Abstract: A data processing method and system for a cloud platform, and an electronic apparatus and a storage medium. The data processing method includes: an acquisition rate of performance data of each data type in a cloud platform is determined (S101); according to the acquisition rate, a shard corresponding to each data type is determined (S102); a database mapping table according to correspondence relationships between shards and databases is generated (S103); and when a data operation request is received, according to the database mapping table, a target database corresponding to the data operation request is determined, and a corresponding operation in the target database is executed (S104). By using the above method, the load balancing of a cloud platform can be realized, and the data processing efficiency can be improved.

Abstract: Dynamic partitioning of a search space of queries is implemented for flexible, heuristic database querying. Search space partitioning refers to dividing the search space for a submitted query into smaller parts by augmenting the queries to append thereto an additional predicate comprising a dynamic partition key and a value(s) selected based on heuristics (e.g., recency and/or relevancy of the value(s)). A plurality of candidate augmentations of the query and corresponding query plans are generated and evaluated based on additional heuristics to determine which can be executed to yield the best results in terms of result quality and latency. This query plan is selected and executed for retrieval of results that satisfy the query, with pagination utilized for presentation of the results. The procedure of generating candidate query plans, selecting one of the candidates for execution, and paginating results is repeated until a search termination criterion is satisfied.

Abstract: Techniques and systems can analyze information associated with instructions to sort data to determine an identifier common to at least a plurality of individual instructions to sort the data. A correspondence of the identifier to a sort identifier used to sort the data can be determined. Based on the determined correspondence, the techniques and systems can sort the data based on the identifier.

Abstract: A system and method to generate and manage clusters of search terms. The system and method generate a cluster including a set of search terms. A size of the cluster is compared to a cluster size threshold level to determine that the cluster comprises a valid-sized cluster. A noise rate of the cluster is compared to a noise rate threshold level to determine that the noise rate is less than the noise rate threshold level, wherein the noise rate relates to a distance of the cluster from a set of other clusters. Performance data associated with the cluster is generated based on a comparison with one or more cluster performance metric thresholds. A user interface is generated to display information associated with the cluster comprising the performance data. Input is received via the user interface comprising an adjustment of the one or more cluster performance metrics thresholds.

Abstract: A computer implemented method for automatically customizing electronic commerce includes analyzing information relevant to the user's purchasing preferences, determining the user's purchasing preferences based on the analysis of the relevant information, determining a shopping list, wherein the shopping list comprises one or more products the user wants to purchase and one or more indications of the user's purchasing preferences for each of the one or more products, generating one or more digital shopping carts based on the shopping list, comparing a product in at least one of the one or more digital shopping carts to a corresponding product and a corresponding purchasing preference on the shopping list, determining one or more preferred digital shopping carts based on the comparison, and displaying to the user the one or more preferred digital shopping carts.

Abstract: A system, apparatus and methods for generating database entries and tools for accessing and searching a database from an Ontology. Starting with an Ontology used to represent data and relationships between data, the system and methods described enable that data to be stored in a desired type of database and accessed using an API and search query generated from the Ontology. Embodiments provide a structure and process to implement a data access system or framework that can be used to unify and better understand information across an organization's entire set of data. Such a framework can help enable and improve the organization and discovery of knowledge, increase the value of existing data, and reduce complexity when developing next-generation applications.

Abstract: Techniques may include receiving a plurality of spans of trace data at a computing system during a first time period. The techniques may include storing the plurality of spans in a span partition of a data store. The data store can contain a plurality of span partitions with spans that are grouped in the partition by trace identifier. The device may include generating a timestamp partition, with an index of timestamps by trace identifiers, for the first time period. The techniques may include storing the timestamp partition in the data store. Also, the techniques may include identifying at least two timestamp partitions that correspond to a second time period that preceded the first time period. The techniques may include generating and storing a primary compacted timestamp partition by combining the at least two timestamp partitions.

Abstract: An enterprise database server may store enterprise information. An application layer server, coupled to the enterprise database server, may use an application programming language (e.g., ABAP) to access the enterprise information. The application layer server may execute a plurality of domain runtimes, each associated with a domain service (e.g., transactional, SQL, analytical, search, etc.). The application layer server may also access, by a first domain runtime, a first data service entity in an infrastructure for data modelling (e.g., CDS) that enables data models to be defined and consumed on the enterprise database server. Similarly, the application layer server may access, by a second domain runtime, a second data service entity in the infrastructure for data modelling. According to some embodiments, the first data service entity includes a first provider contract and the second data service entity includes a second provider contract (different than the first provider contract).

