Network quality data noise reduction method and device and related product

Abstract

The invention discloses a network quality data noise reduction method and device and a related product, and the method comprises the steps: obtaining network quality data corresponding to a detection flow according to the detection flow sent by a source server in a target data center to a to-be-tested server; according to a packet loss index screening condition, identifying target network quality data of which the actual packet loss index meets the packet loss index screening condition from the obtained network quality data; determining a target detection flow corresponding to the target network quality data according to a corresponding relationship between the network quality data and the detection flow; and based on the determined target detection flow, carrying out noise reduction on the obtained network quality data. According to the invention, the noise data in the network quality data causing false alarm can be eliminated. Therefore, when the de-noised network quality data is detected, the requirement on the length of the detection time window is reduced, and an alarm prompt about the network quality data can be sent out in time, so that the fault processing efficiency of the network is improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a network quality data noise reduction method, device and related products. Background Art

[0002] Data can be transmitted between servers, and data transmission will be affected by the network quality of the communication link between servers. For example, a data center includes multiple business servers, and there are communication links between business servers. In order for the data center to provide network services stably and permanently, it is necessary to detect the health of the network quality data of the communication link.

[0003] In the related art, when detecting network quality data, in order to avoid false alarms, the detection time window is usually extended, and an alarm about the network quality data is issued only when the network quality data has a long period of packet loss. However, in the above method, it may take a while to issue an alarm after the network quality data fails, which is easy to affect the timeliness of the alarm, thereby reducing the efficiency of network fault handling. Summary of the invention

[0004] The embodiments of the present application provide a network quality data noise reduction method, device and related products, which are intended to promptly issue an alarm prompt about the network quality data when a network quality data failure occurs, thereby improving the efficiency of network fault handling.

[0005] The first aspect of the present application provides a network quality data noise reduction method, comprising:

[0006] According to the detection stream sent by the source server in the target data center to the server to be tested, network quality data corresponding to the detection stream is obtained; the target data center includes multiple business servers, and the target data center provides network services through data interaction between the multiple business servers; one detection stream corresponds to two different business servers as the source server and the server to be tested among the multiple business servers;

[0007] According to the packet loss index screening condition, identifying the target network quality data whose actual packet loss index satisfies the packet loss index screening condition from the acquired network quality data;

[0008] Determining a target detection flow corresponding to the target network quality data according to a correspondence between the network quality data and the detection flow;

[0009] Based on the determined target detection flow, the acquired network quality data is denoised.

[0010] A second aspect of the present application provides a network quality data noise reduction device, comprising:

[0011] A quality data acquisition module is used to acquire network quality data corresponding to a detection flow according to a detection flow sent by a source server in a target data center to a server to be tested; the target data center includes a plurality of service servers, and the target data center provides network services through data interaction between the plurality of service servers; one detection flow corresponds to two different service servers as a source server and a server to be tested among the plurality of service servers;

[0012] A quality data identification module is used to identify target network quality data whose actual packet loss index satisfies the packet loss index screening condition from the acquired network quality data according to the packet loss index screening condition;

[0013] A target detection flow determination module, used to determine the target detection flow corresponding to the target network quality data according to the corresponding relationship between the network quality data and the detection flow;

[0014] A data denoising module is used to denoise the acquired network quality data based on the determined target detection flow.

[0015] A third aspect of the present application provides a network quality data noise reduction device, the device comprising a processor and a memory:

[0016] The memory is used to store a computer program and transmit the computer program to the processor;

[0017] The processor is used to execute the steps of the network quality data denoising method provided in the first aspect according to the instructions in the computer program.

[0018] In a fourth aspect, the present application provides a computer-readable storage medium, which is used to store a computer program. When the computer program is executed by a network quality data denoising device, the steps of the network quality data denoising method provided in the first aspect are implemented.

[0019] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a network quality data denoising device, implements the steps of the network quality data denoising method provided in the first aspect.

[0020] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0021] In the technical solution of the present application, first, according to the detection stream sent by the source server in the target data center to the server to be tested, the network quality data corresponding to the detection stream is obtained. Among them, the target data center includes multiple business servers, and the target data center provides network services through data interaction between multiple business servers; one detection stream corresponds to two different business servers as the source server and the server to be tested among the multiple business servers. Then, according to the packet loss index screening condition, the target network quality data whose actual packet loss index meets the packet loss index screening condition is first identified from the acquired network quality data. According to the correspondence between the network quality data and the detection stream, the target detection stream corresponding to the target network quality data is determined. Since the target network quality data is identified from the acquired network quality data based on the packet loss index screening condition, and the setting of the packet loss index screening condition takes into account the actual packet loss of the network quality data, the target detection stream corresponding to the target network quality data needs to be further analyzed to see whether it is a detection stream that causes noise data. Subsequently, based on the determined target detection stream, the acquired network quality data is denoised. It can be seen that denoising the acquired network quality data can eliminate noise data that causes false alarms. Therefore, when detecting the network quality data after noise reduction, there is no need to extend the detection time window to avoid false alarms, which reduces the requirements for the length of the detection time window. It is possible to issue alarm prompts about abnormal network quality data in a timely manner, thereby improving the efficiency of network fault handling after a network failure occurs.

