Pressure measurement data processing method and system

Abstract

The invention discloses a pressure measurement data processing method and system, and the method comprises the steps: obtaining uniform resource locators of a plurality of pressure measurement platforms, carrying out the analysis of the uniform resource locators, obtaining a port and a path, determining the protocols of the plurality of pressure measurement platforms according to the port and the path, and collecting multi-source pressure measurement data from the plurality of pressure measurement platforms according to the protocols, the method comprises the following steps: performing data cleaning on multi-source pressure measurement data by using a timestamp alignment algorithm to obtain cleaned multi-source pressure measurement data, and storing the cleaned multi-source pressure measurement data based on a distributed cache architecture, so that the multi-source pressure measurement data can be smoothly obtained from different pressure measurement platforms. The collected multi-source pressure measurement data can be effectively processed by using a timestamp alignment algorithm, and finally the multi-source pressure measurement data is stored based on a distributed cache architecture, so that the multi-source pressure measurement data can be more effectively integrated, and the compatibility and precision of multi-source pressure measurement data integration are improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a stress testing data processing method and system. Background Art

[0002] Existing technologies face poor protocol adaptability when dealing with multi-source stress testing data, resulting in the inability to effectively obtain all stress testing data. After collecting multi-source stress testing data, traditional solutions use fixed time window alignment (such as 1-second granularity), which leads to cross-node data deviation and low data alignment accuracy. Furthermore, when storing stress testing data, the storage architecture lacks scalability. These issues make it difficult to integrate multi-source stress testing data, resulting in data silos. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a stress test data processing method and system, which can improve the compatibility and accuracy of multi-source stress test data integration.

[0004] In order to solve the above technical problems, a technical solution adopted by the present invention is: A stress test data processing method includes the following steps: Obtaining uniform resource locators of multiple stress testing platforms and parsing the uniform resource locators to obtain ports and paths; Determining protocols of the multiple stress testing platforms according to the ports and the paths, and collecting multi-source stress testing data from the multiple stress testing platforms according to the protocols; Cleaning the multi-source stress testing data using a timestamp alignment algorithm to obtain cleaned multi-source stress testing data; The cleaned multi-source stress testing data is stored based on a distributed cache architecture.

[0005] In order to solve the above technical problems, another technical solution adopted by the present invention is: A stress test data processing system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are implemented: Obtaining uniform resource locators of multiple stress testing platforms and parsing the uniform resource locators to obtain ports and paths; Determining protocols of the multiple stress testing platforms according to the ports and the paths, and collecting multi-source stress testing data from the multiple stress testing platforms according to the protocols; Cleaning the multi-source stress testing data using a timestamp alignment algorithm to obtain cleaned multi-source stress testing data; The cleaned multi-source stress testing data is stored based on a distributed cache architecture.

[0006] The beneficial effects of the present invention are: obtaining uniform resource locators of multiple stress testing platforms, parsing the uniform resource locators to obtain ports and paths, determining the protocols of the multiple stress testing platforms according to the ports and paths, collecting multi-source stress testing data from the multiple stress testing platforms according to the protocols, using a timestamp alignment algorithm to clean the multi-source stress testing data to obtain cleaned multi-source stress testing data, and storing the cleaned multi-source stress testing data based on a distributed cache architecture, thereby smoothly obtaining multi-source stress testing data from different stress testing platforms, using a timestamp alignment algorithm to effectively process the collected multi-source stress testing data, and finally storing the multi-source stress testing data based on the distributed cache architecture, which can more effectively integrate the multi-source stress testing data, thereby improving the compatibility and accuracy of the multi-source stress testing data integration. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 This is a flowchart of a method for processing stress test data according to an embodiment of the present invention; Figure 2 The figure is a structural diagram of a stress test data processing system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0008] To illustrate the technical content, achieved objectives and effects of the present invention in detail, the following description is given in conjunction with the embodiments and accompanying drawings.

[0009] Please refer to Figure 1 , a stress test data processing method, comprising the steps of: Obtaining uniform resource locators of multiple stress testing platforms and parsing the uniform resource locators to obtain ports and paths; Determining protocols of the multiple stress testing platforms according to the ports and the paths, and collecting multi-source stress testing data from the multiple stress testing platforms according to the protocols; Cleaning the multi-source stress testing data using a timestamp alignment algorithm to obtain cleaned multi-source stress testing data; The cleaned multi-source stress testing data is stored based on a distributed cache architecture.

