The present disclosure provides a method of reading data maintained in a tree data structure, such as B+ tree, using near data processing (NDP) in a cloud native database. According to embodiments, a desired LSN will be used in NDP page reads on the master computing node (e.g. master SQL node). When the master computing node (e.g. master SQL node) reads the regular page, the maximum desired LSN (e.g. the latest page version number) for that regular page will be used. Embodiments use features of the desired LSN and page locking, wherein correct versions of pages can be obtained by using the desired LSN associated with a page, in combination with page locking, and can enable the reading of a consistent tree structure and achieve good read/write concurrency.

Provided are a method for processing model parameters, and an apparatus. The method comprises: a model parameter set to be sharded is obtained, wherein the model parameter set comprises a multi-dimensional array corresponding to a feature embedding; attribute information for a storage system used for storing the model parameter set to be sharded is obtained, wherein the storage system used for storing the model parameter set to be sharded differs from a system on which a model corresponding to the model parameter set to be sharded is located when operating; the model parameter set to be sharded is stored in the storage system according to the attribute information.

A computer-implemented method can measure query locality during execution of a plurality of incoming queries in a database management system. The database management system includes a query execution plan cache which has a size that can store at least some of query execution plans generated for the plurality of incoming queries. Based on the measured query locality, the method can adjust the size of the query execution plan cache.

Techniques are described herein for prediction of an buffer pool size (BPS). Before performing BPS prediction, gathered data are used to determine whether a target workload is in a steady state. Historical utilization data gathered while the workload is in a steady state are used to predict object-specific BPS components for database objects, accessed by the target workload, that are identified for BPS analysis based on shares of the total disk I/O requests, for the workload, that are attributed to the respective objects. Preference of analysis is given to objects that are associated with larger shares of disk I/O activity. An object-specific BPS component is determined based on a coverage function that returns a percentage of the database object size (on disk) that should be available in the buffer pool for that database object. The percentage is determined using either a heuristic-based or a machine learning-based approach.

An accelerator is disclosed. The accelerator may include an on-chip memory to store a data from a database. The on-chip memory may include a first memory bank and a second memory bank. The first memory bank may store the data, which may include a first value and a second value. A computational engine may execute, in parallel, a command on the first value in the data and the command on the second value in the data in the on-chip memory. The on-chip memory may be configured to load a second data from the database into the second memory bank in parallel with the computation engine executing the command on the first value in the data and executing the command on the second value in the data.

A communications system implements instructions including maintaining a message object that includes an array of entries. Each entry of the array includes a field identifier, a data type, and a next entry pointer. The next entry pointers and a head pointer establish a linked list of entries. The instructions include, in response to a request to add a new entry to the message object, calculating an index based on a field identifier of the new entry and determining whether the entry at the calculated index within the array of entries is active. The instructions include, if the entry is inactive, writing a data type, field identifier, and data value of the new entry to the calculated index, and inserting the new entry into the linked list. The instructions include, if the entry is already active, selectively expanding the size of the array and repeating the calculating and determining.

A method includes determining, in accordance with a first ordering, a plurality of read requests for a memory device. The plurality of read requests are added to a memory device queue for the memory device in accordance with the first ordering. The plurality of read requests in the memory device queue are processed, in accordance with a second ordering that is different from the first ordering, to determine read data for each of the plurality of read requests. The read data for the each of the plurality of read requests is added one of a set of ordered positions, based on the first ordering, of a ring buffer as the each of the plurality of reads requests is processed. The read data of a subset of the plurality of read requests is submitted based on adding the read data to a first ordered position of the set of ordered positions of the ring buffer.

One example method includes receiving data, partitioning the data according to their respective similarity groups, and the similarity groups collectively define a range of similarity groups, deduplicating the data after the partitioning, packing unique data segments remaining after deduplicating into one or more compression regions, compressing the compression regions, and writing an object, that includes the compression regions, to a durable log. The deduplicating and compressing for a similarity group may be performed by a dedup-compression instances uniquely assigned to that similarity group.

Disclosed are some implementations of systems, apparatus, methods and computer program products for executing a process flow represented by a graph or portion thereof using cached subgraphs. A first request to execute a first portion of a process flow is processed, where the first portion of the process flow is represented by a first subgraph of a graph representing the process flow and a final node of the first subgraph corresponds to a set of computer-readable instructions. The first portion of the process flow is executed such that a first output of executing the first portion of the process flow is obtained. The first subgraph is stored in association with the first output in a first cache entry of a cache. A second request to execute a second portion of the process flow is processed, where the second portion of the process flow is represented by a second subgraph of the graph. At least one cache entry for which a corresponding subgraph matches at least a portion of the second subgraph is identified in the cache, where the at least one cache entry includes the first cache entry. The first output is retrieved from the first cache entry, a node of the second subgraph to which the final node of the first subgraph is connected is identified, and the second portion of the process flow is executed by providing the first output as input to the identified node of the second subgraph without executing the set of computer-readable instructions.

Disclosed are methods, systems, and computer-readable medium for providing report results. Viscous attributes and non-viscous may be identified. A smart cube may be received and may include viscous values for the viscous attributes. The smart cube may be stored at a local cache. A report associated with an organization may be initiated. A runtime generation of the report may be generated based on initiating the report. The report may call a viscous attribute from the viscous attributes and call a non-viscous attribute from the non-viscous attributes. The runtime generation may be modified to remove the viscous attribute from the runtime generation. A viscous value for the viscous attribute may be retrieved from the smart cube at the local cache. The modified runtime generation may be executed to retrieve a non-viscous value for the non-viscous attribute from a remote database and a report result may be provided.