LoadRunner- Summary (II)

directory

  • 1. The response time of things

  • 2. Detailed explanation of correlation function

  • 3. 90 Percent (User experience percentage)

1. The response time of things

The system response time obtained by transaction is composed of many parts. Generally speaking, the response time is composed of three parts: network time, server processing time and network delay. Let’s take a look at the path a client has to take when making a request back to the server.

(1) Network time

The client sends a request to the Web Server through the network (N1). The Web Server then sends the processed request to the App Server (N2); The App Server sends the operation data instruction to the Database (N3); The Database Server sends the query result data back to the App Server (N4). The App Server sends the processed page to the Web Server (N5). Finally, the Web Server forwards the HTML to the client (N6 elapsed time). The Nx here is all the time spent on network transmission, not the time spent on business processing.

(2) Server processing time

Another aspect to consider is the time required for each server to process WT, AT, DT.

(3) Network delay

In addition to the above two time costs, there is the problem of network latency to consider.

Therefore, the final response time composition is:

Response time = network delay time + WT+AT+DT + (N1+N2+N3) + (N4+N5+N6) + WT+AT+DT

It is also easy to think of response time as consisting of network overhead (front end) and server-side overhead (back end)

Factors affecting the network generally include the following:

(1) Front-end Network

DNS Lookup

Time to connect

Time to first buffer

Network Time

Download Time

SSL handshake

FTP authentication

Client Time

Error Time

Network latency

(2) Back-end services

Web Server

Servlet Time

Method Time

Static dynamic compression

Web Application Server

EJB Time

Method Time

JNDI Lookup

Database Server

JDBC Time

Connect Time

Execute Time

2. Detailed explanation of correlation function

The correlation function web_reg_save_param_ex()

(1) Parameter Name

Set the name of the parameter, and the associated content will be stored in the parameter. This is affected by the Ordinal option.

Such as:

Set Parameter Name to temp. If the corresponding Ordinal option is any number, only one matching record is associated and the associated value is stored in the temp Parameter. When Ordinal is All, the associated value is stored in the “temp_ number” parameter array, and a temp_count parameter is added to hold the number of associated entries.

(2) Left Boundary

The left boundary is set here, which is used to fill in the left matching content rules of the association for data processing. The left boundary contains a string. For example, the left boundary is converted to the following form:

` web_reg_save_param_ex (` ` ParamName = "test", ` ` "LB = left border", ` ` "RB =", ` ` SEARCH_FILTERS, ` ` LAST); `Copy the code

Note:

If the input contains double quotes, use the escape character \, for example:

` web_reg_save_param_ex (` ` ParamName = "test", ` ` "LB = \" left border, "` `" RB = ", ` ` SEARCH_FILTERS, ` ` LAST); `Copy the code

(3) Match case

By default, the bounds are Match case, that is, case checking. You can override the following option to ignore case checking and see the function change to the following form:

` web_reg_save_param_ex (` ` ParamName = "test", ` ` "LB/IC = left border", ` ` "RB =", ` ` SEARCH_FILTERS, ` ` LAST); `Copy the code

Binary data

Use this option if you want to associate content with non-ASCII characters. Check this option to see the function change to the following form:

`web_reg_save_param_ex(` `"ParamName=test",` `"LB/BIN=\\x3F\\xDD",` `"RB=",` `SEARCH_FILTERS,` `LAST); `Copy the code

5) Regular expression

In LR11, correlation functions provide the function of using regular expressions, but in LR11 Patch3, web_reg_SAVe_param_ex function is removed to support this function, which is implemented by web_reg_save_param_regexp.

(6) Right Boundary

The right boundary is set here, which is used to fill in the right matching content rules for data processing during association. The options are the same as the left boundary.

(7) the Ordinal

This keyword can be used in many functions, such as any integer or All. If you fill in a number, the sequential values are fetched from the returned record, while filling in All returns everything.

(8) Save Offset

Sets the associated content offset, starting with the number of bits.

(9) Save Length

The length of the associated content to save.

(10) Warm if text was not found (Default is Error)

What if the associated object does not exist? The default value is Error, and by default an Error is displayed if nothing is associated. If you select this option, only a simple message is displayed indicating that the content is not caught, and no error is generated.

(11) Filters

The options below are all set to help with the correlation return qualification, which further reduces the scope of the return.

(12) the Scope

The Scope of the associated query has been modified in LR11 to make it more powerful. Here Scope provides 4 options: Body, Headers, Cookies, and All.

1) ALL

The easy way to understand this is to have everything returned by the server treated as a target to be associated with.

2) Headers/Body/Cookies

All three options are associated with everything returned from the request, including images, JavaScript scripts, and so on. The difference lies in the separation of the returned information.

3) the Header

The associated content is the HTTP header that all servers return requests for. You can see by looking at what the server returns that everything before the Body belongs to the Header.

4) the Body

That’s what the server returns after the Body.

5) Cookie

Refers to the part of the Header that defines a Cookie.

(13) Request URL

Filtering for URL addresses is provided to reduce the scope of the association. For example, we can fill in *.php to indicate that only PHP pages are filtered.

(14) Content Type

In web applications, the Content we want to associate is usually stored in HTML pages, so we usually use text/ HTML as Content Type filtering rules.

(15) Frame ID

This option is designed specifically for framed sites. Sometimes the content to be associated with is in a frame, so you need to specify which page is in the frame.

(16) Ignore Redirection

In some cases, HTTP 3XX redirects are used to complete the page jump. This option is used to ignore the page jump information. If this option is checked, pages that pass this redirection technique will not be associated.

3. 90 Percent (User experience percentage)

This statistic is often seen in test reports: 90 Percent.

This value indicates that 90% of the sampled data is smaller than it and 10% of the sampled data is larger than it.

Examples are as follows:

Suppose there is a set of numbers (1, 3, 4, 6, 5, 7, 8, 2, 9, 10), and the ninth largest number in the 10 numbers is 9, so 90 Percent is 9.

Its main purpose is to know what percentage of users are available at a given response time. Of course, this 90% is adjustable in Analysis using the Transaction Percentile option under the Summary Filter in the View menu.

AllTests software tests

Share software testing, automated testing, performance testing, test development and other technical articles and resources, welcome to pay attention to!