Broadcasting and television operation and maintenance supervision software multi-scene performance test method based on TPS mathematical model

Abstract

The invention discloses a radio and television operation and maintenance supervision software multi-scene performance test method based on a TPS mathematical model, and the method comprises the following steps: 1, analyzing the performance requirements of radio and television operation and maintenance supervision software, and determining the performance requirements of the radio and television operation and maintenance supervision software in different scenes; step 2, constructing a TPS mathematical model, analyzing performance requirements by using the TPS mathematical model, and calculating theoretical TPS values of the broadcast and television operation and maintenance supervision software in different scenes; 3, verifying the performance of the broadcast and TV operation and maintenance supervision software in different scenes by using a test tool, and obtaining an actual TPS value; 4, result comparison: if the actual TPS value is higher than or equal to the theoretical TPS value, indicating that the broadcast and TV operation and maintenance supervision software reaches the standard and the performance test passes; if the actual TPS value is lower than the theoretical TPS value, the software does not reach the standard and needs to be returned to the development stage for optimization and correction. The performance of the broadcast and TV operation and maintenance supervision software can be effectively evaluated and improved, and the stability and reliability of the software are ensured.

Description

Technical Field

[0001] The present invention relates to the field of radio and television technology, and in particular to a multi-scenario performance testing method for radio and television operation and maintenance supervision software based on a TPS mathematical model. Background Art

[0002] With the rapid development of the radio and television industry, the performance of radio and television operation and maintenance supervision software has become an important factor in ensuring the overall operating efficiency and stability of the industry. Radio and television operation and maintenance supervision software is responsible for real-time monitoring and management of massive amounts of radio and television content and data. Its performance directly affects the software's response speed, data processing capabilities, and end-user experience. In the process of processing large-scale data and executing complex queries, the software often needs to cope with high concurrent access and intensive data processing tasks, which places strict requirements on the software's performance.

[0003] In order to ensure that the radio and television operation and maintenance supervision software can run stably and efficiently in various complex scenarios, performance testing has become an indispensable link. In the process of users accessing the software, different access volumes bring different pressures on system performance. Although long-term low-frequency access has a small instantaneous load on the system, it will occupy resources for a long time; while high-frequency access in a short period of time will cause greater instantaneous pressure on the system. TPS (transactions per second) is the most direct reflection of user access pressure as a measure of the system's ability to process transactions per unit time. Traditional TPS performance testing methods show certain limitations when dealing with concurrent access and real-time monitoring requirements in the above complex scenarios. Summary of the invention

[0004] The purpose of the present invention is to provide a multi-scenario performance testing method for radio and television operation and maintenance supervision software based on the TPS mathematical model. The present invention can effectively evaluate and improve the performance of radio and television operation and maintenance supervision software and ensure the stability and reliability of the software.

[0005] The technical solution of the present invention is a multi-scenario performance testing method for radio and television operation and maintenance supervision software based on the TPS mathematical model, comprising the following steps:

[0006] Step 1: Analyze the performance requirements of the radio and television operation and maintenance supervision software to determine the performance requirements that need to be met in different scenarios;

[0007] Step 2: Construct a TPS mathematical model, use the TPS mathematical model to analyze performance requirements, and calculate the theoretical TPS value of the radio and television operation and maintenance supervision software in different scenarios;

[0008] Step 3: Use the test tool to verify the performance of the radio and television operation and maintenance supervision software in different scenarios and obtain the actual TPS value;

[0009] Step 4, result comparison: If the actual TPS value is higher than or equal to the theoretical TPS value, it means that the radio and television operation and maintenance supervision software meets the standard and the performance test passes; if the actual TPS value is lower than the theoretical TPS value, it means that the software does not meet the standard and needs to return to the development stage for optimization and correction. The above-mentioned radio and television operation and maintenance supervision software multi-scenario performance testing method based on the TPS mathematical model, in the long-term low-frequency access scenario, the formula of the TPS mathematical model is as follows:

[0010]

[0011] Among them, j and k are any values ​​between 0 and 1, j+k=1, α is the average concurrency, that is, the average number of page accesses per unit time within a period of time; λ is the peak concurrency, that is, the number of page accesses per unit time during the peak period; t is the time to complete a page access.