[0022] In addition, the server serving as the source server is also a business server in the target data center. The detection path corresponding to the detection stream sent by the source server to the server to be tested is consistent with the actual business traffic path of the communication link between the source server and the server to be tested. Therefore, the obtained network quality data can truly reflect the network quality of the communication link, which is conducive to further reducing the noise data in the network quality data, thereby improving the effective utilization rate of the obtained network quality data. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A schematic diagram of a curve of packet loss rate in network quality data;

[0024] Figure 2 A schematic diagram of a local curve of a packet loss rate in network quality data;

[0025] Figure 3 A scenario architecture diagram of a network quality data noise reduction method provided in an embodiment of the present application;

[0026] Figure 4 A schematic diagram of an application scenario of a network quality data noise reduction method provided in an embodiment of the present application;

[0027] Figure 5 A flowchart of a network quality data noise reduction method provided in an embodiment of the present application;

[0028] Figure 6 A schematic diagram of the number of packet losses of a service server provided in an embodiment of the present application;

[0029] Figure 7 A schematic diagram of denoising the acquired network quality data provided in an embodiment of the present application;

[0030] Figure 8 A schematic diagram of denoising the remaining network quality data after exclusion provided in an embodiment of the present application;

[0031] Fig. 9 A flowchart of a method for determining a source server or a server to be tested provided in an embodiment of the present application;

[0032] Fig.10 A schematic diagram of a curve of packet loss rate of network quality data after noise reduction provided in an embodiment of the present application;

[0033] Fig.11 A schematic diagram of a local curve of the packet loss rate of network quality data after noise reduction provided in an embodiment of the present application;

[0034] Fig.12 A schematic diagram of the structure of a network quality data noise reduction device provided in an embodiment of the present application;

[0035] Fig.13 A schematic diagram of the structure of the server in the embodiment of the present application;

[0036] Fig.14 A schematic diagram of the structure of a terminal device in an embodiment of the present application. DETAILED DESCRIPTION

[0037] By detecting the network quality data of the communication link between servers, we can observe the health status of the communication link, such as whether there are delays or packet loss in the communication link. However, network quality data usually contains noise data, which refers to data that causes unstable network quality data due to manual restart of the server, network jitter, etc. Specifically, if the technician proactively shuts down the server, the network quality data related to the server will be temporarily abnormal, but when the technician turns on the server again, its network quality data will return to normal. Therefore, if an alarm about the network quality data is issued immediately after such an abnormal situation, it will be easy to cause a false alarm.

[0038] For details, please refer to Figure 1 A schematic diagram of packet loss rate in network quality data is shown in FIG. Figure 1As shown in the figure, the packet loss rate in the network quality data will appear as a burr-like packet loss rate data between "00:05" and "00:10". For easier understanding, please refer to Figure 2 A schematic diagram of a local curve of packet loss rate in network quality data is shown. Figure 2 It can be understood as Figure 1 The curve diagram after magnification is as follows: Figure 2 As shown in the figure, there are a lot of glitches between "15:30" and "15:55", that is, the packet loss rate corresponding to this moment is greater than 0%. However, the reason for the packet loss may be due to high server load or server restart, not the failure of the communication link between servers. Because if the communication link fails, the packet loss must be continuous. Therefore, if an alarm is issued for the packet loss rate data corresponding to each glitch, a large number of false alarms may occur.

[0039] At present, in order to avoid false alarms, the method of extending the detection time window of network quality data is usually adopted, that is, when the network quality data has abnormal conditions such as long-term packet loss, an alarm will be issued to the network operation and maintenance personnel to prompt the network operation and maintenance personnel to handle the fault. However, this method of extending the detection time window of network quality data is prone to the situation that after the communication link network fails, it takes a while for the network operation and maintenance personnel to receive the alarm about the network quality data, which is easy to affect the timeliness of the alarm, extend the cycle of network fault handling, and thus reduce the efficiency of network fault handling.

[0040] In view of the above problems, a network quality data denoising method, device and related products are provided in the present application, the purpose of which is to issue an alarm about the network quality data in time when a fault occurs in the network quality data, so as to improve the efficiency of network fault handling. In the technical solution provided in the present application, the network quality data corresponding to the detection stream is first obtained according to the detection stream sent by the source server in the target data center to the server to be tested. Among them, the target data center includes multiple business servers, and the target data center provides network services through data interaction between multiple business servers; one detection stream corresponds to two different business servers as source servers and servers to be tested in multiple business servers. Then, according to the packet loss index screening condition, the target network quality data whose actual packet loss index meets the packet loss index screening condition is first identified from the acquired network quality data. According to the corresponding relationship between the network quality data and the detection stream, the target detection stream corresponding to the target network quality data is determined. Then, based on the determined target detection stream, the acquired network quality data is denoised. It can be seen that denoising the acquired network quality data can eliminate the noise data that causes false alarms. Therefore, when detecting the network quality data after noise reduction, there is no need to extend the detection time window to avoid false alarms, which reduces the requirements for the length of the detection time window. It is possible to issue alarm prompts about abnormal network quality data in a timely manner, thereby improving the efficiency of network fault handling after a network failure occurs.

