Performance Test common indicators

transactions (user transaction analysis)
User transaction analysis is a basic performance analysis that is performed at the user's perspective.

1. Transation sunmmary (transaction summary)
Comprehensive analysis of transactions is the first step in performance analysis, through the analysis of the test time of the success and failure of the user transaction, you can directly determine whether the system is functioning properly.

2, Average Transaciton Response time (transaction average response times)
Transaction average response time shows the average time spent executing a transaction during a test scenario run, through which you can analyze the performance trend of the application system during the test scenario run.
Example: With the change of testing time, the speed of the system processing transaction begins to gradually slow down, which indicates that the overall performance will have a downward trend when the application system changes with the production time .

3, transactions per Second (number of transactions per second/tps)
The "Number of transactions per second/tps" Displays the number of passes, failures, and stops of each transaction in the scene every second, making it an important parameter to examine the performance of the system. It allows you to determine the time transaction load of the system at any given moment. The analysis of TPs mainly depends on the performance trend of the curve.
Comparing it to the average transaction response time, you can analyze the impact of the number of transactions on the execution time.
Example: When the pressure increases, the clickthrough rate/tps curve If the change is slow or has a flat trend, it is likely that the server began to appear bottlenecks.

4, total transactions per Second (number of transactions per second)
"Total transactions per second" shows the total number of transactions that were passed in each second during the scene, the total number of failed transactions and transactions that were stopped.

5, Transaction Performance sunmmary (Transactional performance summary)
The Transaction performance summary displays the minimum, maximum, and average execution times for all transactions in the scenario, and can directly determine whether the response time meets the user's requirements.
Focus on the average and maximum execution time of a transaction, and if its scope is not within the user's acceptable time frame, cause analysis is required.

6. Transaction Response time under load (transaction response times and load)
"Transaction response time and load" is the "running virtual user" graph and "average response transaction time" graph, through which we can see the relationship between the transaction response time and the number of users, so as to master the performance data of the system in user concurrency, and provide reference for the extended user system. This figure can see the overall impact of the virtual user load on the execution time, which is useful for analyzing a test scenario with a gradient load.

7, Transaction Response time (percentile) (Transaction response times (%))
"Transaction response Time (percentage)" is a comprehensive analysis based on the results of the test, that is, the tool through a number of statistical analysis methods indirectly obtained by the chart. It allows you to analyze the percentage of transactions that can be performed within a given transaction response time range.

8, Transaction Response Time (distribution) (Transaction response times (distribution))
Transaction response Time (distribution) shows the distribution of the time spent in the execution of a transaction during the run of the scene, through which you can understand the number of transactions for different response times during the test. If the system pre-defines the minimum and maximum transaction response times acceptable to the related transaction, you can use this graph to determine whether the server performance is within acceptable range.

Web Resources (Web resource analysis)
Web resource analysis is a performance analysis of a Web server starting from a server.

1, Hits per Second (clicks per second)
Clicks per second, even if the number of HTTP requests that the virtual user submits per second to the Web server during the run scenario.
It allows you to evaluate the amount of load generated by a virtual user, and compare it to the average transaction response time graph to see how many clicks affect transactional performance. by looking at "clicks per second", you can determine whether the system is stable. Decreased CTR of the system usually indicates that the server is responding slowly and needs further analysis to find the bottleneck of the system.

2, throughput (throughput rate)
The throughput rate shows the throughput per second of the server during the scene's run. The unit of measure is bytes, which represents the amount of data that the virtual uses to obtain from the server every given second.
The amount of load generated by the virtual user can be assessed based on the throughput of the server, as well as the ability of the server to handle traffic and the presence of bottlenecks.
The difference between the throughput graph and the clickthrough rate graph:
The throughput graph, which is the number of HTTP requests processed per second by the server.
The "clickthrough Rate" graph is the total amount of data that the client obtains from the server per second.

3. HTTP status Code Summary (HTTP state Codes Overview)
The HTTP status Code summary shows the number of HTTP status codes returned from the Web server during a scene or session step, grouped by code. The HTTP status code represents the status of the HTTP request.

4, HTTP responses per Second (HTTP responses per second)
HTTP responses per second is the number of different HTTP status codes returned per second from the Web server during the running scenario , as well as information about other types of status codes, by analyzing the status code, You can determine how the server is operating under pressure, or you can locate the code script that generated the error by grouping the results shown in the diagram.

5. Pages Downloader per Second (number of pages downloaded per second)
Download pages per second displays the number of pages downloaded from the server in each second that the scene or session step runs. Use this graph to calculate the amount of load generated by vuser based on the number of pages downloaded.
As with the throughput graph, the number of pages downloaded per second is the amount of data that VUser receives from the server within any given second. However, the throughput considerations for each resource are extremely large (for example, the size of each GIF file, the size of each page). The number of pages downloaded per second only takes into account the number of pages.
Note: To see the number of pages downloaded per second, "pages per second (HTML mode only)" must be set up in R-t-s.