[0012] The aforementioned multi-scenario performance testing method for radio and television operation and maintenance supervision software based on the TPS mathematical model, in a short-term high concurrency and high load scenario, the formula of the TPS mathematical model is as follows:

[0013]

[0014] Among them, j and k are any values ​​between 0 and 1, j+k=1, α is the average concurrency, that is, the average number of page accesses per unit time within a period of time; λ is the peak concurrency, that is, the number of page accesses per unit time during the peak period; t is the time to complete a page access.

[0015] The aforementioned multi-scenario performance testing method for radio and television operation and maintenance supervision software based on the TPS mathematical model, the calculation formula for the average concurrent number is as follows:

[0016]

[0017] Where: α is the average number of concurrent users, i.e., the average number of page visits per unit time within a period of time; u is the maximum number of active users; t is the time to complete a page visit; t c The duration that the page is accessed.

[0018] In the aforementioned multi-scenario performance testing method for radio and television operation and maintenance supervision software based on the TPS mathematical model, the calculation formula for the peak concurrent number is as follows:

[0019]

[0020] The aforementioned multi-scenario performance testing method for radio and television operation and maintenance supervision software based on the TPS mathematical model, the testing tool is JMeter.

[0021] Compared with the prior art, the present invention constructs a TPS mathematical model applied to the performance test of radio and television operation and maintenance supervision software. The TPS mathematical model can calculate the theoretical TPS value of the radio and television operation and maintenance supervision software when it meets the performance requirements, and then use the theoretical TPS value to compare with the actual TPS of the actual test, so as to know the relationship between the actual TPS value and the theoretical TPS value, so that when the software does not meet the standards, it can return to the development stage in time for optimization and correction. Therefore, the present invention can effectively evaluate and improve the performance of radio and television operation and maintenance supervision software and ensure the stability and reliability of the software. In addition, in order to accurately reflect the business processing capabilities of the software in multiple scenarios, the present invention designs a TPS mathematical model based on the average concurrency and the peak concurrency for long-term low-frequency access and short-term high-frequency access scenarios, which adapts to various complex test scenarios to the greatest extent and facilitates the use of software developers. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0023] The present invention is further described below in conjunction with the accompanying drawings and embodiments, but they are not intended to limit the present invention.

[0024] Example: Multi-scenario performance testing method for radio and television operation and maintenance supervision software based on TPS mathematical model, such as Figure 1 As shown, the following steps are included:

[0025] Step 1. Analyze the performance requirements of the radio and television operation and maintenance supervision software to determine the performance requirements that need to be met in different scenarios; for example, the radio and television monitoring and supervision software has 10,000 active users performing regular queries during peak hours (such as 9 o'clock), and the software requires all query operations to be completed within 1 minute. After measurement, it takes 6 seconds to fully load the active page. Therefore, the software performance requirement is to ensure that the software can operate normally under such high load conditions and achieve the expected TPS (transactions per second). Of course, in other embodiments, the software can also be selected for performance analysis in a stable scenario (i.e., long-term low-frequency access).

[0026] Step 2: Construct a TPS mathematical model, use the TPS mathematical model to analyze performance requirements, and calculate the theoretical TPS value of the radio and television operation and maintenance supervision software in different scenarios;

[0027] In this step, under the condition of long-term low-frequency access (this scenario may have functions such as long-term scattered small-volume queries and short-term concentrated queries according to different business needs, and different coefficients j and k can be set to obtain a stable TPS that is more in line with actual production conditions), the formula of the TPS mathematical model is as follows:

[0028]

[0029] Among them, j and k are any values ​​between 0 and 1, j+k=1, α is the average concurrency, that is, the average number of page accesses per unit time within a period of time; λ is the peak concurrency, that is, the number of page accesses per unit time during the peak period; t is the time to complete a page access.

