Browse Prior Art Database

Dynamically Verifying Query Results

IP.com Disclosure Number: IPCOM000240270D
Publication Date: 2015-Jan-20
Document File: 3 page(s) / 37K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method and system is disclosed for dynamically verifying query results. The method and system stores temporarily a master copy of the query result from an initial execution and compares query results from subsequent execution with the master copy to validate correctness of the query results.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 50% of the total text.

Page 01 of 3

Dynamically Verifying Query Results

Typically, query results are verified by comparing an expected result with results obtained in each execution. The expected result is captured after manual examination and added to a validation process for future executions. The manual process may provide exanimate output which may be laboratory intensive and inflexible. When a variation is introduced to existing queries, the manual examination is required to be repeated.

Disclosed is a method and system for dynamically verifying query results. The method and system stores temporarily a master copy of the query result from an initial execution. Thereafter, the method and system compares query results from subsequent execution with the master copy to validate correctness of the query results.

The below figure illustrates a flow diagram to execute workload of a method and system.

Figure

The disclosed method and system provides an execution framework which is set up to execute queries repeatedly and monitor the results for identifying unexpected difference. The execution starts from a routine which drives the execution of queries. Thereafter, the execution framework validates the result repeatedly when variations are fed to a target Database Management System (DBMS) in each execution. The queries result from the initial execution is stored temporarily as a master copy, and results from subsequent execution are compared with the master copy to validate correctness. The execution is carried out repeatedly and the execution framework introduces variation

1


Page 02 of 3

through different system parameters, queries execution preferences, performance adjustment, and different versions of the DBMS. The variations may not inject difference into queries results, in which case the execution framework executes the same workload against numerous sets of data with various differences. Thus, the

workload can be scaled into using different data sources against a set of queries or different sets of queries against data sources.

The execution framework can be used to execute various queries workloads against a DBMS or different DBMS's. At the initiation stage, database objects are created and populated with any set of predefined data. All the workloads are executed at least once and the queries result is captured in the framework as the master copy. Once the initiation stage is completed, the framework issues predefined adjustment against the DBMS in an adjustment phase. The predefined adjustment against the DBMS is performed one at a time. The adjustments against DBMS are important and are carefully chosen.

Every DBMS uses an algorithm to create an access path to filter and retrieve data for a specific query. The algorithm uses many factors to decide the optimal sequence while retrieving data. The factors can be statistics of data or available indexes. Due to change in the statistics of data, the DBMS data distribution is changed. There are opportunities for savi...