问题描述
这是我的测试计划线程属性的配置:
This is the configuration of my test plan thread properties:
Number of Threads (users): 100
Ramp-up Period (in seconds): 10
Loop Count : Forever
Delay thread creation until needed: No
Scheduler: No
我在一夜之间进行了测试，总持续时间为 14 小时 7 分钟(大约 50820 秒).加载 jtl 文件后，汇总报告中显示的样本数为 1050975.我尝试计算，但我无法理解它是如何得出这么多样本的.
I ran the test overnight with the total duration of 14 hours and 7 minutes (approximately 50820 seconds). After loading the jtl file, The number of samples shown in summary report is 1050975. I tried to compute but I can't understand how it came up with that many sample.
如果 Ramp-up Period 是 JMeter 创建每次迭代线程数所用的时间，并且测试的持续时间是 50820 秒，那么我应该只有 508200 个样本 ( 50820/10 * 100 ).我不知道循环计数如何或是否会影响这一点.
If Ramp-up Period is the time taken by JMeter to create the number of threads per iteration and if the duration of the test is 50820 seconds, then I should have 508200 samples only ( 50820/10 * 100 ). I do not know how or if Loop Count affects this.
推荐答案
线程组 是 JMeter 启动线程总数所需的时间.在您的情况下，这意味着每 0.1 秒一个新线程在 10 秒后开始提供 100 个正在运行的线程.这 100 个线程背靠背执行您的测试迭代，因此在加速后 100 个线程在测试期间连续运行.
The ramp-up in the Thread Group is the time JMeter should take to start the total number of threads. In your situation this means that every 0.1 second a new thread starts giving 100 running threads after 10 seconds. These 100 threads perform your test iterations back-to-back, so after the ramp-up 100 threads run continously for the duration of the test.
这篇关于JMeter 理解提升的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！