Abstract: A system includes initiation of recovery from a crash of a database instance and, during the recovery, identification of a subset of pages of a memory image of the database instance which are not reusable and replacement of the identified subset of pages with corresponding pages of the savepoint image.

Abstract: The disclosure generally describes methods, software, and systems, including a method for caching information. A request for information is received at a staging layer. A determination is made whether the requested information already exists entirely in a cache. In response to determining that the requested information does not exist entirely in the cache, at least some of the requested information from the underlying database layer is retrieved. A completeness level to which the requested information exists in the cache is determined. In response to determining that the requested information exists partially in the cache, cache-nonresident information is retrieved from the underlying database layer and merged with information in the cache. In response to determining that none of the requested information exists in the cache, the request is forwarded to the underlying database layer. The requested information is provided by the staging layer to a presentation layer.

Abstract: Disclosed herein are system, method, and computer program product embodiments for resolving an execution order for a plurality of tasks, such as solving complex calculations in order, installing new software in order, or scheduling instructions by compilers in order. The approach uses a dependency graph to perform a topological sort, and applies a heuristic to determine which node to execute next from remaining nodes without dependency issues. The approach favors reducing resource conflicts with a resource used by a previous node, resolving larger numbers of dependencies early, and forcing early resolution of resource conflicts.

Abstract: Intelligent loading of user interface resources is provided herein. Intelligent loading can include generating a usage pattern for a user interface. The usage pattern can be generated based on usage data gathered from use of the user interface. The usage data can be categorized by user, and user-specific usage patterns can be generated based on that user's usage data. Usage patterns can be generated via a machine-learning algorithm. The machine-learning algorithm can be trained on the usage data of a user interface. The machine-learning algorithm can be trained on usage data of a specific user of a user interface. The usage pattern can indicate a predictive order of use of resources in a user interface. Resources for a user interface can be loaded based on the usage pattern. Loading the resources based on a usage pattern can be performed in parallel to loading and using a user interface.

Abstract: Technologies are described for facilitating transaction processing within a database environment that includes one or more processing units and a memory. A request for a database operation, such as a query, is received from a client. Execution statistics associated with the request are retrieved. At least some of the execution statistics are compared with one or more threshold values. Based on results of the comparing, the request is carried out with a first execution component or delegated to a second execution component. The request is executed and execution results are sent to the client.

Abstract: Disclosed herein are system, method, and computer program product embodiments for the translation of a level based format, where each record contains a complete individual path from the highest to the lowest leaf level, to an annotated hierarchy through a generator function which through mapping and encoding on a level-by-level basis, keeping track of sibling links, is able to perform a depth-first pre-ranking order assessment of original data, to produce an annotated hierarchy which may be utilized by a user.

Abstract: A service request is received at an intelligence service server from a user, where the service request includes a number of required inputs associated with the user. The number of required inputs are executed by the intelligence service server to generate an inference, an outcome probability distribution and a price quote, where the price quote corresponds to the outcome probability distribution. The outcome probability distribution and the price quote are returned by the intelligence service server to the user. It is determined by the intelligence service server that whether the user accepts the price quote based on a response from the user. If so, the inference is returned by the intelligence service server to the user. Otherwise, the response from the user is logged in a database associated with the intelligence service server by the intelligence service server.

Abstract: Disclosed is a solution that tags and stores video, audio, pictures, text, and other content, along with their associated metadata to allow for contextual retrieval and customized playback of stored content.

Abstract: Disclosed herein are system, method, and computer program product embodiments for accelerating queries with zone map enhancements. An embodiment operates by receiving a query including a condition on one or more values from a column of a database table. It is determined that the values from the column are divided into a plurality of data zones, and that a sorted dictionary includes an ordered arrangement of the values from the column. A min value and a max value from a satisfaction array that indicates which values satisfy the condition are determined. Which of the data zones are disqualified are determined based on the values of the disqualified data zone falls outside of the range between the minimum value and the maximum value. The values of any remaining zones are evaluated to determine a result of the query. The result set is returned.

Abstract: Techniques and solutions are described for facilitating software testing. A test script is automatically generated from an incident report. Natural language processing can be used, at least in part, to generate the test script. The test script can be manually or automatically executed, including automatically executing the test script upon the occurrence of trigger event or according to a schedule. A trigger event can include determining that a software update has been deployed and is available for testing. It can be determined whether test execution indicates that a particular program behavior, such as an error, has been resolved. Machine learning can be used to classify program behavior. A user interface screen can be rendered to present test results.

Abstract: Techniques and solutions are described for providing high-availability computing resources to service client requests. Groups of computing nodes are organized into loops, a given loop being configured to execute a particular subset of tasks, such as tasks with a hash value in a particular ranged serviced by a loop. Computing nodes within a loop can evaluate a task request to determine whether the task request conflicts with another task currently assigned to a node. If a computing node which sent out a task request determines that no conflict was identified, it can execute the task request. Communications within a loop can occur unidirectionally, such that a node which initiated a communication will receive the communication from the last loop node. Loops can be connected to form a ribbon, the ribbon providing a namespace for task execution, where hash ranges for the namespace are uniquely assigned to loops of the ribbon.

