Database and method for evaluating data therefrom

Abstract

A computer system for use in determining the minimum and / or maximum value of a specified subsequence of a sequence of values. The computer system comprises a database that is configured to store a sequence of values, to identify a minimum and / or maximum value of the sequence of values, and to define at least a first subsequence of values and a second subsequence of values immediately following the first subsequence of values, wherein the boundary between the first and second subsequences is located at the position of either the minimum or maximum value of the sequence.

Description

BACKGROUNDTechnical Field[0001]The present disclosure relates to computer-implemented querying mechanisms for obtaining and / or evaluating data from a database, in particular to querying mechanisms for obtaining minimum or maximum values from a sequence of numerical values stored in the database.Description of the Related Art[0002]A database is a structure for storing and relating data within e.g., a computer system. Different database architectures exist depending on the intended usage. The primary use for general purpose databases is to manage and facilitate data entry and retrieval in relation to the relevant application. A recent trend has been the emergence of specialized database architectures optimized to work with specific application domains.[0003]Complex event processing (CEP) is a technology for low-latency filtering, correlating, aggregating and / or computing on real-world event data, e.g., financial data. Such data is usually generated at high frequencies and so needs to ...

AI Technical Summary

Benefits of technology

[0057]Information in data servers must be protected from network borne attacks [H. Li and S. Lee, Mining frequent itemsets over data streams using efficient window sliding techniques, Expert Systems with Applications, Volume 36, Issue 2, Part 1, March 2009, Pages 1466-1477], [W. Wang, X. Guan and X. Zhang, Processing of massive audit streams for real-time anomaly intrusion detection, Computer Communications, Volume 31, Issue 1 (January 2008), Pages: 58-72], The two basic approaches consist of signature based detection where malicious behavior is detected by matching against previously stored prototypical attacks and anomaly detection where the profile of a normal user is maintained and an attempt is made to identify unacceptable deviation as a possible attack. The signature based approach can be adapted to make use of the invention's distance measurement mechanism where distances to a predefined set of signatures are recorded in real-time. The unique advantage of the invention is that not only distances to the complete signature are obtainable, but also distances to signature subsets are possible. This allows for quick experimentation and testing capabilities that do not require time consuming distance re-computation. Anomaly detection also benefits from the ability to perform calculations over data subsets. This is especially suitable for automated calibration approaches where multiple time periods can be measured simultaneously with no additional computation and vastly reduced bandwidth costs compared to a conventional approach.

[0058]Monitoring technology is used to track the performance of mass damping systems for tall concrete structures to enhance their damping and maintain their safety [J. M. W. Brownjohn, E. P. Carden, C. R. Goddard and G. Oudin, Real time performance monitoring of tuned mass damper system for a 183 m reinforced concrete chimney. Journal of Wind Engineering and Industrial Aerodynamics Vol. 8, No. 3, March 2010, pp. 169-179], Accelerometers are connected to the structure and remotely monitored to provide real-time information to determine whether the structures displacement exceeds critical thresholds. A large-scale application of this monitoring technology over hundreds or even thousands of structures would require a unified data management system such as the invention in order to allow all users to perform their required monitoring tasks simultaneously. With the compute-once, use-by-many approach, it is possible to efficiently have many automated systems monitoring different phenomena simultaneously through the use of different sized windows and data combinations. In addition, the low bandwidth requirement resulting from the cumulative data storage structure means that the monitoring sites can be remotely located with little overhead communications cost.Drilling Optimization

[0059]The goal is to optimize the drilling process to minimize cost while maintaining operational safety standards. This is done by continuously assessing the drilling process throughout the operation such that the overall rate of penetration is maximized based on cumulative footage drilled. The basic decision that must be made is to compromise between maximizing the utilization of the drill head and minimizing project down time due to a damaged drill head and the associated lead time to replace and order new parts. This is performed by continuously analyzing several variables such as weight on bit and rotary speed, applying such data to mathematical models and making a decision as late as possible to maximize bit head usage while not endangering progress of the drilling project. An advantage of a solution based on the invention is that time windows can be customized at any time with a negligible computational penalty. This is important when drilling through different geological formations, varying the drilling speed or even real-time calibration of models as work progresses and ensures the ability to perform comprehensive analysis regardless of the initial strategy. This also allows for evaluating several different models simultaneously using the same initial data. In addition, the low bandwidth requirement is an important feature especially when drilling in harsh remote areas where it is difficult to locate analysts and acquiring large bandwidth communication channels is often cost prohibitive.Scientific Data AnalysisEarthquake Prediction

[0060]Early warning devices can provide a short warning before the arrival of major tremors based on the different propagation speeds of the various vibrations produced. This application is distinguished by large amounts of continuous data that need near instantaneous processing in order to be of value. Measurement centers can also make use of readings from neighboring centers in their effort to increase the detection time before an earthquake. Due to the large amount of data generated by each measurement device, this is prohibitively expensive for all but the largest centers. However, with the use of the invention as the data storage backbone, hundreds or even thousands of neighboring centers can share data making use of the low bandwidth requirement of the invention. In addition, the negligible calculation overhead for different sliding window sizes means that multiple levels of detection can be simultaneously performed in order to detect different size anomalies. This is important for experimentation since a large number of potential models can be simultaneously machine tested and potentially applied.Tropical Atmosphere Ocean

[0061]In-situ environmental sensors are physically located in the environment they are monitoring and their time-series data is transmitted continuously to a single data repository [D. J. Hill and B. S. Minsker, Anomaly detection in streaming environmental sensor data: A data-driven modeling approach, Environmental Modelling &Software, Volume 25, Issue 9, September 2010, Pages 1014-1022], There is a need for automated data quality assurance and control to detect and identify anomalous data that deviates significantly from historical patterns. Such anomaly detection can also be used in the field of adaptive monitoring where anomalous data indicates phenomena that may require further investigation. The invention can be used as the main data depository where the customizable sliding windows mechanism can be used to perform simultaneous data assurance tests, thereby adding an additional layer of sophistication for quality monitoring. In addition, the same mechanism can be used to simultaneously detect multiple abnormal phenomena or allow for straight forward changes in detection strategy with a no additional computational cost. An added advantage is the ability to serve other remote users with their data requests with only a minimal communications investment due to the low bandwidth data transmission mechanism.DETAILED DESCRIPTION

[0062]The following explanation and examples demonstrate how a database can be adapted to allow the minimum value in a specified data string to be identified more easily. The maximum value in a specified data string may be identified in an analogous manner. The algorithms used to construct and query the database are set out below.

Problems solved by technology

While these databases are well-suited to performing traditional transaction-oriented operations, they do not provide an efficient means for allowing large amounts of contiguous data to be accessed and / or evaluated.

Thus, the operation is costly in terms of the I / O bus usage and / or network bandwidth utilization in retrieving the dataset, and in terms of the computation required to evaluate the dataset.

These costs will increase as the number of values in the requested data string increases.

In particular, comparative experimentation may be extremely costly, due to the cost of retrieving and evaluating a number of individual data sequences.