[0041] The execution subject of the network quality data denoising method provided in the embodiment of the present application may be a terminal device. As an example, the terminal device may specifically include but is not limited to a mobile phone, a desktop computer, a tablet computer, a notebook computer, a PDA, an intelligent voice interaction device, a smart home appliance, a vehicle-mounted terminal, an aircraft, etc. The execution subject of the network quality data denoising method provided in the embodiment of the present application may also be a server. In addition, the network quality data denoising method provided in the embodiment of the present application may also be executed collaboratively by the terminal device and the server, that is, the server may obtain the network quality data corresponding to the detection stream according to the detection stream sent by the source server in the target data center to the server to be tested, and the server then sends the obtained network quality data to the terminal device; so that the terminal device can identify the target network quality data whose actual packet loss index meets the packet loss index screening condition from the obtained network quality data according to the packet loss index screening condition; the terminal device then determines the target detection stream corresponding to the target network quality data according to the corresponding relationship between the network quality data and the detection stream; then the terminal device can reduce the noise of the obtained network quality data based on the determined target detection stream. Among them, the terminal device and the server can be directly or indirectly connected through wired or wireless communication, and this application is not limited here. Therefore, the embodiments of the present application do not limit the implementation subject of the technical solution of the present application.

[0042] In addition, the server serving as the source server is also a business server in the target data center. The detection path corresponding to the detection stream sent by the source server to the server to be tested is consistent with the actual business traffic path of the communication link between the source server and the server to be tested. Therefore, the obtained network quality data can truly reflect the network quality of the communication link, which is conducive to further reducing the noise data in the network quality data, thereby improving the effective utilization rate of the obtained network quality data.

[0043] Figure 3 The following is an exemplary scenario architecture diagram of a network quality data noise reduction method, which includes a server and various terminal devices. Figure 3 The server shown can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. In addition, the server can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0044] In order to facilitate understanding of the technical solution of the present application, the network quality data denoising method provided in the embodiment of the present application is introduced below in combination with actual application scenarios.

[0045] See also Figure 4 , which is a schematic diagram of an application scenario of a network quality data noise reduction method provided by an embodiment of the present application. Figure 4 The application scenario shown includes terminal device 100, data centers of city 1 and city 2. DCI is the abbreviation of Data Center Interconnect, which means that city 1 and city 2 can achieve network interconnection through cross-data centers. BR refers to city-level switches, DR refers to regional-level switches, CUF refers to computer room-level switches, and ToR refers to switches inside computer rooms. One ToR is connected to multiple business servers. Figure 4 As shown, it includes communication link 1, indicating that there is at least one business server that can exchange data with another business server through communication link 1; it also includes communication link 2, which also indicates that at least one business server can exchange data with another business server through communication link 2. The network quality data denoising method provided in the embodiment of the present application can be used for Figure 4 When the network quality data of the communication link 1 and the network quality data of the communication link 2 are detected, noise reduction is performed on the obtained network quality data.

[0046] Taking the detection of communication link 1 and communication link 2 as an example, business server A1 is used as the source server, and business server B1 is used as the server to be tested; business server A2 is used as the source server, and business server B2 is used as the server to be tested. Then, according to the detection stream 1 sent by business server A1 to business server B1, and the detection stream 2 sent by business server A2 to business server B2, the terminal device 100 can obtain the network quality data 1 of communication link 1 and the network quality data 2 of communication link 2; the terminal device 100 then identifies the target network quality data whose actual packet loss index meets the packet loss index screening condition from the obtained network quality data 1 and network quality data 2 according to the packet loss index screening condition; then the terminal device 100 determines the target detection stream corresponding to the target network quality data according to the corresponding relationship between the network quality data and the detection stream; finally, the terminal device 100 performs noise reduction on the obtained network quality data based on the determined target detection stream.

[0047] Next, taking the terminal device as the execution subject, the network quality data noise reduction method provided in the embodiment of the present application is specifically described.

[0048] See also Figure 5 , which is a flow chart of a network quality data noise reduction method provided by an embodiment of the present application. Figure 5 The network quality data noise reduction method shown includes the following steps:

[0049] S501: Acquire network quality data corresponding to a detection stream sent by a source server in a target data center to a server to be tested.

[0050] In the embodiment of the present application, the target data center refers to the data center where the network quality needs to be detected, and does not specifically refer to any data center. The target data center includes multiple business servers, and the target data center can provide network services through data interaction between multiple business servers. As an example, the target data center can be the data center mentioned above that realizes network interconnection between city 1 and city 2, and the present application does not limit this.

[0051] In the embodiment of the present application, the source server refers to a server serving as a detection source, the server to be tested refers to a server serving as a detection target, and the detection source may send a detection stream to the detection target. The detection stream refers to detection data sent by the source server to the server to be tested. The network quality data corresponding to the detection stream is used to characterize the network quality of the communication link between the source server and the server to be tested.

[0052] As an example, the network quality data may include a packet loss rate, which refers to the ratio of lost data packets to sent data packets when the source server sends detection data to the target server. Specifically, the detection stream sent by the source server to the target server includes 100 data packets, but the target server only receives 90 data packets, indicating that 10 data packets are lost during the detection data transmission process, and the packet loss rate is: (100-90) / 100=10%. In addition, the network quality data may also include transmission delay data, etc., which is not limited in this application.

[0053] It should be emphasized that the same business server in the target data center can be used as a source server or as a server to be tested. And the same business server may correspond to different detection streams. This application does not limit this. For example, business server A in the target data center can be used as the source server corresponding to detection stream 1, as the source server corresponding to detection stream 2, and as the server to be tested corresponding to detection stream 3.

[0054] S502: According to the packet loss index screening condition, identify target network quality data whose actual packet loss index satisfies the packet loss index screening condition from the acquired network quality data.

[0055] It is understandable that the target data center usually includes a large number of business servers. In order to detect the network quality data of the target data center, even if only some business servers are selected as source servers and servers to be tested, network quality data corresponding to dozens of detection flows may be obtained.