[0030] In a short-term high concurrency and high load scenario (or stress test scenario), the formula of the TPS mathematical model is as follows:

[0031]

[0032] Among them, j and k are any values ​​between 0 and 1, j+k=1, α is the average concurrency, that is, the average number of page accesses per unit time within a period of time; λ is the peak concurrency, that is, the number of page accesses per unit time during the peak period; t is the time to complete a page access.

[0033] The calculation formula of the average concurrent number is as follows:

[0034]

[0035] Where: α is the average number of concurrent users, i.e., the average number of page visits per unit time within a period of time; u is the maximum number of active users; t is the time to complete a page visit, i.e., the unit time required for a single user to access a page, such as the time required for a user to open the homepage from the beginning to the completion of loading; t c The duration of page access, that is, the duration of active users accessing the page during peak hours. For example, if 9:30 to 10 a.m. is the peak access time, the duration is 30 minutes.

[0036] For example, the radio and television operation and maintenance supervision software has 10,000 active users. The regular query at 9 o'clock is expected to last for 1 minute. It has been measured that the activity page takes 6 seconds to fully load. Then the calculation can be performed:

[0037]

[0038] From the above formula, we can conclude that the longer the page loading time is, the greater the concurrency is, and the situation where the previous user's access has not yet ended, and the next user initiates access to the system. The longer the query function takes, the smaller the concurrency is, the user's access becomes sparse, and the software pressure is reduced.

[0039] Then, according to the average concurrent number, an impact factor is set for it to obtain the peak concurrent number. In this embodiment, the calculation formula of the peak concurrent number is as follows:

[0040]

[0041] In this formula, Used as an influencing factor to obtain the peak concurrency value under corresponding performance requirements.

[0042] Following the example mentioned above, further calculation:

[0043]

[0044] Then, in a stable scenario, the theoretical TPS value calculated by the TPS mathematical model (in this embodiment, j and k are both set to 0.5) is:

[0045]

[0046] In high concurrency and high load scenarios, the theoretical TPS value calculated by the TPS mathematical model (in this embodiment, j and k are both set to 0.5) is:

[0047]

[0048] Step 3. Use the test tool to verify the performance of the radio and television operation and maintenance supervision software in different scenarios and obtain the actual TPS value; in this step, the test tool is JMeter, which is a powerful open source performance testing tool that can effectively test the system's TPS (transactions per second) and QPS (queries per second). By simulating a large number of concurrent user accesses, JMeter can generate high-concurrency requests to help evaluate the system's processing capabilities under different loads. It accurately measures the number of transactions and queries processed by the system per second by configuring thread groups and transaction controllers. In addition, the aggregated report and graphical result functions provided by JMeter can display the performance of TPS and QPS in real time, help identify performance bottlenecks, and provide data support for system optimization.

[0049] In this embodiment, the test tool mainly simulates high concurrency and high load scenarios (of course, in other embodiments, the test tool can also simulate long-term low-frequency access scenarios), so the scenario of 10,000 users initiating queries at 9 o'clock is simulated in JMeter to test the actual TPS value of the software under extreme load. By configuring the thread group and transaction controller, the load is gradually increased until the peak concurrency of 1095 calculated in theory is reached, and the actual TPS value is obtained through JMeter.

[0050] Step 4. Compare the results: If the actual TPS value is higher than or equal to the theoretical TPS value, it means that the radio and television operation and maintenance supervision software meets the standards and the performance test passes; if the actual TPS value is lower than the theoretical TPS value, it means that the software does not meet the standards and needs to return to the development stage for optimization and correction.

