A computer-implemented method for efficiently estimating the number of unique elements in a collection of elements comprises generating, via hash logic, hash values associated with the elements. The hash values specify bit positions within an array of bits. Hash values output from the hash logic conform to a geometric distribution such that bit positions of the array of bits corresponding to lower orders bits are more likely to be generated than bit positions corresponding to higher-order bits. Bits of the array of bits corresponding to the bit positions are set. The number of bits of the array of bits that are set is counted. Estimation logic estimates the number of unique elements of the collection of elements as a function of the number of bits of the array of bits that are set.

An electronic device includes a plurality of nodes, each node having a processor that performs operations for processing instances of input data through a model, a local memory that stores a separate portion of model data for the model, and a controller. The controller identifies model data that meets one or more predetermined conditions in the separate portion of the model data in the local memory in some or all of the nodes that is accessible by the processors when processing the instances of input data through the model. The controller then copies the model data that meets the one or more predetermined conditions from the separate portion of the model data in the local memory in the some or all of the nodes to local memories in other nodes. In this way, the controller distributes model data that meets the one or more predetermined conditions among the nodes, making the model data that meets the one or more predetermined conditions available to the nodes without performing remote memory accesses.

Described are methods, systems and computer readable media for providing a current row position query language construct and array processing query language constructs and associated processing.

This disclosure relates to, among other things, scalable data processing, storage, and/or management systems and methods. Certain embodiments disclosed herein provide for a multi-dimensional data storage structure that may facilitate parallel processing of data during both data ingestion and data retrieval and/or access processes involving one or more storage layers In various embodiments, sequence number schemes associated with ingested data records may allow for improved management of data stored in one or more layers (e.g., hot and/or cold data storage layers).

Implementations of the present disclosure include methods, systems, and computer-readable storage mediums for receiving first and second data sets, both the first and second data sets including structured data in a plurality of columns, for each of the first data set and the second data set, inputting each column into an encoder specific to a column type of a respective column, the encoder providing encoded data for the first data set, and the second data set, respectively, providing a first multi-dimensional vector based on encoded data of the first data set, providing a second multi-dimensional vector based on encoded data of the second data set, and outputting the first multi-dimensional vector and the second multi-dimensional vector to a loss-function, the loss-function processing the first multi-dimensional vector and the second multi-dimensional vector to provide an output, the output representing matched data points between the first and second data sets.

Improved access of a multidimensional database can be implemented using a space-filling curve, such as a z-order curve, and a tree structure. A state engine is configured to track the state of the dimensions of the database as the tree structure is traversed. Different states allow optimizations such as avoidance of analyzing a dimension for further nodes of the tree and removal of multiple nodes in a sub-tree.

A system for returning synthetic database query results. The system may include a memory unit for storing instructions, and a processor configured to execute the instructions to perform operations comprising: receiving a query input by a user at a user interface; determining, based on natural language processing, a type of the query input; determining, based on the received query input and a database language interpreter, an output data format; returning, based on a generation model and the output data format, a result of the query input; providing, to a plurality of training models and based on the determined query type, the query input and the result; and training the training models, based on the query input and the result.

A system includes a memory and a node. The memory stores first and second log string correlithm objects. The node receives first and second real-world numerical values, and identifies a first sub-string correlithm object from the first log string correlithm object representing the first real-world numerical value and a second sub-string correlithm object from the second log string correlithm object representing the second real-world numerical value. The node aligns the first and second log string correlithm objects such that the first sub-string correlithm object aligns with the second sub-string correlithm object. The node identifies a sub-string correlithm object from the second log string correlithm object representing the logarithmic value of one. The node determines which sub-string correlithm object from the first log string correlithm object aligns with the identified sub-string correlithm object from the second log string correlithm object. The node outputs the determined sub-string correlithm object.

A computer-implemented method, system and computer program product for handling lock contentions of an index (e.g., SQL index). Leaf pages of an index are monitored for a lock contention during an insert operation of index keys by the transactions. Upon detecting a lock contention of a leaf page, the next index key to be inputted into such a leaf page is routed to a queue of a buffer. The index key that was stored in the queue of the buffer is then mapped to the particular leaf page experiencing the lock contention where the transaction originally attempted to store such an index key, where such mapping is stored in a data structure. Upon such a leaf page no longer experiencing a lock contention, the appropriate index keys are then removed from the buffer and stored in the appropriate leaf page based on the mapping identified in the data structure.

The present invention discloses a method and a system for hybrid query based on a cloud analysis scene, and a storage medium. The method comprises the following steps: obtaining query information, and obtaining an index thereof based on the query information; obtaining meta-information of the index based on pre-computation, and comparing the obtained meta-information with the meta-information of an aggregate index; and determining a query mode corresponding to the meta-information based on a comparison result, the query mode including a query mode of storage-computation separation or a MPP architecture. The present invention solves the technical problem that in an ultra-high-dimensional environment, how to enable a pre-computation query system to most efficiently and more stably utilize the pre-computation result to respond to the query of a client most quickly and avoid generating a large amount of redundant data.