Abstract: Disclosed herein is a method for determining whether to apply a lazy materialization technique to a query run. A data processing service receives a request to perform a query identifying a filter column and a non-filter column in a columnar database. The data processing service accesses a first task of contiguous rows in the filter column from a cloud-based object storage. The data processing service applies a filter defined by the query to the first task. The data processing service generates filter results for the first task that may include a percentage of the first task discarded and a run-time. The data processing service determines, based on the filter results for the first task, a likelihood value that indicates a likelihood of gaining a performance benefit by applying the lazy materialization technique to a second task of the query.

Abstract: A system includes an interface, a processor, and a memory. The interface is configured to receive a version of a model from a model registry. The processor is configured to store the version of the model, start a process running the version of the model, and update a proxy with version information associated with the version of the model, wherein the updated proxy indicates to redirect an indication to invoke the version of the model to the process. The memory is coupled to the processor and configured to provide the processor with instructions.

Abstract: A system for multipass sort includes a communication interface and a processor. The communication interface is configured to receive from a client device a request to sort a dataset that includes a plurality of rows. The processor is configured to perform a first sort pass on the dataset in part by: extracting prefixes associated with a first schema element associated with the dataset for the plurality of rows; and sorting the extracted prefixes utilizing an integer sort algorithm based on a sort order included in the request to sort the dataset, where sorting the extracted prefixes includes utilizing NULL values to resolve a tied range that includes at least two rows of the plurality of rows having a same extracted prefix.

Abstract: A data processing service performs a rebalancing process for rebalancing stateful tasks on a cluster computing system. In one instance, the method for rebalancing stateful tasks is performed such that the per-operator partitions are spread across available executors of a cluster of the cluster computing system with respect to one or more statistics of the tasks. In one instance, the method for rebalancing stateful tasks is also performed such that the total number of stateful tasks are balanced per executor as long as this rebalancing does not imbalance the per-operator placements. In this way, the processing of stateful tasks can be spread across multiple executors in a relatively uniform manner, even though there may be an upfront cost of breaking the local caching on an executor.

Abstract: The interface is to receive an indication to execute an optimize command. The processor is to receive a file name; determine whether adding a file of the file name to a current bin causes the current bin to exceed a threshold; associate the file with the current bin in response to determining that adding the file does not cause the current bin to exceed the bin threshold; in response to determining that adding the file to the current bin causes the current bin to exceed the bin threshold: associate the file with a next bin, indicate that the current bin is closed, and add the current bin to a batch of bins; determine whether a measure of the batch of bins exceeds a batch threshold; and in response to determining that the measure exceeds the batch threshold, provide the batch of bins for processing.

Abstract: A data tree for managing data files of a data table and performing one or more transaction operations to the data table is described. The data tree is configured as a KD-epsilon tree and includes a plurality of nodes and edges. A node of the data tree may represent a splitting condition with respect to key-values for a respective key. A leaf node of the data tree may correspond to a data file for a data table that includes a subset of records having key-values that satisfy the condition for the node and conditions associated with parent nodes of the node. A parent node may correspond to a file including a buffer that stores changes to data files reachable by this parent node, and also includes dedicated storage to pointers of the child nodes. By using the data tree, the data processing system may efficiently cluster the data in the data table while reducing the number of data files that are rewritten.

Abstract: The present application discloses a method, system, and computer system for managing a data in a storage system. The method includes receiving a first transaction that modifies or deletes first data stored in a storage system, determining that the first data is subject to an intervening re-arrangement transaction, and in response to determining that the first data is subject to the intervening re-arrangement transaction, rolling back the re-arrangement transaction at least with respect to the first data and committing the first transaction.

Abstract: The present application discloses a method, system, and computer system for parsing files. The method includes receiving an indication that a first file is to be processed, determining to begin processing the first file using a first processing engine based at least in part on one or more predefined heuristics, indicating to process the first file using a first processing engine, determining whether a particular error in processing the first file using the first processing engine has been detected, in response to determining that the particular error has been detected, indicate to stop processing the first file using the first processing engine and indicate to continue processing using a second processing engine, and storing in memory information obtained based on processing the first file by one or more of the first processing engine and the second processing engine.

Abstract: A system for clustering data into corresponding files comprises one or more processors and a memory. The one or more processors is/are configured to: 1) determine to cluster a set of data into a set of files; 2) determine a set of split points in a corresponding set of dimensions of the set of data to determine the set of files, wherein each file of the set of files has an approximate target size; and 3) store one or more items of the set of data into a corresponding file of the set of files based at least in part on the set of split points. The memory is coupled to the one or more processors and configured to provide the processor with instructions.

Abstract: A method, system, and computer system for performing an operation with respect to a target table are disclosed. The method includes performing first and second jobs, obtaining one or more other resulting files based at least in part on unmatched rows, and obtaining a set of processed files based at least in part on performing a post-processing operation with respect to the set of resulting files. The set of processed files has less files than the set of resulting files. Performing the first job includes determining a set of matching target table files and storing target table information indicating for each of the set of matching target table files, a particular set of rows having matching rows. Performing the second job includes performing a matching action based on matched rows and obtaining the second job resulting file(s).