Abstract: An example method of tuning a machine learning operation can include receiving a data query comprising a reference to an input data set of a database, generating a plurality of unique sets of hyperparameters by varying a hyperparameter value of each set of hyperparameters of the plurality of unique sets of hyperparameters based on the input data set, in response to receiving the data query, training a plurality of machine learning models using the input data set of the data query, each of the plurality of machine learning models configured according to a respective one of a plurality of unique sets of hyperparameters, selecting a first machine learning model of the plurality of machine learning models based on an accuracy of an output of the first machine learning model, and returning the output of the first machine learning model in response to the data query.

Abstract: Methods, systems, devices and/or apparatus related to finding and sharing information on a large-scale or global computer network such as the Internet, and, specifically, to allow for collaboratively searching for, organizing and/or accessing information.

Abstract: In a set of data analytics queries, at least a one of the queries comprising more than one operator, and each query being at least one of i) a producer of data for an other query in the set, and ii) a consumer of data from an other query in the set. In such examples, one or more computing devices identify each producer/consumer relationship between the queries. The one or more computing devices identify one or more optimizations among the queries based on the identified relationships. The one or more computing devices then apply at least one identified optimization to at least one of the queries.

Abstract: A location-aware search assist capability identifies location-aware search query suggestions using location information associated with the location-aware search query suggestions. A user's search query input and location and a location associated with each location-aware search query suggestion candidates may be used to identify a set of search query suggestions for presentation to the user. Location-aware search query suggestion candidates may be ranked in accordance with a closeness of each one's location to the user's location. The ranking may be performed using a score, such as a popularity score associated with each search query suggestion candidate. The location-aware search query suggestion candidates having a location closer to the user's location may be promoted by adjusting each candidate's popularity score upward, and the search query suggestion candidates that are farther away from the user's location may be demoted by adjusting each such candidate's popularity score downward.

Abstract: Systems and methods for processing loops in computational graphs representing machine learning models are disclosed. An example method begins with obtaining data representing a computational graph. Data identifying an allocation of the computational graph across devices is obtained. Additionally, one or more nodes in the computational graph that represent a respective control flow statement are identified. For each identified node, a structure of nodes and edges that represents an operation that provides a current state of recursion or iteration in the respective control flow statement is generated. This structure is inserted into the computational graph and the allocation of nodes to devices is modified to assign the structure to a device.

Abstract: Systems and methods for recommending content to an online service user are presented. In response to a request from a user, a set of n-grams of the request are generated, with each n-gram comprising one or more terms from the request and each n-gram of the set of n-grams being unique. Embedding vectors projecting the n-grams into a content item embedding space are generated, and the embedding vectors are combined into a representative embedding vector for the request. The nearest content items are identified according to a distance measure between a projection of the representative embedding vector and embedding vectors of content items of a corpus of content items in the content item embedding space. At least some of the nearest content items are returned as recommended content in response to the request from the user.

Abstract: This disclosure describes, in part, techniques for estimating frequency values of top contributors of a high cardinality data set in a computationally efficient manner. In one example, the techniques utilize a probabilistic data structure, such as a count-min-sketch, for identifying which contributors are the largest contributors to a particular metric. After identifying these largest contributors using the probabilistic data structure, the techniques may store the contributor IDs and the their respective estimated cumulative value (e.g., frequency counts) in a top-N list.

Abstract: A method for execution by a database system includes determining physical network topology data of a plurality of nodes of the database system utilized to facilitate query executions. Initial route selection data indicating a set of initial relay nodes for each of a plurality of pairs of the plurality of nodes is based on the physical network topology data by selecting one of a plurality of possible initial route options for the each of the plurality of pairs of the plurality of nodes. The initial route selection data is sent to the plurality of nodes. A query resultant is generated by facilitating execution of a query, via the plurality of nodes, by facilitating routing of data blocks between at least one of the plurality of pairs of the plurality of nodes via a set of intermediate nodes selected based on the initial route selection data in accordance with a dynamic data block routing scheme.

