97 results about "Fact table" patented technology

Techniques for eliminating one or more portions of a data object from any join step of an operation that joins multiple data objects include determining that an operation joins a first data object and a second data object. The second data object includes multiple portions. Each of multiple data units of the first data object is scanned. Based on data in the data units of the first data object, information is generated. The information indicates a portion of the second data object for exclusion. The indicated portion is excluded from an output of the operation. Only one or more portions of the second data object that are not indicated for exclusion in the information are included in a particular join step involving the second data object. By pruning a large second table, such as a fact table, the computational resources consumed by the joins are substantially reduced.

A system and method of determining an earliest delivery date for a demand on the basis of a fact table stored in a database, where the demand indicates both a desired date and a desired quantity for items from a stock, where a fact time series is comprised by and / or deducible from the fact table, and where the fact time series comprises fact tuples indicating both a fact date and a change in quantity of stock items that occurred on this fact date. The method includes generating an aggregated time series from the fact time series by defining and / or obtaining discrete time intervals, and generating, for at least one discrete time interval, an aggregated tuple indicating both its corresponding discrete time interval and the aggregated change in quantity of stock items that occurred in its corresponding discrete time interval.

Techniques are described for optimizing a multi-fact star schema query to retrieve data from two or more fact tables of a database. The optimization techniques may include internally transforming a received multi-fact star schema query into a group of single-fact sub-queries, thereby reducing or eliminating certain constraints during query processing. The group of sub-queries may be enumerated to determine a query execution plan. The multi-fact star schema query may be executed according to the query execution plan to retrieve the data from the two or more fact tables of the database.

Methods, systems, and articles of manufacture for constructing and populating data marts with dimensional data models from a set of data repositories that contain factual and association information about a set of related assets are disclosed. An intermediate data warehouse is generated to process the facts and associations for each asset. Using the intermediate warehouse, one or more data marts are generated with fact tables, dimensions, and hierarchies to fully model the information available for each asset.

A fact table is set up without any dimension columns. The data in the table is ordered according to an order corresponding to the ordering of the dimensions. The table is divided into segments corresponding to contiguous portions of data. The portions of the table containing non-null values for the measures are placed into the segments. The size of the segments and the location segments may be determined according to the density of discontinuities in the data. An indexed organized table is used to determine the beginning and end of each segment of the table, the locations of the discontinuities in the data, and to facilitate accessing the measures of the table.

The invention discloses a power grid resource fault diagnosis method based on association rules. The method comprises the following steps: A, drawing off fault data, converting the fault data, loading the fault data to a data warehouse, and establishing a multidimensional data model and a fault fact table; B, performing cluster analysis on the fault data; C, performing association rule analysis on the fault data; and D, abstracting actual fault events as nodes, if it is compared that a high-credibility fault power grid device accords with a calculated aggregation degree, obtaining a diagnosis result, and giving assistance advice for processing of power grid faults. According to the invention, correlation data of power equipment can be comprehensively acquired, the multidimensional data model and the fault fact table can be established, a data base is provided for data mining, data is preliminarily analyzed by use of the cluster analysis, data association analysis is performed by integrating various data, comparative determining is carried out, integration of multiple intelligent criteria is formed, the faults are accurately determined, and the application prospect is good.

The invention discloses an OLAP (On-Line Analytical Processing) star connection query optimizing method under a CPU (Central Processing Unit) and GPU (Graphics Processing Unit) mixing framework. The method comprises the following steps: first filtering and optimizing an OLAP star connection operation through bitmap connection indexing, buffering high-frequently accessed connecting bitmap in a GPU buffer memory, next, loading fact table external key attribute group satisfying the connecting bitmap filtering condition into the GPU buffer memory to carry out star connection filtering, and finally switching the whole table scanning of the big fact table of the internal memory into accessing randomly according to the position through filtering bitmap generated by the GPU, thereby improving the querying and treating performances of the OLAP star connection. The method improves the storage efficiency of the GPU buffer memory and the parallel processing efficiency of the GPU, and improves the OLAP querying and treating performance of the hybrid processor platform on the whole.