6, retries per Second (number of retries per second)
Retry per second displays the number of connections attempted by the server in each second that the scene or session step is running.
The server connection is retried in the following situations:
A. Initial connection Unauthorized
B. Proxy Authentication Required
C, the server closed the initial connection
D. The initial connection cannot connect to the server
E, the server was initially unable to resolve the load generator's IP address

7. Retries Summary (Summary of retry attempts)
The retry summary shows the number of connections attempted by the server during the run of the scene or session step, grouped by retry reason. Use this graph with the retry per second graph to determine at what point the server was retried while the scene or session step was running.

8, Connections (number of connections)
The number of connections displays the number of TCP/IP connections that are open at each point in time during the scene or session step run.
With this diagram, you can know when additional connections need to be added .
Example: when the number of connections reaches a steady state and the transaction response time increases rapidly, adding a connection can greatly improve the performance (transaction response time will be reduced).

9, Connections per Second (number of connections per second)
The number of connections per second displays the number of TCP/IP connections that are established per second during the run of the scenario.
Ideally, many HTTP requests should use the same connection instead of opening a new connection for each request. By the number of connections per second graph can see the server processing, indicating that the performance of the server is gradually declining.

10, SSLs per Second (SSL connections per second)
SSL connections per second displays the number of new and reused SSL connections that are opened in each second that the scene or session step runs. When you open a TCP/IP connection to a secure server, the browser opens the SSL connection.

Web page Breakdown (page element subdivision)
"Page element segmentation" is primarily used to evaluate whether the content of a page affects the response time of a transaction, and it allows you to drill down into the problem of those slow-downloading graphics or interrupted connections on the site.
Elements.

"page decomposition" shows how a particular transaction responds to the test process, and then analyzes whether the related transaction is running properly. The
page decomposition graph can be further subdivided in the following four ways:
1), Download time Breaddown (download times breakdown) The
Download time breakdown graph shows the download time for different elements in a Web page, and also allows you to decompose the time according to the download process. Use different colors to display the DNS resolution time, establish connection time, the first buffer time, and so on each of their respective proportions.
2), Component breakdown (Over time) (component breakdown (with the Change)
Component Breakdown Graph shows a breakdown of the page component over time for a selected page. This diagram makes it easy to see which elements are not stable during the test. This diagram is especially useful for pages that need to download more controls on the client side, and by analyzing the control's response time, it is easy to find that the controls are unstable or time-consuming.
3), Download time breakdown (Over time) (Download period breakdown (with change)
"Download time breakdown (over time)" graph showing page element download time breakdown (over time) for selected pages, It shows very clearly how the various elements of the page are downloaded during the stress test. The
Download time breakdown graph shows the statistical results of the time-of-day statistical analysis of the page element response for the entire test process, "Download time breakdown (over time)", and the statistics of the page element response time in each second during the run of the scenario, each of which analyzes the download time of the page elements from both a macro and a microscopic perspective.
4), time to primary buffer breakdown (over time) (first buffer time subdivision (with change)
"First buffer break (change over time)" The figure shows the time, in seconds, of the server time and network time of each page component in each second that the scene or session step ran, before the first buffer returned from the Web server was successfully received. First buffer time: When a client establishes a connection to the server, the start of a packet is sent from the server, the data is transmitted over the network to the client, and the browser receives the first buffer.

2. Page Component Breakdown (component subdivision)
The page component Breakdown graph shows the average download time (in seconds) for each Web page and its components. You can sort the columns by the average number of seconds that are used to download the component, which helps isolate the problematic component.

3. Page Component breakdown (over time) (paging component decomposition (varies over))
The page component decomposition (over time) graph shows the average response time (in seconds) for each page and its components in each second during the scenario run.

4. Page Download Time Breakdown (page download times breakdown)
The page download time Breakdown graph shows a breakdown of the download time for each page component, depending on whether the slow transaction response time during a Web page download is caused by a network error or by a server error.
The page download time breakdown graph subdivides The download process for each component based on DNS resolution time, connection time, first buffer time, SSL handshake time, receive time, FTP authentication time, client time, and error time.

5. Page Download time breakdown (over time) (page download times breakdown (varies over times))
Page download time breakdown (over time) The graph shows the breakdown of the download time for each page component in each second during the run of the scenario. Use this diagram to determine which point in time the network or server has been in the course of the execution of the scenario.
The page component breakdown (over time) graph and page download time breakdown (over time) graphs are often combined for analysis: Identify problematic components first, then analyze their download process to locate the cause.

6. Time to first buffer breakdown
The first buffer time breakdown graph shows the relevant server/network time for each page component in the period before the first buffer returned from the Web server was successfully received. If the download time for a component is long, you can use this figure to determine whether the problem is related to the server or to the network.
Network time: The average time that is defined as the first HTTP request that begins at the moment until it is confirmed.
Server time: The average time that is defined to begin when the initial HTTP request acknowledgement is received until a buffer from the Web server is received successfully.

7, time-to-first buffer breakdown (over time) (a breakdown of the initial buffer period (with the change))
The first buffer time breakdown (over time) graph shows the server time and network time for each page component in each second that the scene is running, before it successfully receives the first buffer returned from the Web server. You can use this graph to determine the point in time when a server or network problem occurs during a scene run.

8, Downloader Component size (KB) (downloaded component size)
The downloaded component size graph shows the size of each page build that has been downloaded. It allows you to see directly which components are larger and need to be further optimized to improve performance

Performance Test common indicators