Abstract: Techniques for performing machine learning inference calls in database query processing are described. A method for performing machine learning inference calls in database query processing may include generating a query plan to optimize a query for batch processing of data stored in a database service, the query plan including a batch mode operator to execute a function reference and an execution context associated with the batch mode operator, executing the query plan to invoke a function associated with the function reference, wherein the function sends a batch of requests, generated using the execution context, to a remote service and obtains a plurality of responses from the remote service, and generating a query response based on the plurality of responses.

Abstract: Disclosed is system for executing service request. The system comprises a processing arrangement and data sources. The processing arrangement receives the service request and is configured to extract data from the data sources based on the service request. The data sources respond in response to a characteristic framework of the service request. The system further comprises an administrator module to permute the service request received by the processing arrangement in accordance with the characteristic framework employed by the data sources. The administrator module is configured to identify at least one attribute of the service request, obtain data corresponding to the at least one attribute of the service request from the data sources, normalize the obtained data and provide the normalized data to execute the service request, via the processing arrangement.

Abstract: Various examples are directed to systems and methods for operating a database management system (DBMS) in a cloud environment. An assembly worker executing at a first computing device may provide a first database protocol message generated by a process code executing at the first computing device. A translation engine executed at the cloud environment may translate the first database protocol message from a first format to a second format associated with a DBMS instance executing at the cloud environment to generate a translated first database protocol message. The translation engine may cause the translated first database protocol message to be provided to the DBMS instance and may receive, from the DBMS instance, a first reply corresponding to the translated first database protocol message. The first reply may be sent to the process code.

Abstract: The present disclosure is directed towards systems and methods for finding documents that are similar to a reference text. The inventive systems and methods examine a set of collected documents to determine the facts present in those documents by, for example, extracting triplets and expanding them. A user's input reference text is similarly examined to extract and expand triplets therein and the facts identified with respect to the reference text are used as a basis to find documents having similar facts. The present disclosure is also related to systems and methods for mining facts from documents relating to a primary source such as a piece of legislation and using the mined facts to improve the results of subsequent searches.

Abstract: A computer system for processing unstructured data, the computing system comprising a computer processor, a computer memory operatively coupled to the computer processor and the computer memory having disposed within it computer program instructions that, when executed by the processor, cause the computing system to carry out the steps of receiving unstructured data input from a client device, analyzing the unstructured data for features that satisfy logical segment criteria by using natural language processing (NLP), partitioning the unstructured data into logical segments based on satisfaction of the logical segment criteria, and linking data from a repository to the unstructured data based on the logical segments.

Abstract: An association metric for record attributes associated with cardinalities that are not necessarily the same is used for training and/or applying an entity resolution (ER) model. A pair of records includes (a) a first record indicating a first set of values for a first attribute and (b) a second record indicating a second set of values for a second attribute. Each of the first set of values and each of the second set of values are compared to determine individual association metrics. A first-level reduction operation is applied to subsets of the individual association metrics to determine reduced association metrics. A second-level reduction operation is applied to the reduced association metrics to determine an association metric, for the pair of records, for training and/or applying an ER model.

Abstract: A method for processing requests, the method comprising: receiving, from a client application node and by a metadata node, a request for a data layout, determining a QoS tag associated with the request, identifying at least one file system block (FSB) in a sparse virtual space, wherein the FSB is mapped to a location in a storage pool that is associated with the QoS tag, generating the data layout based on the at least one identified FSB; and providing the data layout to the client application, wherein the client application node uses the data layout to directly write data to the location in the storage pool, wherein the metadata node does not participate in the writing of the data to the location in the storage pool.

Abstract: Described is hard link handling when a directory snapshot exists that includes the hard link's connected file object. A hard link is created by allocating a virtual inode number for the hard link, with the virtual inode number mapped to a real inode number that identifies a real inode of the file object; the hard link is assigned weight. A total weight associated with the real inode is increased by the hard link weight, and a hard link data store is updated with an entry for the hard link. Upon receiving data write request to the hard link, weight data determines that the file object is shared as a result of the snapshot; the hard link is disassociated from the real inode file, and associated with a new real inode number and new real inode of a new file object. The data is written based on the new real inode.

Abstract: A first direct index lookup table represents the current state of a storage object using entries with references corresponding to tracks of the storage object. A second direct index lookup table represents a first targetless snapshot of the storage object. A virtual replication data pointer table maps the entries of both the first direct index lookup table and the entries of the second direct index lookup table to backend storage via a system replication data pointer table. Updates to the storage object are represented using new entries in the first direct index lookup table and the system replication data pointer table. Movement of a track that is represented in multiple targetless snapshots that are represented by multiple direct index lookup tables is represented by updating the corresponding virtual replication data pointer table and system replication data pointer table rather than updating each of the direct index lookup tables.