A method of data migration and integration with a data load mechanism. The first primary step is to extract the data from a named source by remote queries. The follow-up logic is used for any miscellaneous or supplemental transformations to inbound data for data renaming, null scrubbing, and data polishing. The filtration step allows for any extraneous data to be filtered out. The compression step consolidates any fragmented balances residual in the remote queries, localize remote data, data map, or follow-up logic. As a part of the compression process, the result set table signature is explicitly cast to that of the destination fact table. The result of this is an exact replica of the destination fact table format, data types and column order. The kickout handling provides referential integrity of the inbound data flow against the named BPM Destination System. Finally, the deployment step limits the dynamic reload or sync determination to certain fields.

A fact table is set up without any dimension columns. The data in the table is ordered according to an order corresponding to the ordering of the dimensions. The table is divided into segments corresponding to contiguous portions of data. The portions of the table containing non-null values for the measures are placed into the segments. The size of the segments and the location segments may be determined according to the density of discontinuities in the data. An indexed organized table is used to determine the beginning and end of each segment of the table, the locations of the discontinuities in the data, and to facilitate accessing the measures of the table.

The invention provides a grid fault diagnosis method based on data mining. The method comprises the steps of establishing a multidimensional data model and a fault fact table, conducting clustering analysis on fault data, conducting association rule analysis on the fault data, conducting Bayesian network establishment and training on the fault data, and obtaining a result. By means of the method, the problem of multiple sources of grid fault data is solved, data marts with corresponding themes are established according to different requirements, and therefore data retrieval is more convenient. Furthermore, data are analyzed by means of the three classic algorithms in data mining, namely clustering analysis, association rules and the Bayesian network, unknown knowledge is finally discovered through comprehensive analysis of the results of the three algorithms, and then auxiliary suggestions are provided for grid fault handling.

The invention discloses a method and device for generating a report template. The method comprises the following steps of: determining a fact table and counting indexes to be counted of the report template, and selecting a counting condition; associating the counting condition with the fact table by using a snowflakes model; and sequencing and grouping the counting indexes in the fact table according to the counting condition, and dynamically memorizing sequencing and grouping results to generate the report template. In the invention, the counting condition is associated with the fact table to be counted of the report template by using the snowflakes model, and the indexes to be counted in the fact table are sequenced and grouped according to the counting condition to dynamically generate the report template. Compared with the conventional fixed report, the invention has the advantages that: implementation is easier and more flexible, a report with a high custom degree can be provided without secondary development, and the development speed of the report template is greatly increased.

The invention provides a method and system for rapidly constructing (KPIs) of enterprises to solve the problem of heavy indicator development workloads when statistical caliber changes of indicators are frequent with more and more KPIs. The method comprises the steps of configuring a grouping configuration table and an indicator realization configuration table in advance; looking up a corresponding SQL-ID from the grouping configuration table according to identification parameters in a calling order; obtaining a corresponding SQL script in the indicator realization configuration table according to the SQL-ID, wherein the SQL script corresponding to the SQL-ID contains SQL statements of a plurality of indicators; finishing calculation of the indicators by analyzing and executing the statements in the SQL script, and writing calculation results of the indicators in a middle process table; generating data in the middle process table into a fact table. By means of the method and the system, the development efficiency of the indicators is improved, and workloads are reduced greatly.

A system and method of determining an earliest delivery date for a demand on the basis of a fact table stored in a database, where the demand indicates both a desired date and a desired quantity for items from a stock, where a fact time series is comprised by and / or deducible from the fact table, and where the fact time series comprises fact tuples indicating both a fact date and a change in quantity of stock items that occurred on this fact date. The method includes generating an aggregated time series from the fact time series by defining and / or obtaining discrete time intervals, and generating, for at least one discrete time interval, an aggregated tuple indicating both its corresponding discrete time interval and the aggregated change in quantity of stock items that occurred in its corresponding discrete time interval.

Certain embodiments contemplate systems and methods for improving the speed and efficiency of a data warehouse. In some embodiments, an ETL process is modified to perform a joined indexing operation which reduces the number of lookup requests required. Certain embodiments contemplate a date dimension and hierarchical data structure which improve operation speed. Still other embodiments contemplate structural organizations of biographical fact tables to better improve data access.

