Font Size: a A A

Design And Implementation Of Test Data Generation Tool For Graph Database System

Posted on:2020-08-10Degree:MasterType:Thesis
Country:ChinaCandidate:K C PengFull Text:PDF
GTID:2428330590483220Subject:Computer technology
Abstract/Summary:PDF Full Text Request
In the era of big data,it has become a trend to use graph database system to represent and store multi-source and heterogeneous linked data.The graph database product GDM requires a large amount of real data for system testing before going online,but it is impossible to get large-scale real data,and most of the current test data generation tools generate random data,without considering the characteristics of the relationship between graph data and the simulation of generated data,which can not meet the test requirements of GDM.So the test data generation tool for graph database system is designed and implemented.The test data generation tool for graph database system implemented can generate similar large-scale test graph data based on small-scale business data provided by users,that is,the frequencies of simple patterns in the generated graph is consistent with the corresponding proportion of the frequencies in the initial small graph.The tool consists of five modules: business data graph data modeling,statistical analysis of small-scale graph data,expansion and transformation of statistical information,large-scale graph data generation and verification of generated graph data.Modeling module removes invalid data from small-scale business data and extracts key attribute columns to construct vertex and edge data to construct small graph data;Statistical analysis module makes statistics on the indicators of the constructed small graph data,obtains the data and the pattern distribution information of the small graph data,then fits them;Extended transformation module expands statistical information of small graph data according to user needs to construct statistical information of large graph data;Large graph data generation module constructs vertex data and edge data of large graph data in turn based on statistical information of large graph data;Finally,the data verification module verifies the generated large-scale test data,checks whether it meets the user's expected scale,and maintains similarity with the small graph to ensure the rationality and generality of the tool.On the basis of the design and implementation,the feasibility test and data similarity verification of the test data generation tool for graph database system are carried out.The test results show that the tool can generate test graph data of user's required scale and satisfy the similarity of the data at the same time.Therefore,the system can meet the expected design principles and related requirements.
Keywords/Search Tags:Graph Database, Test Data, Data Generation, Similarity of Graph Data
PDF Full Text Request
Related items