[0051] In this step, in high concurrency and high load scenarios, if the actual TPS value obtained by JMeter is less than the theoretical TPS value (698.3) calculated in step 2, it indicates that the system fails to meet the performance requirements and needs to return to the development stage for optimization, especially for high concurrency scenarios. If the actual TPS value obtained is greater than the theoretical TPS value (698.3) calculated in step 2, it indicates that the software has passed the performance test, the performance test process is completed, and the software is confirmed to meet the requirements. Similarly, if the actual TPS value obtained by JMeter is less than the theoretical TPS value (174.6) calculated in step 2 in a long-term low-frequency access scenario, it indicates that the system fails to meet the performance requirements and needs to return to the development stage for optimization. If the actual TPS value obtained is greater than the theoretical TPS value (174.6) calculated in step 2, it indicates that the software has passed the performance test, the performance test process is completed, and the software is confirmed to meet the requirements.

[0052] In summary, the present invention constructs a TPS mathematical model applied to the performance test of radio and television operation and maintenance supervision software. The TPS mathematical model can calculate the theoretical TPS value of the radio and television operation and maintenance supervision software when it meets the performance requirements, and then use the theoretical TPS value to compare with the actual TPS of the actual test, so as to know the relationship between the actual TPS value and the theoretical TPS value, so that when the software does not meet the standards, it can return to the development stage in time for optimization and correction. Therefore, the present invention can effectively evaluate and improve the performance of radio and television operation and maintenance supervision software, and ensure the stability and reliability of the software. In addition, in order to accurately reflect the business processing capabilities of the software in multiple scenarios, the present invention designs a TPS mathematical model based on the average concurrency and the peak concurrency for long-term low-frequency access and short-term high-frequency access scenarios, which adapts to various complex test scenarios to the greatest extent and facilitates the use of software developers.

Claims

1. A multi-scenario performance testing method for radio and television operation and maintenance supervision software based on the TPS mathematical model, characterized by: The steps include: Step 1: Analyze the performance requirements of the radio and television operation and maintenance supervision software to determine the performance requirements that need to be met in different scenarios; Step 2: Construct a TPS mathematical model, use the TPS mathematical model to analyze performance requirements, and calculate the theoretical TPS value of the radio and television operation and maintenance supervision software in different scenarios; Step 3: Use the test tool to verify the performance of the radio and television operation and maintenance supervision software in different scenarios and obtain the actual TPS value; Step 4: Compare the results: If the actual TPS value is higher than or equal to the theoretical TPS value, it means that the radio and television operation and maintenance supervision software meets the standards and the performance test has passed; If the actual TPS value is lower than the theoretical TPS value, it means that the software does not meet the standard and needs to return to the development stage for optimization and correction.

2. According to the multi-scenario performance testing method of radio and television operation and maintenance supervision software based on the TPS mathematical model according to claim 1, it is characterized by: Under long-term low-frequency access, the formula of the TPS mathematical model is as follows: Among them, j and k are any values ​​between 0 and 1, j+k=1, α is the average concurrency, that is, the average number of page accesses per unit time within a period of time; λ is the peak concurrency, that is, the number of page accesses per unit time during the peak period; t is the time to complete a page access.

3. According to the multi-scenario performance testing method of radio and television operation and maintenance supervision software based on the TPS mathematical model according to claim 1, it is characterized by: In a short-term high concurrency and high load scenario, the formula of the TPS mathematical model is as follows: Among them, j and k are any values ​​between 0 and 1, j+k=1, α is the average concurrency, that is, the average number of page accesses per unit time within a period of time; λ is the peak concurrency, that is, the number of page accesses per unit time during the peak period; t is the time to complete a page access.

4. The multi-scenario performance testing method for radio and television operation and maintenance supervision software based on the TPS mathematical model according to claim 2 or 3 is characterized in that: The calculation formula of the average concurrent number is as follows: Where: α is the average number of concurrent users, i.e., the average number of page visits per unit time within a period of time; u is the maximum number of active users; t is the time to complete a page visit; t c The duration that the page is accessed.

5. According to claim 4, the multi-scenario performance testing method for radio and television operation and maintenance supervision software based on the TPS mathematical model is characterized by: The calculation formula of the peak concurrent number is as follows:

6. The multi-scenario performance testing method for radio and television operation and maintenance supervision software based on the TPS mathematical model according to claim 1 is characterized in that: The testing tool is JMeter.

Images