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: One embodiment provides a method, including: utilizing at least one processor to execute computer code that performs the steps of: providing, on a display device, a communications window, wherein the communications window comprises a request for user input to start a conversation with an online assistant; receiving a user input identifying a request by the user to be completed by the online assistant; updating, based upon the request, the communications window, wherein the updated communications window comprises a summary of the conversation including prepopulated variable terms and allows user interaction to adjust the summary including adjustment of the prepopulated variable terms; and iteratively updating the communications windows based upon user input adjusting the summary. Other aspects are described and claimed.

Abstract: One embodiment of the invention is directed to a method for performing efficient data searches in a distributed computing system. The method may comprise, receiving a search request including a key. The key may be provided to a block-based table manager via a programming interface external to a virtual machine executing on a computer system. The programming interface may provide a translation between a first programming framework of the virtual machine and a second programming framework of the block-based table manager. Providing the key may cause the block-based table manager to conduct a search for a value corresponding to the key. The value may be provided in response to the search request. Utilizing such block-based tables may enable a data search to be performed using on-board memory of computing node operating within a distributed computing system.

Abstract: Aggregation in a computing system can include receiving, at a service node of the computing system, a first query specifying aggregation and translating the first query into a second query having a first canonical format and specifying the aggregation. The method can include forwarding the second query to a first subset of a plurality of endpoint nodes and translating, at each endpoint node of the first subset, the second query into a third query having a format executable by a data source connected to the endpoint node. The third query can specify a level of the aggregation to be performed by the data source determined based upon a processing capability of the data source. The endpoint nodes can initiate execution of the third query by the data sources and provide an aggregated result including a result from the data source(s) to the service node.

Abstract: In one embodiment, the present disclosure includes a cloud based management system controlling a plurality of automated food production systems residing at different locations. A machine learning algorithm may be used to modify recipe data to customize recipes.

Abstract: A user may submit a search query to a search engine, which may process the search query and generate a set of results. Previous actions of the user or of other users may be used to identify recommended search queries. The recommended search queries may be presented to the user along with the search results for the query. The location of the recommended search queries may be fixed, set by the user, or determined on a query-by-query basis. A user interface element operable to cause the presentation of one or more recommended search queries may be presented after a delay. After an additional or independent delay, the recommended search queries may be presented automatically. The delays may be fixed, set by the user, or determined on a query-by-query basis.

Abstract: Methods, systems, and computing platforms for data communication are disclosed. The processor(s) may be configured to electronically receive a SQL query request for a relational computer readable database having plurality of computer readable records. The processor(s) may be configured to electronically process the SQL query request with a machine learning controller based on a set of machine learning training data so as to output at least one probable smart execution plan for executing the SQL query request. The processor(s) may be configured to electronically output the at least one probable smart execution plan for executing the SQL query request to a relational database management processing module.

Abstract: Methods and systems for querying application metadata objects are provided. The method may comprise receiving a query identifying an application; executing the query against a metadata model for the application to obtain a result, the result identifying a first metadata object satisfying the query, wherein the metadata model identifies a plurality of metadata objects and hierarchical relationships between the plurality of metadata objects, and wherein the metadata model comprises information about structure and semantics of data used by the application; determining, from the plurality of metadata objects, a second metadata object referenced by the first metadata object; determining, from the plurality of metadata objects, a third metadata object that is a parent object of the first metadata object; building a directed graph including the first metadata object, the second metadata object, and the third metadata object; and providing a response including the directed graph to the query.

Abstract: Systems and methods that offer significant improvements to current content processing management systems for heterogeneous and widely distributed content sources are disclosed. The proposed systems and methods are configured to provide a framework and libraries of extensible components that together are designed enable creation of solutions to acquire data from one or more content repositories, possibly distributed around the world across a wide range of operating systems and hardware, process said content, and publish the resulting processed information to a search engine or other target application. The proposed embodiments offer an improved architecture that incorporates manager nodes and worker (processing) nodes, where worker nodes are configured to scan and process data, while manager nodes are configured to handle all allocation of work (including throttling) and control state and failover. Such an arrangement enables the system to perform with greater scalability and reliability.

Abstract: Described are methods, systems and computer readable media for computer data distribution architecture connecting an update propagation graph through multiple remote query processors.