[0056] Therefore, in order to further identify the target network quality data that needs to be further analyzed from the acquired network quality data, a packet loss index screening condition can be pre-set to screen out the target network quality data that may need to be denoised. As an example, the packet loss index screening condition can be that the packet loss rate in the network quality data is greater than 0%, then the target network quality data whose actual packet loss rate is greater than 0% needs to be identified from the acquired network quality data summary. This application does not limit this.

[0057] It should be noted that the packet loss index-based screening condition is to first screen out the target network quality data to facilitate the subsequent steps. However, the remaining network quality data after screening is not excluded, and all the obtained network quality data needs to be denoised later.

[0058] S503: Determine a target detection flow corresponding to the target network quality data according to the correspondence between the network quality data and the detection flow.

[0059] Different detection flows correspond to different network quality data. After the target network quality data is determined, the target detection flow corresponding to the target network quality data can be determined based on the corresponding relationship between the two.

[0060] S504: Based on the determined target detection flow, denoising is performed on the acquired network quality data.

[0061] In the process of detecting network quality, the business server as the server to be tested may not be able to receive all the data packets in the detection stream due to high load or restart, that is, detection packet loss occurs. This will cause an abnormal packet loss rate in the network quality data and issue an alarm prompt, but the network quality data will return to normal due to the successful restart of the business server or the reduction of load, and the network maintenance personnel do not need to perform fault recovery on the communication link. In other words, the obtained network quality data cannot fully represent the network quality, and there may be noise data that causes false alarms. Therefore, it is necessary to reduce the noise of the obtained network quality data. The determined target detection stream is conducive to determining the noise data as soon as possible.

[0062] As mentioned above, a business server may correspond to multiple detection streams. If the network quality data corresponding to multiple detection streams corresponding to a business server all have detection packet loss, it indicates that the business server may be affected by server-specific factors such as restart, that is, the network quality data corresponding to the business server is likely to be noise data. Therefore, the number of packet losses corresponding to multiple business servers corresponding to the target detection stream can be calculated to reduce the noise of the obtained network quality data.

[0063] In practical applications, when the actual packet loss rate of network quality data is 0%, it indicates that the network quality is normal. However, when the actual packet loss rate of network quality data is relatively small, it will not have much impact on the network quality. In other words, network quality data with a relatively small actual packet loss rate will not cause false alarms, and will not be noise data. Therefore, in a possible implementation of the present application, the packet loss indicator screening condition can be that the actual packet loss rate in the network quality data is greater than the first packet loss rate threshold, and the target network quality data with an actual packet loss rate greater than the first packet loss rate threshold is first screened out. For example, the first packet loss rate threshold can be 0, and the present application does not limit this. Accordingly, based on the determined target detection flow, the step of denoising the acquired network quality data can be specifically subdivided into the following steps:

[0064] Step 1: According to the correspondence between the detection flow and the service server, determine multiple service servers corresponding to the target detection flow.

[0065] In practical applications, the target detection stream is usually multiple target detection streams. Therefore, multiple business servers corresponding to the target detection stream are determined, that is, two business servers corresponding to the multiple target detection streams are determined; the two business servers corresponding to one target detection stream include the source server that sends the target detection stream and the server to be tested that receives the target detection stream.

[0066] Step 2: Based on multiple service servers corresponding to the target detection flow, the number of packet losses of each service server is calculated.

[0067] It should be understood that a service server may correspond to multiple detection streams, so the number of packet losses of each service server is the sum of the number of packet losses when each service server sends the detection stream as a source server and the number of packet losses when the service server receives the detection stream when it is a server under test.

[0068] Step 3: Based on the calculated number of packet losses of the service server, the obtained network quality data is denoised.

[0069] The packet loss index screening condition mentioned above is not used to screen out network quality data. Therefore, based on the calculated number of packet losses of the business server, all the obtained network quality data can be denoised.

[0070] It can be seen that first, based on the packet loss index screening conditions, the target network quality data that needs to be further analyzed is determined from a large amount of network quality data, which can reduce the amount of calculation of the number of packet losses of the service server in the subsequent calculation, saving computing resources. Then, based on the number of packet losses corresponding to multiple service servers corresponding to the target detection flow, the noise data that causes the false alarm can be determined from the acquired network quality data. The network quality data after noise reduction is more real and accurate, which reduces the requirements for the length of the detection time window, and is conducive to issuing correct alarms about network quality data.

[0071] As an example, the terminal device may randomly select some service servers from multiple service servers in the target data center for mutual detection. Figure 6 , which is a schematic diagram of the number of packet loss of a service server provided in an embodiment of the present application. The packet loss index screening condition can be that the actual packet loss rate in the network quality data is greater than 0%. Figure 6As shown, the detection stream may include detection stream 1, detection stream 2, detection stream 3, detection stream 4 and detection stream 5. Detection stream 1 corresponds to business server A1 and business server B1. The direction of detection stream 1 is from business server A1 to business server B1, indicating that business server A1 as the source server sends detection stream 1 to business server B1 as the server to be tested. The packet loss rate in the network quality data corresponding to detection stream 1 is 100%. The packet loss rates corresponding to detection stream 2, detection stream 3, detection stream 4 and detection stream 5 can be specifically referred to Figure 6 As shown, no further details will be given here. Figure 6 As shown, the service server A1 corresponds to the detection stream 1 and the detection stream 2, and the remaining service servers correspond to only one detection stream.