[0010] From the above description, it can be seen that the beneficial effects of the present invention are: obtaining the uniform resource locators of multiple stress testing platforms, and parsing the uniform resource locators to obtain ports and paths, determining the protocols of multiple stress testing platforms according to the ports and paths, and collecting multi-source stress testing data from multiple stress testing platforms according to the protocols, using a timestamp alignment algorithm to clean the multi-source stress testing data to obtain cleaned multi-source stress testing data, and storing the cleaned multi-source stress testing data based on a distributed cache architecture, so that multi-source stress testing data can be smoothly obtained from different stress testing platforms, and the timestamp alignment algorithm can be used to effectively process the collected multi-source stress testing data. Finally, storing the multi-source stress testing data based on the distributed cache architecture can more effectively integrate the multi-source stress testing data, thereby improving the compatibility and accuracy of the multi-source stress testing data integration.

[0011] Furthermore, the protocol for determining the multiple stress testing platforms according to the ports and the paths includes: Get the preset protocol rule library; A protocol corresponding to the port and the path is determined from the preset protocol rule library as a protocol for the multiple stress testing platforms.

[0012] From the above description, it can be seen that the protocol corresponding to the port and path of the stress testing platform is determined from the preset protocol rule library, which realizes the automatic and accurate identification of the stress testing platform protocol and ensures that stress testing data can be collected from different stress testing platforms.

[0013] Furthermore, the multi-source stress testing data is cleansed using a timestamp alignment algorithm to obtain the cleansed multi-source stress testing data, including: Using the Precision Time Protocol to synchronize the clocks of the multi-source stress testing data; Determine whether the clock synchronization is successful. If so, perform timestamp alignment on the multi-source stress test data after the clock synchronization is successful to obtain the aligned multi-source stress test data. If not, calculate the network time protocol offset and use a linear regression algorithm to compensate for the network time protocol offset to obtain a compensation value. Perform timestamp alignment on the multi-source stress test data according to the compensation value to obtain the aligned multi-source stress test data. Calculate the mean and standard deviation of the aligned multi-source stress testing data; Anomalies are identified on the aligned multi-source stress testing data based on the mean and the standard deviation according to a triple standard deviation rule to obtain abnormal data, and the abnormal data in the aligned multi-source stress testing data is marked to obtain cleaned multi-source stress testing data.

[0014] As can be seen from the above description, precise clock synchronization ensures the reliability and accuracy of data processing. By calculating the mean and standard deviation and applying the triple standard deviation rule to identify data anomalies, abnormal data can be effectively identified and marked, improving data quality and thus the accuracy of multi-source stress testing data integration.

[0015] Furthermore, before storing the cleaned multi-source stress testing data based on the distributed cache architecture, the method further includes: Analyze the players corresponding to the cleaned multi-source stress testing data to obtain behavior patterns; Calculating the weight of each stress testing data in the cleaned multi-source stress testing data according to the behavior pattern; The storing of the cleaned multi-source stress testing data based on the distributed cache architecture includes: The cleaned multi-source stress testing data is stored based on a distributed cache architecture according to the weight.

[0016] As can be seen from the above description, the weight of the stress testing data is calculated according to the behavior pattern, and then the cleaned multi-source stress testing data is stored based on the weight based on the distributed cache architecture. The value of the data can be identified by the weight, so that high-weight data can be cached first in the future, thereby optimizing the distributed cache performance.

[0017] Furthermore, storing the cleaned multi-source stress testing data based on the weight and distributed cache architecture includes: Determining data partitions and priorities of the data partitions; Determine a target data partition to which the cleaned multi-source stress testing data is to be written according to the weight and the priority; Writing the cleaned multi-source stress testing data into the target data partition; Determine whether data has been consumed from the data partition. If so, write the consumed stress test data to the shard cluster.

[0018] From the above description, we can know that the target data partition for the cleaned multi-source stress testing data to be written is determined based on the weight and priority, so that high-weight data will be assigned to the high-priority partition to ensure fast access. After the data is consumed, it is written to the shard cluster. The data partition is responsible for responding to stress testing requests with low latency. The consumed data is asynchronously persisted to the shard cluster to achieve read-write separation and improve overall throughput.

[0019] Please refer to Figure 2 A stress test data processing system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are implemented: Obtaining uniform resource locators of multiple stress testing platforms and parsing the uniform resource locators to obtain ports and paths; Determining protocols of the multiple stress testing platforms according to the ports and the paths, and collecting multi-source stress testing data from the multiple stress testing platforms according to the protocols; Cleaning the multi-source stress testing data using a timestamp alignment algorithm to obtain cleaned multi-source stress testing data; The cleaned multi-source stress testing data is stored based on a distributed cache architecture.