Abstract: Embodiments of the disclosure provide systems and methods for identifying hierarchically related files stored in a database. According to one embodiment, this can be accomplished using a pipeline of worker applications which each use a specific table stored in the database 35 and a sequence or series of queues. Generally speaking, each worker application can read a message identifying tasks to be performed from one queue in the sequence of queues, perform those tasks using the appropriate table from the database to identify a set of files, and write one or more messages to a subsequent queue in the sequence or series of queues to pass results and/or further tasks to be performed to a subsequent worker application in the pipeline of worker applications.

Abstract: This document describes techniques and devices for a refined search with machine learning. These techniques improve computer-aided searches through enabling selection of search criteria used in a prior search and providing a refined search result based on that selection. Furthermore, a machine-learning component of a search engine can be altered to improve future search results based on the selection and an indication of the desirability of the refined search result.

Abstract: According to various embodiments, the Query Context Translation Engine identifies a topic of a search query history received during a current user session. The search query history in a first language. The Query Context Translation Engine identifies, in a translation table, target text that corresponds with a query in the search query history, the target text comprising at least one word. The Query Context Translation Engine obtains at least one search result based on a translation of the target text in a second language.

Abstract: Techniques are described for responding to aggregate queries using optimizer statistics already available in the data dictionary of the database in which the database object targeting by the aggregate query resides, without the user creating any additional objects (e.g. materialized views) and without requiring the objects to be loaded into volatile memory in a columnar fashion. The user query is rewritten to produce a transformed query that targets the dictionary tables to form the aggregate result without scanning the user tables. “Accuracy indicators” may be maintained to indicate whether those statistics are accurate. Only accurate statistics are used to answer queries that require accurate answers. The accuracy check can be made during runtime, allowing the query plan of the transformed query to be used regardless of the accuracy of the statistics. For queries that request approximations, inaccurate statistics may be used so long as the statistics are “accurate enough”.

Abstract: Embodiments of the present disclosure describe systems, methods, and computer program products for executing and tuning a machine learning operation within a database. An example method can include receiving a data query referencing an input data set of a database, executing a plurality of machine learning operations to generate, in view of the input data set, a plurality of output data sets each having a respective accuracy value, wherein each of the plurality of machine learning operations is executed by a processing device according to one of a plurality of unique sets of hyperparameters, selecting a first output data set of the plurality of output data sets in view of the accuracy values, and returning the first output data set in response to the data query.

Abstract: Operations include halting a runaway query in response to determining that a performance metric of the query exceeds a performance threshold. The runaway query halting system receives a query execution plan associated with a query and divides the received execution plan into one or more components. For each component, the system determines a predicted resource usage associated with executing the component. The system further determines a predicted resource usage associated with the query execution plan based on the predicted resource usage associated with each component. The system executes the query associated with the received query execution plan and compares the predicted resource usage associated with the query to a resource usage threshold. In response to determining that the predicted resource usage of the query execution plan exceeds the resource usage threshold, the system halts execution of the query associated with the query execution plan.

Abstract: Methods and apparatus are provided for using machine learning to estimate query resource consumption in a massively parallel processing database (MPPDB). In various embodiments, the machine learning may jointly perform query resource consumption estimation for a query and resource extreme events detection together, utilize an adaptive kernel that is configured to learn most optimal similarity relation metric for data from each system settings, and utilize multi-level stacking technology configured to leverage outputs of diverse base classifier models. Advantages and benefits of the disclosed embodiments include providing faster and more reliable system performance and avoiding resource issues such as out of memory (OOM) occurrences.

Abstract: The present invention relates to a system and method of selling goods or services, or collecting recycle refuse using mechanized mobile merchantry, comprising positioning, by self-propelling, at least one of a mechanized mobile merchantry within a geographical boundary, allowing interaction with consumers, and effectuating selling of goods or services, or collection of recycle refuse with consumers. Other exemplary embodiments can include signaling a mechanized mobile merchantry with a consumer's mobile device to direct the merchantry to self-propel to the consumer's location, and utilizing usage logs and algorithms to optimize functionality of a fleet of merchantry and reposition the merchantry, as necessary, within a geographical boundary, to increase sales and consumer convenience.