[0072] Combination Figure 6 As shown, based on the packet loss index screening condition, the target network quality data with an actual packet loss rate greater than 0% is first identified, and then the target detection stream corresponding to the target network quality data is determined, and the target detection stream includes detection stream 1, detection stream 2, detection stream 4, and detection stream 5. Then, multiple business servers corresponding to detection stream 1, detection stream 2, detection stream 4, and detection stream 5 are determined, and the number of packet losses of each business server is calculated. Figure 6 As shown in the figure, the number of packet losses corresponding to the service server A1 is 2, and the service server A1 does not receive the detection stream, that is, the number of packet losses when the service server A1 receives the detection stream is 0. Therefore, the number of packet losses corresponding to the service server A1 is the number of packet losses 2 when the service server A1 sends the detection stream 1 and the detection stream 2. The number of packet losses corresponding to the service servers A3, A4, and B1-B5 are all 1. Finally, based on the calculated number of packet losses of the service servers, all the acquired network quality data are denoised.

[0073] In the embodiment of the present application, there are multiple possible implementations of the above-mentioned step 3, which are introduced below. It should be noted that the implementations given in the following introduction are only for illustrative purposes and do not represent all implementations of the embodiment of the present application.

[0074] Then the first optional implementation method of step 3 is: based on the number of packet losses of the business server corresponding to the target detection flow, the number of packet losses of each business server is compared with the packet loss threshold. If the number of packet losses of the business server is greater than the packet loss threshold, the business server is used as the target server; then, according to the correspondence between the detection flow and the business server, the detection flow to be denoised corresponding to the target server is determined; then, according to the correspondence between the network quality data and the detection flow, the network quality data to be denoised corresponding to the detection flow to be denoised is determined from the acquired network quality data; then, the network quality data to be denoised is excluded from the acquired network quality data. Among them, the packet loss threshold can be pre-set by technical personnel such as network operation and maintenance personnel based on experience, and this application does not limit this.

[0075] The second optional implementation method of step 3 is: first, based on the number of packet losses of each business server, sort the business servers in order from small to large or from large to small, and take the business server corresponding to the maximum number of packet losses as the target server; then, based on the correspondence between the detection flow and the business server, determine the detection flow to be denoised corresponding to the target server; then, based on the correspondence between the network quality data and the detection flow, determine the network quality data to be denoised corresponding to the detection flow to be denoised from the acquired network quality data; finally, exclude the network quality data to be denoised from the acquired network quality data.

[0076] It should be noted that, for the above two optional implementation methods, the terminal device can choose one or a combination of multiple implementation methods, and this application does not impose any restrictions on this.

[0077] It can be seen that based on the calculated number of packet losses of the business servers, the network quality data to be denoised corresponding to the business servers with the maximum number of packet losses is determined and excluded, or the network quality data to be denoised corresponding to the business servers with a number of packet losses greater than a threshold is determined and excluded. This can identify and exclude noise data in the acquired network quality data, and effectively denoise the noise data in the acquired network quality data caused by abnormalities in the server to be tested or network jitter, so that the denoised network quality data can accurately reflect the health status of the communication link in the target data center, and then when a network failure occurs in the target data center, correct alarms about the network quality data can be issued in a timely manner, thereby improving the efficiency of network fault handling.

[0078] As an example, see Figure 7, which is a schematic diagram of denoising the acquired network quality data provided by an embodiment of the present application. Based on the above example, the target detection stream includes detection stream 1, detection stream 2, detection stream 4 and detection stream 5. The number of packet losses corresponding to business server A1 is 2, and the number of packet losses corresponding to business server A3, business server A4 and business server B1-business server B5 are all 1. Then, the business servers can be sorted in order from large to small according to the number of packet losses, and the business server A1 with the largest number of packet losses is determined as the target server; then, according to the correspondence between the detection stream and the business server, the detection stream 1 and detection stream 2 corresponding to the business server A1 are used as the detection streams to be denoised; then, according to the correspondence between the network quality data and the detection stream, the packet loss rates corresponding to the detection stream 1 and the detection stream 2 are determined from the acquired network quality data as the network quality data to be denoised; finally, the packet loss rate of 100% corresponding to the detection stream 1 and the packet loss rate of 80% corresponding to the detection stream 2 are excluded. Finally, the remaining network quality data after exclusion can be summarized as the denoised network quality data.

[0079] It is understandable that there may still be noise data in the remaining network quality data after the exclusion in the above steps. Therefore, in a possible implementation of the present application, after excluding the network quality data to be de-noised, the network quality data with the actual packet loss rate as the maximum packet loss rate can be excluded from the remaining network quality data after the exclusion.

[0080] As an example, see Figure 8 , which is a schematic diagram of a method of reducing noise of the remaining network quality data after exclusion provided by an embodiment of the present application. Based on the above example, the remaining network quality data after exclusion includes the network quality data corresponding to detection stream 3, detection stream 4, and detection stream 5, respectively. The network quality data with the actual packet loss rate of the maximum packet loss rate can be excluded in descending order, such as Figure 8 As shown, the packet loss rate corresponding to detection stream 5 is the largest, so the network quality data corresponding to detection stream 5 is excluded, and finally the network quality data corresponding to detection stream 3 and the network quality data corresponding to detection stream 4 remain.

[0081] In another possible implementation of the present application, after excluding the network quality data to be de-noised, network quality data whose actual packet loss rate is greater than the second packet loss rate threshold may also be excluded from the remaining network quality data after exclusion. The second packet loss rate threshold may be pre-set by technical personnel such as network operation and maintenance personnel, and the present application does not impose any restrictions on this.

