SolarWinds has been granted a patent for a method, computer program-product, and system that generates predicate-weighted histograms in a database management system. The invention aims to improve query performance by generating histograms with greater resolution near predicate literals of the queries submitted to the system. The method involves initializing a working histogram, capturing queries with predicate literals, generating a weight vector, calculating cost values for each row of the histogram, identifying the row with the lowest cost value, and merging rows based on information loss. GlobalData’s report on SolarWinds gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
According to GlobalData’s company profile on SolarWinds, AI assisted network mgmt was a key innovation area identified from patents. SolarWinds's grant share as of June 2023 was 1%. Grant share is based on the ratio of number of grants to total number of patents.
Generating predicate-weighted histograms in a database management system
A recently granted patent (Publication Number: US11669519B2) describes a method for generating a histogram. The method involves initializing a working histogram for a data column, where each row in the histogram corresponds to a unique data value. The method captures queries that include predicate literals and generates a weight vector based on these literals. The weight vector consists of weight values corresponding to each row in the histogram.
The method then calculates a cost value for each row in the histogram, taking into account information loss and the weight value associated with the row. The cost value helps identify the row with the lowest cost value in the histogram. This row is then merged with another row in the histogram. The information loss is calculated as the entropy value of the merged row minus the sum of the entropy of the two original rows.
If the number of rows in the working histogram exceeds a predetermined threshold, the method repeats the steps of generating the weight vector, calculating the cost value, identifying the row with the lowest cost value, and merging rows until the total number of rows in the histogram is below the threshold.
The patent also describes comparing the distribution of weight values to the distribution of rows in the histogram. This comparison involves performing a statistical non-parametric test to generate a comparison score, which is then compared to a predetermined threshold score.
The method further includes capturing queries that meet or exceed a predetermined threshold, such as a runtime or CPU time for the queries. It also involves generating a kernel density function for the predicate literals and evaluating this function at each row in the histogram. The bandwidth for the kernel density function is proportional to the variance in the predicate literals.
The patent also mentions calculating various penalties and parameters, such as a range penalty, information loss, legacy information loss, and scaling parameter. These factors contribute to the cost function used in determining the cost value for each row in the histogram.
Overall, this patent presents a method for generating histograms that takes into account various factors, including information loss, weight values, and distribution comparisons. The method can be implemented using computer-readable program code and can be executed on a processing device within a system for generating histograms.
To know more about GlobalData’s detailed insights on SolarWinds, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