Abstract: A device may receive source code from a source data structure, and may receive information associated with a target data structure. The device may analyze the source code to extract statements, and may utilize natural language processing on the statements to identify functions and keywords associated with the source data structure. The device may train a machine learning model with the functions and the keywords to generate a trained machine learning model, and may process the information associated with the target data structure, with the trained machine learning model, to transform a source query to a target query compatible with the target data structure. The device may process the target query, with a neural network model, to generate an optimized target query, and may cause data from the source data structure to be migrated to the target data structure based on the optimized target query.

Abstract: Resource provisioning systems and methods are described. In an embodiment, a system includes a plurality of shared storage devices collectively storing database data, an execution platform, and a compute service manager. The compute service manager is configured to determine a task to be executed in response to a trigger event and determine a query plan for executing the task, wherein the query plan comprises a plurality of discrete subtasks. The compute service manager is further configured to assign the plurality of discrete subtasks to one or more nodes of a plurality of nodes of the execution platform, determine whether execution of the task is complete, and in response to determining the execution of the task is complete, store a record in the plurality of shared storage devices indicating the task was completed.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining query refinements using search data. In one aspect, a method includes receiving a first query and a second query each comprising one or more n-grams for a user session, determining a first set of query refinements for the first query, determining a second set of query refinements from the first set of query refinements, each query refinement in the second set of query refinements including at least one n-gram that is similar to an n-gram from the first query and at least on n-gram that is similar to an n-gram from the second query, scoring each query refinement in the second set of query refinements, selecting a third query from a group consisting of the second set of query refinements and the second query, and providing the third query as input to a search operation.

Abstract: A method, a computer program product, and a computer system co-assign operators to service an incident with a first operator educating a second operator of skills used in servicing the incident. The method includes determining a type of incident and determining a command group used for the type of incident according to a first mapping based on historical resolutions of historical incidents. The method includes determining the first operator for co-assignment according to a second mapping based on successful historical resolutions of historical incidents by the first operator. The method includes determining the second operator for the co-assignment according to the second mapping that further maps the command group and the second operator based on a lack of experience in using the command group. The method includes co-assigning the incident to the first operator and the second operator.

Abstract: Various examples are directed to systems and methods optimizing database queries. A database management system may receive a first query comprising a plurality of query expressions. The database management system may determine that a first expression of the first query is nullable and that the first expression is null preserving. The database management system may generate optimized query code for the first query. The optimized query code may comprise a first code segment and a conditional jump instruction. The first code segment that, when executed by a processor, may cause the processor to perform operations comprising determining a value of the first expression. The conditional jump instruction may, when executed by the processor, cause the processor to perform operations comprising: skipping execution of at least a portion of the first code segment and returning null for the first expression.

Abstract: A database engine includes one or more computing devices, each having one or more processors and memory. The memory stores programs configured for execution by the processors. The database engine receives a database query from a client, and parses the database query to build a query operator tree. The query operator tree includes a plurality of query operators. The database engine performs one or more optimization passes on the query operator tree, including a deduplication optimization pass, to form an optimized execution plan. The deduplication optimization pass includes determining that a first query operator is equivalent to a second query operator during a traversal of the query operator tree, and replacing the second query operator with a link to reuse results from the first query operator. The database engine executes the optimized execution plan to retrieve a result set from the database and returns the result set to the client.

Abstract: Techniques and solutions are described for converting data models between formats, such as between a conceptual data model and a physical data model for a database system, or between a conceptual data model and artefacts to be implemented in the database system. The conceptual data model is annotated with annotations that allow the physical data model or database artefacts to be automatically generated from the conceptual data model. The annotations can reflect relationships between entity types in the physical data model, such as inheritance relationships, header/item relationships, or one-to-one cardinality relationships. Annotations can also indicate attributes that should be added to entity types in the conceptual data model, such as attributes for versioning or data governance, that may not be used in the conceptual data model. Annotations can be used to determine how entity types in the conceptual data model will be denormalized in the physical data model.

Abstract: A system and method for creating and editing graph schema data structures in a collaborative, real time graphical editing environment is disclosed. The system and method further operates to integrate external sources of data as components of the graph schema. The system and method outputs definition files for the graph schema data structure and mapping files.that enable the population of a graph database.