[0082] It can be seen that, based on the calculated number of packet losses, the obtained network quality data is subjected to denoising, and further denoising is performed on the remaining network quality data after exclusion based on the packet loss rate in the remaining network quality data after exclusion. This enables the remaining network quality data to reflect the health status of the communication link in the target data center more accurately, which is conducive to issuing accurate alarm prompts in a timely manner.

[0083] In the related art, a fixed probe is usually selected, that is, an independent server as the detection source, and some or all of the business servers in the target data center are used as detection targets (that is, the servers to be tested mentioned above). A fixed probe is a dedicated server used for detection, which can avoid the problem of detection packet loss due to unstable detection sources. However, when the fixed probe sends a detection stream to the business server in the target data center, the detection path passed by the detection stream is not exactly the same as the path passed by the actual business traffic, and may not reflect the true and accurate network quality. It is easy for the fixed probe to obtain normal detection results, but the actual communication link has failed.

[0084] In order to solve this problem, in an embodiment of the present application, some or all of the service servers are selected from the target data center for mutual detection. In other words, the detection source and the detection target are both service servers in the target data center. The detection path from the detection source to the detection target is completely consistent with the actual path of the service traffic between the service servers, so the obtained network quality data can truly reflect the health status of the target data center network.

[0085] In practical applications, it is considered that the business servers in the target data center may have detection packet loss due to their own instability, which in turn leads to an increase in noise data in the network quality data. In order to solve this problem, when determining the source server or the server to be tested from the target data center, the influence of the business server's own factors can be comprehensively considered. Therefore, in a possible implementation of the present application, the source server or the server to be tested can be specifically determined in the following manner: according to the server screening condition, the business servers that meet the server screening condition are screened out from the multiple business servers in the target data center, and the remaining business servers are obtained as the source server or the server to be tested. It can be understood that the business server's own factors may include the hardware settings, software settings or usage performance of the business server, etc. Therefore, the server screening condition can be a screening condition regarding the hardware settings, software settings or usage performance of multiple business servers in the target data center.

[0086] In a possible implementation manner of the present application, the server screening condition may include at least one of a first server screening condition, a second server screening condition, a third server screening condition, and a fourth server screening condition.

[0087] Among them, the above-mentioned first server screening condition can be that the usage rate of the central processing unit of the business server is greater than the usage rate threshold, or the data transmission volume of the business server is greater than the transmission volume threshold. When a high-load business server is used as a source server or a server to be tested, it is easy to cause detection packet loss due to its own excessive load, so it is necessary to screen out the high-load business server. As an example, the usage rate threshold can be 80%, and the transmission volume threshold can be 40% of the bandwidth of the computer room, which is not limited in this application.

[0088] The above-mentioned second server screening condition is that the identifier of the business server is in the server blacklist. The fact that the identifier of the business server is in the server blacklist indicates that when the business server is used as a source server or a server to be tested, it is easy to cause detection packet loss. As an example, the identifier in the server blacklist may be added manually. When technical personnel such as network operation and maintenance personnel find a faulty business server, their corresponding identifier can be added to the server blacklist to prevent the faulty business server from being identified as a source server or a server to be tested, thereby increasing the noise data in the network quality data.

[0089] The third server screening condition is that the uplink switch of the service server does not contain the correct topology information. That is, when querying the topology information contained in the uplink switch of the service server, the topology information of the uplink switch of the service server is displayed as wrong, or the topology information of the uplink switch of the service server cannot be queried, indicating that the service server may have information entry errors or is in a state of maintenance and replacement. If it is used as a source server or a server to be tested, it is easy to affect the accuracy of the network quality data due to its own instability.

[0090] The fourth server screening condition is that a virtual machine is deployed on the business server. Considering that business services deployed with virtual machines may often be rebuilt or shut down by users, business services deployed with virtual machines are not stable enough when used as source servers or servers to be tested, which can easily affect the accuracy of network quality data. For example, virtual machines include but are not limited to KVM virtual machines (Keyboard Video Mouse), CVM virtual machines (Cloud Virtual Machine), etc.

[0091] As an example, the server screening condition may include four server screening conditions: a first server screening condition, a second server screening condition, a third server screening condition, and a fourth server screening condition. Fig. 9 , which is a flow chart of a method for determining a source server or a server to be tested provided by an embodiment of the present application. Fig. 9 As shown, the following steps may be specifically included:

[0092] S1: Screening out a business server that meets a first server screening condition from multiple business servers in a target data center;

[0093] S2: Filter out the service servers that meet the second server filtering condition from the multiple service servers in the target data center;

[0094] S3: Screening out the business servers that meet the third server screening condition from the multiple business servers in the target data center;

[0095] S4: filtering out the business servers that meet the fourth server filtering condition from the multiple business servers in the target data center;

[0096] S5: Determine the remaining business servers after screening as source servers or servers to be tested.

[0097] It should be noted that the embodiment of the present application does not limit the execution order of S1-S4. S1-S4 can be executed sequentially according to the preset execution order, or S1-S4 can be executed simultaneously. The present application does not limit this.

[0098] It can be seen that based on multiple server screening conditions, the business servers in the target data center that may be unstable due to server factors are screened out, and the remaining business servers after screening are determined as source servers or servers to be tested. This can not only ensure the stability of the source server or the server to be tested and avoid the generation of noise data, but also make the detection path from the detection source to the detection target completely consistent with the actual path of the business traffic between the business servers, thereby improving the accuracy of the obtained network quality data, that is, further reducing the noise data in the network quality data.