Abstract: Disclosed herein are system, method, and computer program product embodiments for generating a single transaction data stream from multiple database logs. An embodiment operates by assigning a first unique identifier to a first log entry associated with a disk store, a second unique identifier to a second log entry associated with in an in memory row store, and a third unique identifier to a third log entry based on the second unique identifier. The embodiment further operates by determining a first stream package based on the first log entry, a second stream package based on the third log entry, and a replay order for the first stream package and the second stream package based on the unique identifiers. In some embodiments, the operation further includes sending the first stream package and the second stream package to a replicant server in accordance with the replay order.

Abstract: Disclosed herein are system, method, and computer program product embodiments for appropriately routing requests for data stored in multiple storage mediums. An embodiment operates by maintaining a first and second data stored on a first storage medium in communication with a second storage medium. Thereafter, a replicate of the first data stored in the first storage medium may be created for the second storage medium to store a replica data mirroring the first data. Subsequently, a request for retrieval of the first data may be received. Afterward, a previous update time of the second storage medium in receiving the replicate of the first data stored in the first storage medium may be determined. Lastly, based on the previous update time, the request may be forwarded to the first storage medium or second storage medium.

Abstract: Techniques and solutions are described for facilitating the distribution of tables and table data between a shared container and one or more tenant containers. The tables can include a sharing indicator that can designate a table (or a view on a table) as having a materializable type. Materializable types can initially be deployed in the shared container, and a tenant container can have read access to the table via a view. If the tenant attempts to modify the materializable table, the table can be materialized in the tenant container. Metadata for a materializable table can include materialization instructions. Techniques and solutions are also provided for converting between sharing types, and for updating shared containers or table schemas.

Abstract: A system and method including authoring a semantic layer universe and at least one semantic layer set container created on top of the objects of the semantic layer universe and concurrently authoring of resources to the universe by one or more users based on a merge strategy based on at least a status of the resource being authored relative to the published universe.

Abstract: Some embodiments provide a non-transitory machine-readable medium that stores a program. The program receives from a client device a request for data stored in a database categorized based on a hierarchical dimension. The hierarchical dimension is configured to store values from a plurality of values organized in a hierarchy comprising a plurality of levels. The program further generates an intermediary data structure by combining a first set of the data stored in the database and a second set of data based on a level in the hierarchy. The program also generates results for the request by aggregating the data in the intermediary data structure based on the level in the hierarchy. The program further provides the results for the request to the client device.

Abstract: Disclosed herein are system, method, and computer program product embodiments for generating a data structure and responding to an aggregation query. An embodiment operates by a computer implemented method that includes receiving, by at least one processor, a request to enter data in a block in a data structure. The method further includes reading, by the at least one processor, a first aggregated value stored in the data structure. The first aggregated value represents a sum over at least values associated with data that is stored in the block in the data structure. The method further includes calculating, by the at least one processor, a second aggregated value by adding the first aggregated value and a value associated with the data to be entered in the data structure. The method further includes storing, by the at least one processor, at least part of the data to be entered and the second aggregated value in the block in the data structure.

Abstract: Techniques and solutions are described for recording document transactions in a blockchain. Document transactions can include sending or receiving a document, or various workflow steps involving a document. Other steps in a workflow that includes a document can be recorded in the blockchain. Blockchain entries related to a document can include a unique identifier of the document, such as a hash value. When a document is sent between two computing systems, the document can be encrypted using a public encryption key of the receiving computing system. Sending and receiving of documents, and recording of blockchain transactions, can be facilitated by an intermediary service. A service providing recording of blockchain transactions can abstract recording details from applications calling the service. A party having the unencrypted document, or the unique document identifier (such as legitimately retaining or receiving the identifier), can retrieve a transaction history of the document from the blockchain.

Abstract: A system includes reception of a database query, determination of a first logical query execution plan to execute the database query, determination of a second logical query execution plan to execute the database query, determination of a first plurality of resource usage estimates associated with the first logical query execution plan, determination of a second plurality of resource usage estimates associated with the second logical query execution plan, and determination, based on the first plurality of resource usage estimates and the second plurality of resource usage estimates, whether it is determinable to a suitable degree of certainty that one of the first logical query execution plan and the second logical query execution plan will be associated with a smaller amount of resource usage than the other one of the first logical query execution plan and the second logical query execution plan.

Abstract: User-generated input is received that includes a sequence of words associated with initiation of a computer-implemented event. Thereafter, such input is parsed using at least one natural language processing (NLP) model. This parsed input is then used by a machine learning model to determine a suggested template having a plurality of fields for initiating the event. The template can then be presented in a graphical user interface. Related apparatus, systems, techniques and articles are also described.