Abstract: A method, system, and computer system for performing an operation with respect to a target table are disclosed. The method includes performing first and second jobs, and persist, in one or more deletion vector files, one or more deletion vectors for corresponding rows of the one or more target table files, and obtaining a resulting table based at least in part on the second job resulting file(s). Performing the first job includes determining a set of matching target table files and storing target table information indicating for each of the set of matching target table files, a particular set of rows having matching rows. Performing the second job includes performing a matching action based on matched rows and one or more deletion of vectors associated with previously removed rows of the matching target table files and obtaining the second job resulting file(s).

Abstract: The present application discloses a method, system, and computer system for building a model associated with a dataset. The method includes receiving a data set, the dataset comprising a plurality of keys and a plurality of key-value relationships, determining a plurality of models to build based at least in part on the dataset, wherein determining the plurality of models to build comprises using the dataset format information to identify the plurality of models, building the plurality of models, and optimizing at least one of the plurality of models.

Abstract: A system for executing a streaming query includes an interface and a processor. The interface is configured to receive a logical query plan. The processor is configured to determine a physical query plan based at least in part on the logical query plan. The physical query plan comprises an ordered set of operators. Each operator of the ordered set of operators comprises an operator input mode and an operator output mode. The processor is further configured to execute the physical query plan using the operator input mode and the operator output mode for each operator of the query.

Abstract: A system for dataflow graph processing comprises a communication interface and a processor. The communication interface is configured receive an indication to generate a dataflow graph, wherein the indication includes a set of queries and/or commands. The processor is coupled to the communication interface and configured to: determine dependencies of each query in the set of queries on another query; determine a DAG of nodes based at least in part on the dependencies; determine the dataflow graph by determining in-line expressions for tables of the dataflow graph aggregating calculations associated with a subset of dataflow graph nodes designated as view nodes; and provide the dataflow graph.

Abstract: A system for dataflow graph processing comprises a communication interface and a processor. The communication interface is configured receive an indication to generate a dataflow graph, wherein the indication includes a set of queries. The processor is coupled to the communication interface and is configured to: determine dependencies of each query in the set of queries on another query; determine a DAG of nodes based at least in part on the dependencies; insert a node in the DAG of nodes to generate an updated DAG to enforce an expectation; determine a dataflow graph based on the updated DAG; and provide the dataflow graph.

Abstract: The system is configured to: 1) receive a client request; 2) determine executor(s) to generate a response to the user request; 3) provide each of the executor(s) with an indication; 4) receive for each indication a response including an output of either a cloud output or an in-line output to generate a group of in-line outputs and a group of cloud outputs; 5) determine whether the group of in-line outputs comprises all outputs; and 6) in response to the group of in-line outputs not comprising all the outputs for the client request: a) convert the group of in-line outputs to a converted group of cloud outputs; b) generate metadata for the converted group of cloud outputs and the group of cloud outputs; and c) provide response to the client request including the metadata for the converted group of cloud outputs and the group of cloud outputs.

Abstract: A function creation method is disclosed. The method comprises defining one or more database function inputs, defining cluster processing information, defining a deep learning model, and defining one or more database function outputs. A database function is created based at least in part on the one or more database function inputs, the cluster set-up information, the deep learning model, and the one or more database function outputs. In some embodiments, the database function enables a non-technical user to utilize deep learning models.

Abstract: A system comprises an interface, a processor, and a memory. The interface is configured to receive a query. The processor is configured to: determine a set of nodes for the query; determine whether a node of the set of nodes comprises a first engine node type or a second engine node type, wherein determining whether the node of the set of nodes comprises the first engine node type or the second engine node type is based at least in part on determining whether the node is able to be executed in a second engine; and generate a plan based at least in part on the set of nodes. The memory is coupled to the processor and is configured to provide the processor with instructions.

Abstract: A system includes an interface, a processor, and a memory. The interface is configured to receive a version of a model from a model registry. The processor is configured to store the version of the model, start a process running the version of the model, and update a proxy with version information associated with the version of the model, wherein the updated proxy indicates to redirect an indication to invoke the version of the model to the process. The memory is coupled to the processor and configured to provide the processor with instructions.

Abstract: A system includes an interface and a processor. The interface is configured to receive a table indication of a data table and to receive a transaction indication to perform a transaction. The processor is configured to determine a current position N in a transaction log; determine a current state of the metadata; determine a read set associated with a transaction; attempt to write an update to the transaction log associated with a next position N+1; in response to a transaction determination that a simultaneous transaction associated with the next position N+1 already exists, determine a set of updated files; and in response to a determination that there is not an overlap between the read set associated with the current transaction and the set of updated files associated with the simultaneous transaction, attempt to write the update to the transaction to the transaction log associated with a further position N+2.