[0099] In addition, for the server blacklist mentioned above, the identifiers therein may also be automatically imported by the system. In a possible implementation of the present application, the network quality data denoising method may also include: identifying, from the acquired network quality data, the network quality data to be processed whose actual packet loss rate is greater than 0 for m consecutive detection moments within a first preset time period; wherein m can be any integer greater than 1. Then, based on the correspondence between the network quality data and the detection stream, determine the detection stream to be processed corresponding to the network quality data to be processed. Then, based on the correspondence between the detection stream and the business server, determine the server to be tested that receives the detection stream to be processed from multiple business servers. Subsequently, add the identifier of the server to be tested that receives the detection stream to be processed to the server blacklist.

[0100] The detection moment refers to the moment of detecting the network quality, which can be understood as the moment of detecting the network quality data of the target data center in the embodiment of the present application. For example, the m detection moments can be each moment within the first preset time period, indicating that the network quality data is continuously detected at this time; the m detection moments can also be moments with a preset time interval within the first preset time period; the m detection moments can also be moments randomly selected within the first preset time period. This application does not limit this.

[0101] As an example, the first preset time period can be 12 hours, and the m detection moments can be each moment within the first preset time period. Then, the network quality data to be processed whose packet loss rate is greater than 0 at each moment within 12 hours can be identified from the acquired network quality data, indicating that the server to be tested corresponding to the network quality data to be processed may be unstable due to network isolation or system settings, and is not suitable as a source server or a server to be tested. The terminal device can then automatically add the identifier of the server to be tested to the server blacklist.

[0102] It can be seen that the terminal device can automatically add the identifier of the unstable business server to the server blacklist, further avoiding the unstable business server from being identified as the source server or the server to be tested, which is conducive to reducing noise data and obtaining more accurate network quality data.

[0103] In addition, after the acquired network quality data is denoised, it is necessary to detect the denoised network quality data so that an alarm prompt can be issued after a network failure occurs in the target data center, and network operation and maintenance personnel can promptly handle the network failure. Therefore, in a possible implementation of the present application, the network quality data denoising method can also include: in multiple detection moments, detecting the actual packet loss rate of the denoised network quality data at each detection moment; if the actual packet loss rate of the denoised network quality data at n consecutive detection moments within the second preset time period is greater than 0, then issuing an alarm prompt about the denoised network quality data.

[0104] Among them, the second preset time period is a detection time window for detecting network quality data. Compared with the extended detection time window in the related art, the second preset time period can be a shorter time window. In the related art, the detection time window for network quality data is usually 3 minutes or 5 minutes, while the second preset time period (detection time window) in the present application can be set to 1s-5s, etc., and the length of the detection time window is greatly shortened. The detection moment also refers to the moment of detecting network quality, which can be understood as the moment of detecting network quality data of the target data center in the embodiment of the present application. For example, n detection moments can be each moment within the second preset time period, which indicates that the network quality data of the target data center is continuously detected. This application does not limit this.

[0105] It can be seen that when detecting the network quality data after noise reduction, the detection time window is 1s-5s, etc. Therefore, when the network quality data is abnormal, an alarm prompt about the network quality data after noise reduction can be issued in time, so that network operation and maintenance personnel and other technical personnel can repair the network as soon as possible and improve the efficiency of fault handling.

[0106] See also Fig.10 , which is a curve diagram of the packet loss rate of network quality data after noise reduction provided by an embodiment of the present application. It represents the packet loss rate of network quality data corresponding to a communication link from city 1 to city 2 in the target data center, such as Fig.10 As shown in the figure, the packet loss rate curve will no longer have burr-like noise data. For details, please refer to Fig.11 A local curve schematic diagram of the packet loss rate of network quality data after noise reduction is shown. Fig.11 It can be understood as Fig.10 The schematic diagram after magnification is as follows: Fig.11 As shown in Figure 2, the packet loss rate is a stable and smooth curve. Fig.10 and Fig.11 It can be seen that the network quality data denoising method provided in the embodiment of the present application effectively eliminates noise data in the network quality data. Even if the requirement for the detection time window length is reduced, false alarms can be reduced, and alarm prompts about network quality data can be issued in time, which is conducive to improving the efficiency of network fault handling.

[0107] Based on the network quality data denoising method provided in the above embodiment, the present application also provides a network quality data denoising device. The network quality data denoising device provided in the embodiment of the present application is specifically introduced below.

[0108] See also Fig.12 , which is a schematic diagram of the structure of a network quality data noise reduction device provided in an embodiment of the present application. Fig.12As shown, the network quality data noise reduction device specifically includes:

[0109] The quality data acquisition module 1210 is used to acquire network quality data corresponding to the detection flow according to the detection flow sent by the source server in the target data center to the server to be tested; the target data center includes multiple business servers, and the target data center provides network services through data interaction between the multiple business servers; one detection flow corresponds to two different business servers as the source server and the server to be tested among the multiple business servers;

[0110] The quality data identification module 1220 is used to identify target network quality data whose actual packet loss index meets the packet loss index screening condition from the acquired network quality data according to the packet loss index screening condition;

[0111] The target detection flow determination module 1230 is used to determine the target detection flow corresponding to the target network quality data according to the corresponding relationship between the network quality data and the detection flow;

[0112] The data noise reduction module 1240 is used to reduce noise on the acquired network quality data based on the determined target detection flow.

