressed information. The lightweight backup objects may be configured to provide the differences between revisions, and/or may be configured to replicate or rebuild the repository from scratch. Implementations may be configured to backup and restore any single or revision set in any repository state either online or offline.

Abstract: A method of operating a storage system is provided. The method includes detecting differing amounts of storage memory on two or more of a plurality of blades of the storage system and forming a plurality of resiliency groups of blades, based on the differing amounts of storage memory, each resiliency group comprising a subset of the plurality of blades of the storage system supporting data recovery in case of failure of two of the blades of the resiliency group.

Abstract: If a query indicates joining of N (N is an integer of no less than 3) relation tables, a database management system calculates a query cost that is based on a power consumption amount for query execution, for each of a first query plan candidate in which a temporary table is generated and the temporary table is stored in a storage device unit with a first access cost, a second query plan candidate in which a temporary table is generated and the temporary table is stored in a storage device unit with a second access cost that is higher than the first access cost and a third query plan candidate in which the N relation tables are joined in a pipelined manner without generation of a temporary table, selects a relatively low-query cost query plan candidate and executes the query based on the query plan candidate.

Abstract: A device automatically extracts a data file from an upstream source based on ingestion parameters. The data file is in a first format that is not readable by a downstream data analysis utility and includes a plurality of data records. Each record includes one or more entries. The ingestion parameters include a file identifier, validation criteria, transformation instructions, and storage instructions for the extracted data file. The device generates table(s) based on the extracted data file by, determining, based on the validation criteria, whether the record is permitted to be ingested. The device determines, based on the transformation instructions, a table in which to store each entry and a row and column of the determined table in which to store the entry. Entries are input into the determined tables, rows, and columns. The table(s) are stored in a second format that is readable by the downstream data analysis utility.

Abstract: A method is described. The method includes generating an access model that simulates a transformation of existing new technology file system (NTFS) permissions for a plurality of shared folders. The method also includes creating permission groups for the plurality of shared folders based on the access model. The method further includes updating the NTFS permissions of the shared folders based on the access model and permission groups.

Abstract: Techniques for processing relational data with a user-defined function (UDF) are provided. Relational input data being requested by the UDF, from within a relational database system, is intercepted and normalized. The UDF is called with the normalized input data and as the UDF produces output data in response to the normalized input data that output data is captured and normalized. In an embodiment, the normalized output data is used to dynamically update a data model within the relational database for the input data.

Abstract: Example methods and apparatus release storage locations in a tiered storage system by deleting expired tier-copies of a file stored in the tiered storage system. One embodiment includes a policy-class circuit that defines a tier-copy policy associated with a tier, where the tier-copy policy includes an expiration interval (EI) associated with the tier, and a schedule associated with the tier. A copy expiration circuit checks, according to the schedule, for expiration-eligible tier-copies of a file stored in the tiered storage system based, at least in part, on an EI associated with a tier, and on an access time (AT) associated with the file, where the AT denotes the last time the file was accessed. The copy expiration circuit, upon determining that a tier-copy is expiration eligible, declares the tier-copy as expiration eligible. A delete/restore circuit, upon the copy expiration circuit declaring that the tier-copy is expiration eligible, deletes the tier-copy.

Abstract: The MULTI-SOURCE, MULTI-DIMENSIONAL, CROSS-ENTITY, MULTIMEDIA ANALYTICAL MODEL SHARING DATABASE PLATFORM APPARATUSES, METHODS AND SYSTEMS (“AMSDB”) transform data aggregated from various computer resources using AMSDB components into updated entity profiles and/or social graphs. In one implementation, the AMSDB aggregates data records including search results, purchase transaction data, service usage data, service enrollment data, and social data. The AMSDB identifies data field types within the data records and their associated data values. From the data field types and their associated data values, the AMSDB identifies an entity. The AMSDB generates correlations of the entity to other entities identifiable from the data field types and their associated data values. The AMSDB also associates attributes to the entity by drawing inferences related to the entity from the data field types and their associated data values.

Abstract: A mechanism is provided for determining distinct values of an attribute in a data table including the attribute and data records stored in a set of data blocks. Attribute value information is provided for each data block that comprises information indicating distinct values of the attribute in the data blocks. At least part of a current data block of the data table is scanned to determine current distinct values of the attribute. The current distinct values of the attribute of the current data block are added to a list of current distinct values of the data table. The scanning and adding are repeatedly performed for a further data block of the data table if the attribute value information of the further data block indicates at least one distinct value that is not present in the list. The list of current distinct values are then provided.