The invention discloses a data warehouse based financial data integration method. The data warehouse based financial data integration method includes the following steps that step one, a data warehouse data model is designed, data segmentation and data granularity and theme confirmation are performed, and a concept model, a logical model and a physical model are designed; step two, data integration is performed, namely data extraction, data cleaning, data conversion and data loading are performed; step three, data analysis is performed, a used fact table and dimension are selected and analyzed according to a created data source view, and a multi-dimensional dataset is created. The data warehouse based financial data integration method is a group enterprise financial data integration method which adopts an independent architecture, is quick and convenient to develop and use and capable of making data consistent and extendible and is based on a data warehouse and aims at establishing an analysis system oriented to the characteristics of a group enterprise and meeting the specific demands for financial and economic information analysis and management decision making of the group enterprise.

Load Plan Generator (LPG) is a BIAPPS utility for generating ODI load plans based on desired subset of fact tables for loading BIAPPS Data Warehouse. The tool simplifies the configurations process by minimizing the manual steps and configurations and provides a guided list of configurations steps and checklists. The load plan components are basically different sets of load plans that will be stitched together by the load plan generator to create one load plan for loading chosen fact groups in the warehouse sourcing from different transaction systems.

Methods, systems, and articles of manufacture for constructing and populating data marts with dimensional data models from a set of data repositories that contain factual and association information about a set of related assets are disclosed. An intermediate data warehouse is generated to process the facts and associations for each asset. Using the intermediate warehouse, one or more data marts are generated with fact tables, dimensions, and hierarchies to fully model the information available for each asset.

The invention discloses a wind power equipment fault diagnosis method and system. According to the invention, related data of wind power equipment is comprehensively collected, a data warehouse is built, the data integrity is ensured, a data basis is provided for data mining, fault data is extracted from historical data of the data warehouse, the fault data is converted and loaded to the data warehouse so as to build a multi-dimensional data model and a fault fact table, preliminary analysis is performed on the data by adopting clustering analysis, data correlation analysis is performed by integrating various types of data, comparative judgment is performed, integration of multiple intelligent criteria is formed, and wind power fault diagnosis example explanation is performed according to the intelligent criteria. The wind power equipment fault diagnosis method has the advantages of complete criterion, advanced method and accurate judgment, provides a new method for power equipment fault judgment, overcomes defects and disadvantages of a single method and improves the accuracy of fan equipment fault diagnosis.

A method for managing a logical statement within a distributed database includes checking, responsive to receipt of a first logical statement for by first database management system node, whether the first logical statement is stored within a segment of the distributed database; and storing, responsive to a determination that the first logical statement is not stored within a segment of the distributed database, storing the first logical statement in at least one of non-unique fact table and unique fact table.

The invention provides an intelligent analysis method and system based on large data acquisition. The method comprises the steps that user data are collected; the data attributes of a user in different channels are merged to form a user data fact table; data precipitation is carried out on the user data fact table; data statistics is carried out on the user data fact table; data analysis is carried out on the user data fact table; data mining is carried out on the user data fact table; and a multi-dimensional user portrait is generated according to the data acquired through data precipitation, data statistics, data analysis and data mining. According to the invention, data precipitation, data statistics, data analysis and data mining are used to acquire the natural attribute, the page view, the click rate, the classification weight, the short-term demand, the user group characteristic, the user tag and other data of the user based on the acquired and aggregated user network behavior data, and the data of the corresponding dimension in the user portrait are filled.

The embodiment of the invention discloses a data warehouse construction method and apparatus, and a data query method, apparatus and device. The data warehouse construction method comprises the steps:acquiring a physical fact table; obtaining and establishing an association relationship between the logic topic table and the physical fact table in the metadata according to the logic topic table configuration information; and obtaining and establishing an association relationship between the indexes and the logic topic table in metadata according to the index configuration information. The dataquery method is applied to the data warehouse constructed by the data warehouse construction method, and comprises the following steps: receiving a query request which comprises indexes; determininga corresponding logic topic table according to an association relationship between the index and the logic topic table in the metadata; determining a corresponding physical fact table according to theassociation relationship between the logic theme table and the physical fact table; and determining a query result of the index according to the content of the corresponding physical fact table.