[0020] From the above description, it can be seen that the beneficial effects of the present invention are: obtaining the uniform resource locators of multiple stress testing platforms, and parsing the uniform resource locators to obtain ports and paths, determining the protocols of multiple stress testing platforms according to the ports and paths, and collecting multi-source stress testing data from multiple stress testing platforms according to the protocols, using a timestamp alignment algorithm to clean the multi-source stress testing data to obtain cleaned multi-source stress testing data, and storing the cleaned multi-source stress testing data based on a distributed cache architecture, so that multi-source stress testing data can be smoothly obtained from different stress testing platforms, and the timestamp alignment algorithm can be used to effectively process the collected multi-source stress testing data. Finally, storing the multi-source stress testing data based on the distributed cache architecture can more effectively integrate the multi-source stress testing data, thereby improving the compatibility and accuracy of the multi-source stress testing data integration.

[0021] Furthermore, the protocol for determining the multiple stress testing platforms according to the ports and the paths includes: Get the preset protocol rule library; A protocol corresponding to the port and the path is determined from the preset protocol rule library as a protocol for the multiple stress testing platforms.

[0022] From the above description, it can be seen that the protocol corresponding to the port and path of the stress testing platform is determined from the preset protocol rule library, which realizes the automatic and accurate identification of the stress testing platform protocol and ensures that stress testing data can be collected from different stress testing platforms.

[0023] Furthermore, the multi-source stress testing data is cleansed using a timestamp alignment algorithm to obtain the cleansed multi-source stress testing data, including: Using the Precision Time Protocol to synchronize the clocks of the multi-source stress testing data; Determine whether the clock synchronization is successful. If so, perform timestamp alignment on the multi-source stress test data after the clock synchronization is successful to obtain the aligned multi-source stress test data. If not, calculate the network time protocol offset and use a linear regression algorithm to compensate for the network time protocol offset to obtain a compensation value. Perform timestamp alignment on the multi-source stress test data according to the compensation value to obtain the aligned multi-source stress test data. Calculate the mean and standard deviation of the aligned multi-source stress testing data; Anomalies are identified on the aligned multi-source stress testing data based on the mean and the standard deviation according to a triple standard deviation rule to obtain abnormal data, and the abnormal data in the aligned multi-source stress testing data is marked to obtain cleaned multi-source stress testing data.

[0024] As can be seen from the above description, precise clock synchronization ensures the reliability and accuracy of data processing. By calculating the mean and standard deviation and applying the triple standard deviation rule to identify data anomalies, abnormal data can be effectively identified and marked, improving data quality and thus the accuracy of multi-source stress testing data integration.

[0025] Furthermore, before storing the cleaned multi-source stress testing data based on the distributed cache architecture, the method further includes: Analyze the players corresponding to the cleaned multi-source stress testing data to obtain behavior patterns; Calculating the weight of each stress testing data in the cleaned multi-source stress testing data according to the behavior pattern; The storing of the cleaned multi-source stress testing data based on the distributed cache architecture includes: The cleaned multi-source stress testing data is stored based on a distributed cache architecture according to the weight.

[0026] As can be seen from the above description, the weight of the stress testing data is calculated according to the behavior pattern, and then the cleaned multi-source stress testing data is stored based on the weight based on the distributed cache architecture. The value of the data can be identified by the weight, so that high-weight data can be cached first in the future, thereby optimizing the distributed cache performance.

[0027] Furthermore, storing the cleaned multi-source stress testing data based on the weight and distributed cache architecture includes: Determining data partitions and priorities of the data partitions; Determine a target data partition to which the cleaned multi-source stress testing data is to be written according to the weight and the priority; Writing the cleaned multi-source stress testing data into the target data partition; Determine whether data has been consumed from the data partition. If so, write the consumed stress test data to the shard cluster.

[0028] From the above description, we can know that the target data partition for the cleaned multi-source stress testing data to be written is determined based on the weight and priority, so that high-weight data will be assigned to the high-priority partition to ensure fast access. After the data is consumed, it is written to the shard cluster. The data partition is responsible for responding to stress testing requests with low latency. The consumed data is asynchronously persisted to the shard cluster to achieve read-write separation and improve overall throughput.