Abstract: Implementations of the present disclosure include receiving a merger request including a first identifier of a first node, a second identifier of a second node, and an identifier of a first smart asset, the first smart asset being associated with the first node, determining whether the first smart asset is capable of merger with a second smart asset associated with the second node at least partially based on the identifier of the first smart asset and an identifier of the second smart asset, and, in response to determining that the first smart asset is capable of being merged with the second smart asset, automatically: updating the second smart asset, the updated second smart asset being assigned a value, and including properties of the first smart asset, recording the updated second smart asset to the second node, and deleting the first smart asset.

Abstract: Technologies are described herein for executing queries expressed with reference to a structured query language against unstructured data. A user issues a structured query through a traditional structured data management (“SDM”) application. Upon receiving the structured query, an SDM driver analyzes the structured query and extracts a data structure from the unstructured data, if necessary. The structured query is then converted to an unstructured query based on the extracted data structure. The converted unstructured query may then be executed against the unstructured data. Results from the query are reorganized into structured data utilizing the extracted data structure and are then presented to the user through the SDM application.

Abstract: Techniques for splitting a time-range query into sub-queries for parallel execution are provided. In one embodiment, a user query requesting items from a database is received. A total number of items in the database that correspond to a time range of the user query is determined. Based on the total number of items, target values are determined. Each target value represents a target number of items in the database to be assigned to a respective sub-query. Based on the target values, a plurality of time periods within the time range are determined. Each time period corresponds to a respective set of items in the database platform, and a total number of items in the respective set of items corresponds to a respective target value. Sub-queries defining respective time periods of the plurality of time periods are then generated from the user query and executed in parallel.

Abstract: Techniques for selecting a storage node of a storage system to store data include applying a first function to at least some data chunks of an extent to provide respective first values for each of the at least some data chunks. A storage node, included within multiple storage nodes of a storage system, is selected to store the extent based on a majority vote derived from the respective first values.

Abstract: Techniques for splitting a time-range query into sub-queries for serial execution are provided. In one embodiment, a user query is received requesting items within a time range from a database. The time range is divided into a plurality of time periods within the time range. Sub-queries defining respective time periods of the plurality of time periods are generated from the user query, and a first sub-query is executed. The first sub-query defines a first time period of the plurality of time periods, where the first time period is a most-recent time period or a least-recent time period among the plurality of time periods. If it is determined that a number of items obtained from executing the first sub-query is greater than or equal to a predetermined result target, then the items obtained from executing the first sub-query are provided and subsequent sub-queries are not executed.

Abstract: A method for clustering product media files is provided. The method includes dividing each media file corresponding to one or more products into a plurality of tiles. The media file include one of an image or a video. Feature vectors are computed for each tile of each media file. One or more patch clusters are generated using the plurality of tiles. Each patch cluster includes tiles having feature vectors similar to each other. The feature vectors of each media file are compared with feature vectors of each patch cluster. Based on comparison, product groups are then generated. All media files having comparison output similar to each other are grouped into one product group. Each product group includes one or more media files for one product. Apparatus for substantially performing the method as described herein is also provided.

Abstract: The subject matter described herein relates to implementation of a dictionary in a column-based, in-memory database where values are not stored directly, rather, for each column, a dictionary is created with all distinct values. For each row, a reference to the corresponding value in the dictionary is stored. In one aspect, data is stored in a memory structure organized in a column store format defined by a plurality of columns and a plurality of rows. A dictionary for each column in the memory structure is generated. The dictionary has distinct values for each column. A reference to the dictionary is generated for each column in the memory structure. The dictionary and the reference to the dictionary are stored in the memory structure.

Abstract: Computer file copy systems and methods in which a list of files to be transferred from storage on a source file storage computer is sorted by one or more processors of a copy computer in ascending order from a smallest file size to a largest file size. Thereafter, files having the smallest file sizes are selected from the sorted files such that a total size of the selected files is equal to or less than a total size of memory available on the copy computer, the selected files are copied concurrently with one another from storage on the source file storage computer to the available copy computer memory by the one or more processors, the files are written concurrently with one another from the memory on the copy computer to storage on a target file storage computer by the one or more processors, and the process is performed iteratively.

Abstract: Disclosed is a method, apparatus, and system for a method for implementing a mutual-exclusion semantic for target update, the method comprising: receiving a tagset update request for a tagset associated with an object, the tagset update request comprising one or more metadata tags to be updated; determining whether updating the tagset based on the received tagset update request would cause a value of any existing metadata tag in the tagset to change; and in response to determining that updating the tagset based on the received tagset update request would cause a value of at least one existing metadata tag in the tagset to change, rejecting the tagset update request.