The invention relates to a data processing method and a data processing device. The method includes the steps of first, configuring basic information and calculation rules of indexes, wherein the indexes comprise number indexes and formula indexes; second, automatically loading index configuration information in the above step and completing the calculation of the indexes to obtain results of the indexes. The data processing device comprises a configuration module and a running module capable of automatically loading the configuration module. The data processing method can efficiently change original data in a fact table into final index data, thereby improving the calculation efficiency of the index data. The data processing device is simple in structure and easy to operate.

The invention discloses a BMS data storage system and method based on a database. The system comprises a data collection module, a data classification module, a storage rule module, a basic table module and a fact table module. The data collection module is used for collecting data from all target sources, the data classification module is used for receiving the data and dividing the received data into different categories, the storage rule module is used for storing storage rule information corresponding to the data of all the categories and storage rule indexing information, the basic table module is used for storing the data of the different categories in a classified mode, and the fact table module is used for storing the storage states and storage position information of the data of the different categories. The BMS data collected on site are stored in a novel storage mode, and classified storage, reliable management, safety backup and other operation on the BMS data are completed by means of the powerful data processing tool (the database).

The invention discloses a data analyzing method based on an fMRI brain activation data warehouse. The method includes the following steps that a basic framework of the fMRI data warehouse is built, data extraction is carried out, and dimensions and a fact table are introduced; data processing and multi-dimensional analysis are carried out. Basic principles and technologies of the data warehouse are used for secondary processing of data in brain activation researches, and a multi-dimensional analysis technology is introduced into analysis and processing of result data of brain activation, so that secondary analysis, processing and integration can be carried out, new rules are found, and deep mechanisms of advanced brain activities are disclosed. Accordingly, scattered brain function imaging data and results in the prior art can be used orderly.

A method includes identifying a source fact table, where the source fact table includes a first set of dimensions and corresponds to a source multidimensional cube that is associated with a first set of measures. The method also includes identifying a mapped multidimensional cube, where the mapped multidimensional cube is associated with a second set of dimensions and a second set of measures. The method further includes adding, to the source fact table, a one or more mapped dimensions corresponding to the second set of dimensions. The method includes associating, with the source multidimensional cube, one or more mapped measures corresponding to the second set of measures.

An embodiment of the invention pertains to processing messages received from a set of devices. A mapping of message fields to fact table columns is stored in a mapping table of a relational database. The method comprises, for each received message, determining all fields of the received message. If it is determined that the mapping table does not assign a fact table column to each of the determined fields of the message, a new mapping between message fields to respective fact table is automatically stored in the mapping table. A mapping is identified in the mapping table that assigns each of the determined fields of the message to a respective one of the columns of the fact table. The field values of the message are stored in the fact table in accordance with the identified mapping.

The invention relates to a financial statement generation technology and aims to solve the problems of high workload and high error rate in a process of obtaining existing financial statement consolidation data. The invention provides an XBRL intelligent report platform based statement consolidation data extraction method. The method comprises the following steps: firstly, obtaining fixed dimensions and row dimensions from a statement configuration table, a statement row configuration table and a statement column configuration table, generating and executing a dynamic SQL, automatically retrieving data available for a current statement from a statement consolidation fact table, and inserting the data into a fact middle table; secondly, performing Cartesian join on the statement row configuration table and the statement column configuration table through statement IDs, generating a dynamic SQL of extracted data, inserting the dynamic SQL into a row-column dynamic SQL temporary table, obtaining path and fact value information, and inserting the path and fact value information into a fact value table; and finally, performing circulation on each statement to obtain complete fact value table data identifiable for an XBRL platform, and storing the complete fact value table data in a statement value table. The method is suitable for extracting statement consolidation data of the XBRL intelligent report platform.

The invention relates to a dimension abnormity aggregation method of an OLAP system, which is suitable for the situation in which non-mutual exclusion of indicators is present between dimension values in the OLAP system, in particular in the process of multidimensional analysis which generates a large amount of data and which is timely required. The invention generates gathering data of a fact table in a database in advance, dimension level codes of an OLAP tool is directed to well-gathered codes for association in order to achieve the requirement of generating correct gathering data. The beneficial effect of the invention is to thoroughly solve the problem that the current OLAP system performs indicators gathering too long in case that the dimension values are not mutually exclusive and guarantees timeliness of data, in addition, the invention is simple and easily practicable. The method has extremely excellent effect as represented by the application to IMEI analysis in Hebei Mobile.