[0029] The stress test data processing method and system described above are applicable to multi-source stress test data integration scenarios, such as stress test data integration in high-concurrency scenarios such as financial transactions and e-commerce promotions. The following describes the method and system through specific implementation methods: Please refer to Figure 1 , embodiment 1 of the present invention is: A stress test data processing method includes the following steps: S1. Obtain the Uniform Resource Locators (URLs) of multiple stress testing platforms and parse the URLs to obtain ports and paths.

[0030] S2. Determine protocols of the multiple stress testing platforms according to the ports and the paths, and collect multi-source stress testing data from the multiple stress testing platforms according to the protocols.

[0031] The protocol for determining the multiple stress testing platforms according to the ports and the paths includes: Get the preset protocol rule library; A protocol corresponding to the port and the path is determined from the preset protocol rule library as a protocol for the multiple stress testing platforms.

[0032] For example, if the protocol matched by the preset protocol rule library is of the REST (Representational State Transfer) type, a GET / POST request is initiated to the corresponding stress testing platform. If the protocol matched by the preset protocol rule library is of the gRPC (a general remote procedure call framework) type, a Channel (communication channel) is established and a Stub (a proxy function that contains a call to a service method) is called. If the protocol matched by the preset protocol rule library is of the WebSocket (a TCP-based communication protocol) type, a long connection is created. If the corresponding protocol cannot be matched from the preset protocol rule library, a ProtocolNotSupportedException (protocol not supported) exception is output.

[0033] In an optional embodiment, the method further includes: Distribute the load of network requests according to the load balancing strategy.

[0034] Specifically, the weight of each stress testing platform is calculated, and an independent thread pool is allocated to each stress testing platform according to the calculated weight.

[0035] The weight calculation formula is: Weight = Current stress testing platform TPS (Transactions Per Second) / Σ Total stress testing platform TPS. This means that the weight of each stress testing platform is dynamically calculated based on its transaction processing capacity relative to the total capacity of all platforms. The maximum concurrency is set to four times the number of CPU cores to fully utilize the capabilities of multi-core processors while avoiding resource contention and performance degradation caused by excessive concurrency.

[0036] S3. Clean the multi-source stress test data using a timestamp alignment algorithm to obtain cleaned multi-source stress test data, specifically including S31-S34: S31. Use Precision Time Protocol (PTP) to synchronize the clocks of the multi-source stress testing data.

[0037] S32. Determine whether the clock synchronization is successful. If so, perform timestamp alignment on the multi-source stress testing data after the clock synchronization is successful to obtain the aligned multi-source stress testing data. If not, calculate the Network Time Protocol (NTP) offset, and use a linear regression algorithm to compensate for the NTP offset to obtain a compensation value. Perform timestamp alignment on the multi-source stress testing data according to the compensation value to obtain the aligned multi-source stress testing data.

[0038] S33: Calculate the mean and standard deviation of the aligned multi-source stress testing data.

[0039] S34 . Perform anomaly identification on the aligned multi-source stress testing data based on the mean and the standard deviation according to a triple standard deviation (3σ) rule to obtain abnormal data, and mark the abnormal data in the aligned multi-source stress testing data to obtain cleansed multi-source stress testing data.

[0040] Specifically, transient anomaly data corresponding to network outages is marked, and all other data except for these transient anomaly data is removed to obtain cleaned multi-source stress testing data. This is because network outages are often transient and recoverable (such as a router restart). Retaining the marking allows root cause tracing without affecting normal processes. Other persistent anomalies, such as server downtime, are discarded and alerted to avoid interference from dirty data.

[0041] S4. Analyze the players corresponding to the cleaned multi-source stress testing data to obtain behavior patterns.

[0042] Among them, the behavioral patterns include high-frequency operations (such as skill release), resource consumption (such as mall shopping), long-link interactions (such as cross-server team formation), etc.

[0043] In an optional implementation, an LSTM (Long Short-Term Memory) model is used to analyze players corresponding to the cleaned multi-source stress testing data to obtain behavior patterns.

[0044] S5. Calculate the weight of each stress testing data in the cleaned multi-source stress testing data according to the behavior pattern.

[0045] Specifically, the data timeliness and abnormal volatility are determined, the behavior correlation is determined according to the behavior pattern, and the weight of each stress testing data in the cleaned multi-source stress testing data is calculated according to the data timeliness, the abnormal volatility, and the behavior correlation.

[0046] The weights are specifically: Weight = α × data timeliness + β × behavior correlation + γ × abnormal volatility; Wherein, α represents the first coefficient, β represents the second coefficient, and γ represents the third coefficient.

