Abstract: A supervised similarity measure machine learning method, system, and computer program product that includes generating embeddings by training a supervised deep neural network (DNN) on a feature data to determine which nodes correspond to which clustered learning group of clustered learning groups, performing half-distributed learning by distributing data in a time-series database to the clustered learning groups, and evaluating, based on the embeddings, new tenant data in the clustered learning groups with an upward bow pose.

Abstract: A computer-implemented method selects an access path for high cost and/or complex queries. The method includes building a classification model configured to identify a lowest cost access path. The method further includes receiving a query, where the query is configured to retrieve a set of data from a database. The method also includes generating an access map for the query, where the access map includes one or more potential access paths to execute the query. The method includes collecting, for the query, a set of data for each potential access path. The method further includes classifying, by the classification model, the query. The method also includes selecting a first access path of the one or more potential access paths and executing the query.

Abstract: A computer-implemented method includes: collecting, by a computing device, database activities and database structure information of a database; identifying, by the computing device, related columns in the database; determining, by the computing device, one or more data types for column transference of the identified related columns; generating, by the computing device, a super union column based on the column transference and the identified related columns; and updating, by the computing device, the database with the super union column.

Abstract: An embodiment includes generating a partition schema for a distributed database based on historical usage data indicative of usage of the distributed database, where the generating of the partition schema comprises determining a partition range of a partition of the partition schema. The embodiment also includes generating a node identifier for the partition using a hash function and a first weight value assigned to the partition. The embodiment also includes monitoring performance data indicative of a performance of the distributed database, the monitoring comprising detecting a failure of the performance to satisfy a performance threshold. The embodiment also includes initiating, responsive to detecting the failure, a redistribution procedure by changing the node identifier of the partition by replacing the first weight value with a second weight value.

Abstract: Query content-based data generation includes obtaining a query having an outer query and one or more subqueries, converting predicates of a subquery to be predicates of the outer query, and thereby obtain a transformed query, generate from the transformed query a query block having a list of predicate(s) selected from the transformed query, obtaining column information based on the list of predicate(s) and using the column information to select data for a dataset such that data records from the dataset are produced as results to executing the obtained query against the dataset.

Abstract: Disclosed are techniques for relational database locks based on columns. Database transactions may be targeted to specific columns of one or more records, instead of the entire row for those records, using primary keys. Column locks on specific keys are stored separately than column locks on ranges of keys, which are both checked when requesting a new column lock for either a single key or a range of keys. When a threshold number of columns for a given record, or range of records/keys, have been locked, the column locks for that record, or range of records, can be combined into a single row level lock to reduce resource costs for maintaining multiple concurrent locks.

Abstract: A computer-implemented method that includes managing a buffer pool of pages into a ring sub-chain comprising pages linked in a ring, and a linear sub-chain comprising pages linked in a line from a header, and moving a page between the linear sub-chain and the ring sub-chain based on a moving schema evaluating a chain management characteristic.

Abstract: In a computer-implemented method for improving performance of a database, a processor receives batch jobs for a relational database. The batch jobs may include a first member with a first buffer pool, and a second member with a second buffer pool. The processor may also identify a first actual object and an isolation level for the batch jobs, generate related queries based on the first actual object and the isolation level, calculate a cost for the first member and the second member to run the batch jobs based on the related queries, and assign the batch jobs to the first member based on a lower calculated cost.

Abstract: A method of managing time series data workload requests includes receiving a workload job request from a user in a multi-tenant network, the request specifying a plurality of workloads, each workload including time series data configured to be stored in a time series database (TSDB), inputting workload information to a workload model that is specific to the user, and classifying each workload according to the workload model, the workload model configured to classify each workload based on a plurality of parameters, the plurality of parameters including at least a workload type and an amount of storage associated with each workload. The method also includes assigning each workload of the plurality of workloads into one or more workload groups based on the classifying, and executing each workload according to the workload type and the storage size.

Abstract: An approach is provided for optimizing a database buffer pool. Based on statistics about read and write operations in a range of pages, the range of pages is determined to be a candidate for a hot read range for which locks and latches are avoided in processing operations in the hot read range. Using an on-time trigger or pre-trigger process, the hot read range is created from the range of pages by marking start and end points in the range of pages. Write operation(s) are determined to be included in the hot read range by marking an object control block. The write operation(s) are added to a write pending list. The read operations in the hot read range are performed without a latch or lock. The write operation(s) are merged from the write pending list to the range of pages and the write operation(s) are performed.

Abstract: Embodiments of the present invention relate to a method, system, and computer program product for multiplexing data operation. In some embodiments, a method is disclosed. A query for at least one table comprising a plurality of data records is received. The query indicating a plurality of data operations to be performed on the plurality of data records. The plurality of data operations are combined into a target data operation. An intermediate result of the query is generated by performing the target data operation on the plurality of data records. A final result of the query is determined based on the intermediate result. In other embodiments, a system and a computer program product are disclosed.

