JSON_Partial_Update

This file describes how the performance tests in the paper "The Advantage of Partial Updates for Modification of JSON Data" can be reproduced. (All necessary files can be found in the zip file called "JSON_Partial_Update").Creation of the tables<br>First, the four tables must be created. The corresponding SQL statements can be found in the file "!Figure2a.docx" (see cases CT1 till CT4).Initial load<br>Second, the data must be loaded in these four tables. To load data into the "large" tables, use the SQL statements given in the file "Figure2a.docx" (see cases I3 and I4). To load the tables table_text_30K and table_json_30K with data, execute the statements given in the files INSERT_table_text and INSERT_table_json, respectively. To reach the size of the tables used in the paper, execute each INSERT statement 25 times.Different UPDATE statements<br>All UPDATE statements used for performance measurement are given in the files "!Figure2a.docx" and "!Figure2b.docx". The paper describes which UPDATE statement is used for a particular measurement. For instance, to measure the performance of the topic "Insert an Array's Element" (section 5.2.7), the UPDATE statements in IA1 till IA4 should be used. <br>

本文件详述了如何复现论文《JSON数据修改的部分更新优势》（The Advantage of Partial Updates for Modification of JSON Data）中的性能测试。所有必要文件均可在名为"JSON_Partial_Update"的压缩包中获取。

表格创建
首先需创建四张数据表，对应的SQL语句可在文件"!Figure2a.docx"中找到（对应案例CT1至CT4）。

初始数据加载
其次需向上述四张数据表中导入数据。针对"大型"数据表，可使用文件"Figure2a.docx"中给出的SQL语句（对应案例I3与I4）完成数据导入。若需向table_text_30K与table_json_30K两张数据表导入数据，则需分别执行文件INSERT_table_text与INSERT_table_json中的语句。为达到论文中使用的表规模，需将每条INSERT语句重复执行25次。

性能测试所用更新语句
所有用于性能测试的UPDATE语句均收录于文件"!Figure2a.docx"与"!Figure2b.docx"中。论文中说明了每项性能测试对应的UPDATE语句。例如，若需测试"插入数组元素"主题的性能（5.2.7节），则应使用IA1至IA4对应的UPDATE语句。