[0047] In this embodiment, α=0.5, β=0.3, and γ=0.2, which are optimized by gradient descent.

[0048] Data timeliness means that the data in the last 5 minutes has a higher priority than historical data (time decay factor e -0.1t ).

[0049] Determining the behavior relevance according to the behavior pattern includes: The correlation between the behavior pattern and server indicators (such as TPS and memory leakage) is quantified by the Pearson correlation coefficient to obtain the behavior correlation.

[0050] S6. Storing the cleaned multi-source stress testing data based on a distributed cache architecture.

[0051] Specifically, storing the cleaned multi-source stress testing data based on the distributed cache architecture according to the weights specifically includes S61-S64: S61: Determine data partitions and priorities of the data partitions.

[0052] The data partition is a Kafka partition, and the number of data partitions = the number of stress testing platforms × 2. Parallel processing and message orderliness are achieved through multiple partitions.

[0053] For example, three stress testing platforms correspond to six Kafka partitions to ensure that the data on each platform is independent and orderly.

[0054] S62: Determine a target data partition to which the cleaned multi-source stress testing data is to be written according to the weight and the priority.

[0055] In an optional embodiment, a modulo operation is performed on the number of data partitions using the ID hash value of the stress testing platform corresponding to the cleaned multi-source stress testing data to obtain an operation result, and the target data partition to which the cleaned multi-source stress testing data is to be written is determined based on the operation result.

[0056] S63: Write the cleaned multi-source stress testing data into the target data partition.

[0057] S64. Determine whether data has been consumed from the data partition. If so, write the consumed stress test data into the shard cluster.

[0058] The sharded cluster is a Redis sharded cluster, with each shard storing ≤8 hours of data and a TTL (data time to live) of 72 hours. Each consumer group subscribes to all partitions, using a range allocation strategy.

[0059] Kafka processes real-time streaming data (high throughput, low latency), and Redis provides persistent storage and fast query (memory-level response). The two are connected through consumer groups to form a "buffer-persistence" closed loop.

[0060] Table 1 shows the comparison results between the stress test data processing method of the present invention and the traditional solution in terms of four indicators: protocol compatibility, timestamp alignment error, 1000TPS message delay, and resource occupancy.

[0061] Table 1 Comparison of indicators

[0062] Please refer to Figure 2 , the second embodiment of the present invention is: A stress test data processing system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, each step of the stress test data processing method in the first embodiment is implemented.

[0063] In summary, the present invention provides a stress test data processing method and system, which obtains the uniform resource locators of multiple stress test platforms, parses the uniform resource locators to obtain ports and paths, determines the protocols of the multiple stress test platforms according to the ports and paths, and collects multi-source stress test data from the multiple stress test platforms according to the protocols, uses a timestamp alignment algorithm to clean the multi-source stress test data to obtain cleaned multi-source stress test data, and stores the cleaned multi-source stress test data based on a distributed cache architecture. In this way, multi-source stress test data can be smoothly obtained from different stress test platforms, and the timestamp alignment algorithm can be used to effectively process the collected multi-source stress test data. Finally, the multi-source stress test data is stored based on the distributed cache architecture, which can more effectively integrate the multi-source stress test data, thereby improving the compatibility and accuracy of the multi-source stress test data integration. In addition, through precise clock synchronization, the reliability and accuracy of data processing can be ensured. By calculating the mean and standard deviation and applying the triple standard deviation rule to identify anomalies in the data, abnormal data can be effectively identified and marked, thereby improving data quality and thus improving the accuracy of multi-source stress test data integration.

[0064] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's description and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A stress test data processing method, characterized in that: Including steps: Obtaining uniform resource locators of multiple stress testing platforms and parsing the uniform resource locators to obtain ports and paths; Determining protocols of the multiple stress testing platforms according to the ports and the paths, and collecting multi-source stress testing data from the multiple stress testing platforms according to the protocols; Cleaning the multi-source stress testing data using a timestamp alignment algorithm to obtain cleaned multi-source stress testing data; The cleaned multi-source stress testing data is stored based on a distributed cache architecture.

2. A stress test data processing method according to claim 1, characterized in that: The protocol for determining the multiple stress testing platforms according to the ports and the paths includes: Get the preset protocol rule library; A protocol corresponding to the port and the path is determined from the preset protocol rule library as a protocol for the multiple stress testing platforms.