Abstract: A method, system, and computer program product to plan and schedule executions of various utility tasks of a utility command during a maintain window, the method including receiving a utility command. The method may also include identifying possible utility tasks used to execute the utility command. The method may also include determining preferred utility tasks. The method may also include calculating a degree of parallelism for the preferred utility tasks. The method may also include generating a utility execution plan for the utility command. The method may also include analyzing the utility execution plan against resource constraints of a time window and sub time windows of the time window. The method may also include generating a time window execution plan for each sub time window of the sub time windows. The method may also include updating the utility execution plan with the time window execution plans.

Abstract: An embodiment includes executing a querying process that returns database documents containing text associated with a database environment. The embodiment tokenizes the database documents into a series of n-gram tokens and groups the tokens into topic classes using natural language processing (NLP). The embodiment also generates a feature map by applying a convolution layer to an image that depicts a database structure graph of the database environment. The embodiment detects an architectural element in the image by applying a region proposal network (RPN) to the feature map and classifies the architectural element into one of the plurality of topic classes using a trained neural network. The embodiment renders, responsive to a user selection of the architectural element, an overlay depicting information about the architectural element using text and other architectural elements that are in a same topic class as the user-selected architectural element.

Abstract: A system, program product, and method for enhancing automatic multidimensional query processing. The method includes executing a database query including semi-joining a plurality of dimension tables with a fact table. The method also includes identifying for extraction one or more data values from each dimension table of the plurality of dimension tables. The data values from each dimension table of the plurality of dimension tables are associated with a respective record identification (RID), thereby defining one or more RIDs. The method further includes generating a plurality of RID lists. Each RID list of the plurality of RID lists includes a collection of the one or more RIDs for the respective dimension table. The method also includes merging the plurality of RID lists, sorting, subject to the merging, the plurality of RIDs as a function of data location, and fetching the data values from the fact table.

Abstract: Systems and methods for high availability distributed data storage are provided. In embodiments, a method includes: receiving, by a remote direct memory access (RDMA) switch operatively coupled to a computing device, a request to access a page of a database; determining, by the RDMA switch, a validation state of the page; determining, by the RDMA switch, a status of the page; updating, by the RDMA switch, the status of the page based on the validation state and the request; and reporting, by the RDMA switch, the validation state.

Abstract: A method includes: obtaining, by a database server, security policies associated with a database hosted by the database server; obtaining, by the database server, first database data associated with the database; creating, by the database server, a knowledge distillation model using the security policies and the first database data; obtaining, by the database server, second database data associated with the database; creating, by the database server, a classification model using the second database data and an output of the knowledge distillation model; receiving, by the database server, a client database request to the database; determining, by the database server, a new security policy rule set using the classification model; and applying, by the database server, the new security policy rule set to the client database request.

Abstract: Systems and methods are provided to predict connection configurations in a distributed database environment. In embodiments, a method includes: building, by a computing device, a dynamic machine learning predictive model for a first new node connection in a distributed database environment based on configuration data of the first new node connection; determining, by the computing device, a group of nodes in the distributed database environment of a same type of node as the new node connection; federating, by the computing device, the dynamic machine learning predictive model of the first new node connection with other dynamic machine learning predictive models of other nodes in the group of nodes, thereby generating a federated predictive model for the same type of node; and generating, by the computing device, a predicted connection configuration for a second new node connection of the same type of node using the federated predictive model.

Abstract: A computer implemented system and method allow database performance testing and improvement. The method comprises using a database system (DBS) for embedding a virtual resource simulation system (VRSS) into a node of the DBS, loading a new configuration parameter of a database into the VRSS, receiving, by the VRSS from the DBS, an action control block (ACB), and dynamically activating and assigning a service task by a service task dispatcher of the VRSS to process the ACB. The method further comprises simulating a process of database execution based on a content of the ACB by creating a limited database internal data structure that is based on the ACB and the new configuration parameter, and manipulating the limited database internal structure to produce a simulation result. The method further comprises utilizing a metric associated with the simulation result of the new configuration parameter in the simulated process of database execution.

Abstract: A method includes training, by a computing device, a model to identify SQL queries; identifying, by the computing device, incoming SQL queries; grouping, by the computing device, the incoming SQL queries into groups; monitoring, by the computing device, resource usage of the incoming SQL queries in the groups; and allocating, by the computing device, resources to an SQL query of the incoming SQL queries based on the monitoring.

Abstract: A simulated dataset is queried for regression by validating a structured query language (SQL) statement, determining a pattern type of the SQL statement, reconstructing the SQL statement according to a predetermined process for the pattern type, creating a mutated SQL statement for querying a simulated dataset, and validating the mutated SQL statement. The simulated dataset is based on a confidential dataset having the confidential elements removed or replaced.