Shulou(Shulou.com)06/02 Report--

This article mainly introduces "what is the jmeter aggregation report". In the daily operation, I believe many people have doubts about what the jmeter aggregation report is. The editor consulted all kinds of data and sorted out simple and easy-to-use methods of operation. I hope it will be helpful to answer the doubts about "what is the jmeter aggregation report?" Next, please follow the editor to study!

The most commonly used listener-- aggregation report, Aggregate Report.

Let's first look at the meaning of the main names in the aggregation report:

Label: name of each request

# Samples: the number of requests sent

Average: average response time (in milliseconds). The default is the average response time of a single Request. When Transaction Controller is used, the average response time can also be displayed in Transaction units.

Median: median, more on later

90%Line: later on

95%Line: later on

99%Line: later on

Min: minimum response time

Max: maximum response time

Error rate of Error%: requests = number of error requests / total number of requests

Throughput: throughput. By default, it represents the number of requests completed per second, which is generally considered TPS. (note the change in units, as shown in the figure above. When the TPS is very low, the default in jmeter will be counted as the value per minute. In this case, we need to convert it to seconds.)

KB/sec: the amount of data received from the server per second.

90%Line

Here I think it is necessary to talk about the understanding of 90%Line:

Many people understand it as the average response time of 90% of users. (I always thought so before), but then I found out I was wrong.

Then take a look at what the JMeter website says?

90% Line-90% of the samples took no more than this time. The remaining samples at least as long as this.

"90% of the requests take no more than this time, and the remaining requests take at least more time." In other words, 90% of the request time is below this time.

Here's a mathematical concept: percentiles.

Percentile:

In statistical terms, if a group of data is sorted from large to small and the corresponding cumulative percentile is calculated, the value of the data corresponding to a 100th percentile is called the percentile. It can be expressed as: a group of n observations are arranged according to numerical size, and the value in the p% position is called the p percentile.

The percentile is usually expressed as the percentile, such as the fifth percentile, which indicates that the cumulative frequency of all measurements is up to 5%. Taking height as an example, the fifth percentile of the height distribution indicates that 5% of people are smaller than this measured value, and 95% of the height is greater than this measured value.

The percentile is the actual value corresponding to the percentile.

Percentile is a measure of the position of the data, but it is not necessarily the central position. Percentiles provide information about how each data item is distributed between the minimum and maximum values. For data without a large number of duplicates, the p percentile divides it into two parts. About p% of the data items have a value less than the p percentile, while about (100%) of the data items have a value greater than the p percentile. The p percentile is strictly defined as follows:

The p percentile is such a value that at least p% of data items are less than or equal to this value, and at least (100%)% of data items are greater than or equal to this value.

The scores of entrance examinations in colleges and universities are often reported in percentiles. For example, suppose an examinee's original score in the Chinese part of the entrance exam is 54. Compared with other students who take the same exam, it is not easy to know his grades. But if the original score of 54 corresponds to the 70th percentile, we can know that about 70% of the students have lower test scores than him, while about 30% of the students have higher scores.

Give me another fake son:

There are 10 numbers:

1, 2, 3, 4, 5, 6, 7, 8, 9, 10 arrange them from small to big.

Find that its 90th percentile, that is, the ninth number happens to be 9, then his 90%Line is 9.

Another set of numbers:

2, 2.1, 2.5, 3, 3.4, 3.4, 4, 4, 4, 4, 5, 5, 5, 5.9, 5.91, 6.8, 8, 12, 24, 24.1 are arranged from small to large.

Find that its 90th percentile, the 18th number, is 12, then his 90%Line is 12.

So what's the point of using percentiles in performance testing?

It can use our analysis results to be more accurate!

Because when evaluating the results of a test, only the average response time is not enough. If there is a test in which a total of 100 requests are responded to, the minimum response time is 0.02 seconds, the maximum response time is 110 seconds, and the average transaction response time is 4.7 seconds, would you think of such a large deviation between the minimum and maximum response time? Will it result in that the average itself is not credible?

If we count the response time of each request with excel, we will find that the probability of occurrence of that maximum value is only 1/1000 or 1/10000, and the response time of the remaining 99% of user requests is within the range defined by performance requirements. Therefore, in order to more accurately measure the overall request response time, in addition to the average response time, 90%Line, 95%Line and 99%Line should be used to assist the statistics.

To sum up, the meaning of the percentile in the aggregate report:

Median: median, the response time of 50% users is less than this value, note the difference between it and the average response time of average

The response time of 90% of the users in LinePlux is less than this value.

95% of the response time of 95% of the users of LineGroupe is less than this value.

The response time of 99% of LineGroupe 99% users is less than this value.

At this point, the study of "what is the jmeter aggregation report" is over. I hope to be able to solve your doubts. The collocation of theory and practice can better help you learn, go and try it! If you want to continue to learn more related knowledge, please continue to follow the website, the editor will continue to work hard to bring you more practical articles!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.