Abstract: The present disclosure is directed towards systems and methods for finding documents that are similar to a reference text. The inventive systems and methods examine a set of collected documents to determine the facts present in those documents by, for example, extracting triplets and expanding them. A user's input reference text is similarly examined to extract and expand triplets therein and the facts identified with respect to the reference text are used as a basis to find documents having similar facts. The present disclosure is also related to systems and methods for mining facts from documents relating to a primary source such as a piece of legislation and using the mined facts to improve the results of subsequent searches.

Abstract: Disclosed are various approaches for providing a natural language interface for searching databases. A natural language query is parsed to identify a plurality of tokens. At least one operator is identified with a machine-learning model based at least in part on the plurality of tokens. Next, at least one attribute and at least one respective attribute value are identified with a machine-learning model based at least in part on the plurality of tokens. Then, at least one constraint is identified with a machine-learning model based at least in part on the plurality of tokens. Finally, a machine language query is generated based at least in part on the at least one operator, the constraint(s), the attribute(s), and the respective attribute value(s).

Abstract: A method for execution by a query processing system includes receiving a query expression that includes a call to a computing window function. The computing window function is executed in accordance with execution of the query expression against a database. Execution of the query expression includes accessing an ordered set of rows of the database indicated in the call to the computing window function, and applying a recursive definition indicated in the call to the computing window function to each row in the ordered set of rows to generate output for each row in the ordered set of rows. A query resultant for the query expression is generated based on the output for each row in the ordered set of rows.

Abstract: In one example, a method involves performing an initial discovery process that includes querying a storage array, and identifying, based on the query, one or more hosts that are registered with the storage array. This initial discovery process is performed automatically without requiring user action to identify the one or more hosts. The method additionally includes presenting a list of discovered hosts, receiving a selection input from a user specifying one or more of the hosts in the list, retrieving, from the storage array, information associated with each of the respective hosts, and making the information available to a user.

Abstract: A file storage application that processes file operations is communicably connected with a block storage application that processes block operations by establishing multiple communication sessions between the file storage application and the block storage application. Multiple logical volumes provided by the block storage application are exposed to the file storage application over the multiple communication sessions established between the file storage application and the block storage application using a total number of logical paths to the logical volumes that is equivalent to the total number of the logical volumes provided by the block storage application to the file storage application.

Abstract: A client computer establishes one or more sessions with a DBMS. Session context information for each session is cached in a client-side session cache. When a session disconnection is detected, the session recovery includes establishing a new session with the DBMS, and restoring the session context of the disconnected session from the session information of the disconnected session that is stored in the session cache.

Abstract: This disclosure is directed to a system and method that detects duplicated content and/or media items. A media item can be split into media item portions. Based on the media item portions, features can be determined. Using the features, media item portion signatures can be determined to generate a media item signature. The media item signature can be compared with a different media item signature to determine duplicated content within the media items.

Abstract: A method, computer program product, and computing system for maintaining a first cursor pointer for a memory system, wherein the first cursor pointer concerns a first search criteria and identifies an initial first cursor starting location; receiving a first search request for the memory system that satisfies the first search criteria; executing the first search request within an initial first search window starting at the initial first cursor starting location; and updating the first cursor pointer to identify an updated first cursor starting location based, at least in part, upon the initial first search window.

Abstract: The disclosure herein describes writing data to a log-structured merge (LSM) tree file system on an object storage platform. Write data instructions indicating data for writing to the LSM tree file system are received. Based on the received instructions, the data is written to the first data cache. Based on an instruction to transfer data in the live data cache to the LSM tree file system, the first data cache is converted to a stable cache. A second data cache configured as a live data cache is then generated based on cloning the first data cache. The data in the first data cache is then written to the LSM tree file system. Use of a stable cache and a cloned live data cache enables parallel writing data to the file system by the stable cache and handling write data instructions by the live data cache.

Abstract: Technologies for performing stochastic similarity searches in an online clustering space include a device having a column addressable memory and circuitry. The circuitry is configured to determine a Hamming distance from a binary dimensionally expanded vector to each cluster of a set of clusters of binary dimensionally expanded vectors in the memory, identify the cluster having the smallest Hamming distance from the binary dimensionally expanded vector, determine whether the identified cluster satisfies a target size, and add or delete, in response to a determination that the identified cluster does not satisfy the target size, the binary dimensionally expanded vector to or from the identified cluster.

Abstract: A selection of content from a content presentation is received. At least one topic from the selected content is extracted using natural language processing (NLP). The at least one topic is representative of a subject conveyed within the selected content. At least one perspective associated with the at least one topic is extracted using NLP. The at least one perspective is representative of a point of view conveyed within the selected content regarding the at least one topic. A topic rating of the extracted topics and associated perspectives is determined based upon the extracted topics and associated perspectives. The topic rating is representative of a topic diversity among the extracted topics and associated perspectives. The topic rating is presented within a graphical user interface (GUI).