A friend asked me how to set parameters in lr to determine whether to end the operation based on the number of parameters. I never thought about this problem because I have never had this requirement at work. We always think that the running time of LR is controlled by the running time.
1. File-format parameter settings
1. Select next row
Sequence
Random
Unique
2. In terms of "order" and "unique ",
1) Select next row = sequential undate value on = Iteration
This setting indicates that each virtual user reads data from the beginning to the end of the file. For example, if two concurrent users are added, each user reads the parameters from 1-10 lines of the parameter file sequentially, so that the parameters in the loop of each virtual user are the same. For example, the parameter for the first loop reading of virtual user 1 is key1, the parameter for the second loop reading is key2, and the parameter for the first loop reading of virtual user 2 is key1, the parameter for the second loop reading is also key2.
2) Select next row = SEQUENCE undate value on = occurrance
This setting indicates that every keyword function is updated. In this way, the request parameters of each virtual user are different.
3) Select next row = unique undate value on = Iteration
This setting indicates that the parameter file is segmented by the number of concurrent threads, and each virtual user is assigned a parameter, which ensures that the running parameters are unique. For example, if there are five concurrent threads and 100 parameters, the parameters allocated by virtual user 1 are 1-20 rows, and the parameters allocated by virtual user 2 are 21-40 rows, and so on.
4) Select next row = unique undate value on = occurrance allocate x values on each vuser
The difference is that you can set the block size (X)