Abstract: Systems, methods, and computer-readable media for pinning query execution plans are described. A database system may obtain a query, and determine a organization identifier and/or user identifier associated with the query. The database system may identify an optimized plan among a set of query plans that correspond with the query, and may generate a query execution plan based on the query and the optimized plan. The database system may determine a pinned plan that is associated with the optimized plan, and may generate a query execution plan based on the pinned plan when the pinned plan is associated with the optimized plan and corresponds with the organization identifier and/or the user identifier. The database system may generate a query execution plan using optimization functions when the pinned plan is not associated with the pinned plan, or the organization identifier and/or the user identifier. Other embodiments may be described and/or claimed.

Abstract: The present invention relates to optimized access of a database. Herein are techniques to accelerate execution of any combination of ad hoc query, heterogenous hardware, and fluctuating workload. In an embodiment, a computer receives a data access request for data tuples and compiles the data access request into relational operators. A particular implementation of a particular relational operator is dynamically selected from multiple interchangeable implementations. Each interchangeable implementation contains respective physical operators. A particular hardware operator for a particular physical operator is selected from multiple interchangeable hardware operators that include: a first hardware operator that executes on first processing hardware, and a second hardware operator that executes on second processing hardware that is functionally different from the first processing hardware.

Abstract: Systems and methods are provided for automatically orchestrating the handling of events through a processing pipeline without limitation (or without a substantial limitation) as to the size of the event payload associated with the event. The event pipeline system stores event payloads in data stores and generates notifications regarding the events. The notifications may be placed into event streams for handling by various processing components of the event pipeline system. The processing components may receive notifications or events that they are to process, and may separately access event payloads from the data stores. The processing components may generate and save processed event payloads to the data stores in a streaming fashion such that the computing resources of the processing components do not limit (or substantially limit) the size of the event payloads that the processing components may handle.

Abstract: Techniques are described herein for building a framework for declarative query compilation using both rule-based and cost-based approaches for database management. The framework involves constructing and using: a set of rule-based properties tables that contain optimization parameters for both logical and physical optimization, a recursive algorithm to form candidate physical query plans that is based on the rule based tables, and a cost model for estimating the cost of a generated physical query plan that is used with the rule based properties tables to prune inferior query plans.

Abstract: In one aspect, a method includes generating a design space for a peptide for an application. The generating includes identifying sequences for the peptide, and updating the sequences by determining, for each of the sequences, a respective set of activities pertaining to the application. The updating produces updated sequences each having updated respective activities. The method includes generating, based on the updated sequences, a solution space within the design space. The solution space includes a target subset of the updated sequences. The method includes performing, using a machine learning model to process the solution space, trials to identify a candidate drug compound that represents a sequence having a level of activity that exceeds a threshold level, and transmitting information describing the candidate drug compound to a computing device.

Abstract: A user interface (UI) for visualizing search data provides techniques for grouping and organizing aggregate data that shows the categories of topics included in search queries from a large number of individual users. Raw search queries are categorized into one of a number of topical categories. The search queries are assigned to a geographic location based on geolocations of computing devices generating the search queries. The UI presents a map that shows the number of search queries per topical category for each geographic location displayed in the current UI view. As a result of this UI design, a user can easily understand the interaction between geographic location and frequency of search query topics. Trends in the geographic distribution of searches and in the categories of topics searched are also easily understood from this UI design by changing the time range of the search queries displayed.

Abstract: Technologies for dynamic accelerator selection include a compute sled. The compute sled includes a network interface controller to communicate with a remote accelerator of an accelerator sled over a network, where the network interface controller includes a local accelerator and a compute engine. The compute engine is to obtain network telemetry data indicative of a level of bandwidth saturation of the network. The compute engine is also to determine whether to accelerate a function managed by the compute sled. The compute engine is further to determine, in response to a determination to accelerate the function, whether to offload the function to the remote accelerator of the accelerator sled based on the telemetry data. Also the compute engine is to assign, in response a determination not to offload the function to the remote accelerator, the function to the local accelerator of the network interface controller.

Abstract: A large highly parallel database management system includes thousands of nodes storing huge volume of data. The database management system includes multiple query optimizers for determining low cost execution plans for queries. The database management system is adapted to receive a data query. An execution plan generator component of the database management system generates an initial execution plan for the query. The initial execution plan is fed as input to more than one query optimizers. Each optimizer starts with the initial execution plan, generates alternative execution plans, and determines a satisfactory execution plan that incurs the lowest cost. The database management system compares the selected execution plans by the optimizers and selects one with the lowest cost. The multiple query optimizers run in parallel.