3. A stress test data processing method according to claim 1, characterized in that: The multi-source stress testing data is cleansed using a timestamp alignment algorithm to obtain the cleansed multi-source stress testing data, including: Using the Precision Time Protocol to synchronize the clocks of the multi-source stress testing data; Determine whether the clock synchronization is successful. If so, perform timestamp alignment on the multi-source stress test data after the clock synchronization is successful to obtain the aligned multi-source stress test data. If not, calculate the network time protocol offset and use a linear regression algorithm to compensate for the network time protocol offset to obtain a compensation value. Perform timestamp alignment on the multi-source stress test data according to the compensation value to obtain the aligned multi-source stress test data. Calculate the mean and standard deviation of the aligned multi-source stress testing data; Anomalies are identified on the aligned multi-source stress testing data based on the mean and the standard deviation according to a triple standard deviation rule to obtain abnormal data, and the abnormal data in the aligned multi-source stress testing data is marked to obtain cleaned multi-source stress testing data.

4. A stress test data processing method according to claim 1, characterized in that: Before storing the cleaned multi-source stress testing data based on the distributed cache architecture, the method further includes: Analyze the players corresponding to the cleaned multi-source stress testing data to obtain behavior patterns; Calculating the weight of each stress testing data in the cleaned multi-source stress testing data according to the behavior pattern; The storing of the cleaned multi-source stress testing data based on the distributed cache architecture includes: The cleaned multi-source stress testing data is stored based on a distributed cache architecture according to the weight.

5. A stress test data processing method according to claim 4, characterized in that: Storing the cleaned multi-source stress testing data based on the distributed cache architecture according to the weights includes: Determining data partitions and priorities of the data partitions; Determine a target data partition to which the cleaned multi-source stress testing data is to be written according to the weight and the priority; Writing the cleaned multi-source stress testing data into the target data partition; Determine whether data has been consumed from the data partition. If so, write the consumed stress test data to the shard cluster.

6. A stress test data processing system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the following steps are implemented: Obtaining uniform resource locators of multiple stress testing platforms and parsing the uniform resource locators to obtain ports and paths; Determining protocols of the multiple stress testing platforms according to the ports and the paths, and collecting multi-source stress testing data from the multiple stress testing platforms according to the protocols; Cleaning the multi-source stress testing data using a timestamp alignment algorithm to obtain cleaned multi-source stress testing data; The cleaned multi-source stress testing data is stored based on a distributed cache architecture.

7. A stress test data processing system according to claim 6, characterized in that: The protocol for determining the multiple stress testing platforms according to the ports and the paths includes: Get the preset protocol rule library; A protocol corresponding to the port and the path is determined from the preset protocol rule library as a protocol for the multiple stress testing platforms.

8. A stress test data processing system according to claim 6, characterized in that: The multi-source stress testing data is cleansed using a timestamp alignment algorithm to obtain the cleansed multi-source stress testing data, including: Using the Precision Time Protocol to synchronize the clocks of the multi-source stress testing data; Determine whether the clock synchronization is successful. If so, perform timestamp alignment on the multi-source stress test data after the clock synchronization is successful to obtain the aligned multi-source stress test data. If not, calculate the network time protocol offset and use a linear regression algorithm to compensate for the network time protocol offset to obtain a compensation value. Perform timestamp alignment on the multi-source stress test data according to the compensation value to obtain the aligned multi-source stress test data. Calculate the mean and standard deviation of the aligned multi-source stress testing data; Anomalies are identified on the aligned multi-source stress testing data based on the mean and the standard deviation according to a triple standard deviation rule to obtain abnormal data, and the abnormal data in the aligned multi-source stress testing data is marked to obtain cleaned multi-source stress testing data.

9. A stress test data processing system according to claim 6, characterized in that: Before storing the cleaned multi-source stress testing data based on the distributed cache architecture, the method further includes: Analyze the players corresponding to the cleaned multi-source stress testing data to obtain behavior patterns; Calculating the weight of each stress testing data in the cleaned multi-source stress testing data according to the behavior pattern; The storing of the cleaned multi-source stress testing data based on the distributed cache architecture includes: The cleaned multi-source stress testing data is stored based on a distributed cache architecture according to the weight.

10. A stress test data processing system according to claim 9, characterized in that: Storing the cleaned multi-source stress testing data based on the distributed cache architecture according to the weights includes: Determining data partitions and priorities of the data partitions; Determine a target data partition to which the cleaned multi-source stress testing data is to be written according to the weight and the priority; Writing the cleaned multi-source stress testing data into the target data partition; Determine whether data has been consumed from the data partition. If so, write the consumed stress test data to the shard cluster.

Images