Tag Archives: Big data processing

Summary and statistics of large amount of data

1. Try to implement group grouping in SQL. Although the database needs grouping operation, there is index in the database, so the operation speed is fast, and the number of records fetched to the report server side is greatly reduced, and the fetching speed is greatly accelerated. Therefore, when grouping operation is carried out on the report side, only a small number of records are needed, and the operation speed of the report is greatly accelerated.

2. To modify the expression of background color, use row () as little as possible. For example, calculate the line number in a grid of each line, and then judge the background color expression. The odd and even line judgment of background color, such as: if (row ()% 2 = = 0, – 3342337), is mainly row (), because this function can not optimize the calculation, and the number of expressions must be calculated as many times, and the calculation must be delayed after the expansion. In this way, if you expand more, it will have a greater impact on performance.

3. For cross grouping, the number of data sets should be reduced as much as possible, and single data set should be used as much as possible.