[0113] As an implementation method, the packet loss indicator screening condition in the quality data identification module 1220 is that the actual packet loss rate in the network quality data is greater than the first packet loss rate threshold;

[0114] Accordingly, the data noise reduction module 1240 may specifically include:

[0115] A server determination unit is used to determine multiple business servers corresponding to the target detection flow according to the corresponding relationship between the detection flow and the business server; the multiple business servers corresponding to the target detection flow include a source server that sends the target detection flow and a server to be tested that receives the target detection flow;

[0116] The packet loss number calculation unit is used to calculate the number of packet losses of each business server corresponding to the target detection flow; the number of packet losses of each business server is the sum of the number of packet losses when each business server sends the detection flow and the number of packet losses when receiving the detection flow;

[0117] The data noise reduction unit is used to reduce the noise of the acquired network quality data based on the calculated number of packet losses of the service server.

[0118] As an implementation manner, the data noise reduction unit may specifically include:

[0119] A first server determination subunit is used to determine, based on the number of packet losses of the service server corresponding to the target detection flow, a service server corresponding to a number of packet losses greater than a packet loss number threshold as a target server;

[0120] A first detection stream determination subunit is used to determine the detection stream to be denoised corresponding to the target server according to the corresponding relationship between the detection stream and the service server;

[0121] A first quality data determination subunit, configured to determine, from the acquired network quality data, the network quality data to be denoised corresponding to the detection flow to be denoised, according to the correspondence between the network quality data and the detection flow;

[0122] The first data exclusion subunit is used to exclude the network quality data to be de-noised from the acquired network quality data.

[0123] As an implementation manner, the data noise reduction unit may specifically include:

[0124] The second server determination subunit is used to sort the packet loss times of the service servers corresponding to the target detection flow, and determine the service server corresponding to the maximum packet loss time as the target server;

[0125] A second detection flow determination subunit is used to determine the detection flow to be denoised corresponding to the target server according to the corresponding relationship between the detection flow and the service server;

[0126] A second quality data determination subunit is used to determine the network quality data to be denoised corresponding to the detection flow to be denoised from the acquired network quality data according to the correspondence between the network quality data and the detection flow;

[0127] The second data exclusion subunit is used to exclude the network quality data to be de-noised from the acquired network quality data.

[0128] As an implementation method, the network quality data noise reduction device may further include:

[0129] The first data exclusion module is used to exclude the network quality data whose actual packet loss rate is the maximum packet loss rate from the remaining network quality data after exclusion.

[0130] As an implementation method, the network quality data noise reduction device may further include:

[0131] The second data exclusion module is used to exclude network quality data whose actual packet loss rate is greater than a second packet loss rate threshold from the remaining network quality data after exclusion.

[0132] As an implementation mode, the source server or the server to be tested in the quality data acquisition module is determined by the following modules:

[0133] The server screening module is used to screen out business servers that meet the server screening conditions from multiple business servers in the target data center according to the server screening conditions, and obtain the remaining business servers as source servers or servers to be tested; the server screening conditions are screening conditions regarding the hardware settings, software settings or usage performance of multiple business servers in the target data center.

[0134] As an implementation manner, the server screening condition in the server screening module includes at least one of a first server screening condition, a second server screening condition, a third server screening condition, and a fourth server screening condition;

[0135] The first server screening condition is that the usage rate of the central processing unit of the business server is greater than the usage rate threshold, or the data transmission volume of the business server is greater than the transmission volume threshold;

[0136] The second server screening condition is that the identifier of the business server is in the server blacklist;

[0137] The third server screening condition is that the uplink switch of the service server does not contain correct topology information;

[0138] The fourth server screening condition is that a virtual machine is deployed on the business server.

[0139] As an implementation method, the network quality data noise reduction device may further include:

[0140] A data identification module, used to identify the network quality data to be processed whose actual packet loss rate is greater than 0 for m consecutive detection moments within a first preset time period from the acquired network quality data; m is any integer greater than 1;

[0141] A detection flow determination module to be processed, used to determine the detection flow to be processed corresponding to the network quality data to be processed according to the corresponding relationship between the network quality data and the detection flow;

[0142] A server to be tested determining module, used to determine a server to be tested that receives a detection stream to be processed from a plurality of business servers according to a correspondence between the detection stream and the business server;

[0143] The identification adding module is used to add the identification of the server to be tested that receives the detection flow to be processed to the server blacklist.

[0144] As an implementation method, the network quality data noise reduction device may further include:

[0145] A packet loss rate detection module is used to detect the actual packet loss rate of the network quality data after noise reduction at each detection time at multiple detection times;

[0146] The alarm prompt issuing module is used to issue an alarm prompt about the network quality data after noise reduction if the actual packet loss rate of the network quality data after noise reduction is greater than 0 for n consecutive detection moments within a second preset time period.

[0147] An embodiment of the present application provides a network quality data noise reduction device, which may be a server. Fig.13 : This is a schematic diagram of a server structure provided in an embodiment of the present application. The server 900 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 922 (for example, one or more processors) and a memory 932, and one or more storage media 930 (for example, one or more mass storage devices) storing application programs 942 or data 944. Among them, the memory 932 and the storage medium 930 can be short-term storage or permanent storage. The program stored in the storage medium 930 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the server. Furthermore, the central processing unit 922 can be configured to communicate with the storage medium 930 and execute a series of instruction operations in the storage medium 930 on the server 900.

[0148] The server 900 may also include one or more power supplies 926, one or more wired or wireless network interfaces 950, one or more input and output interfaces 958, and / or one or more operating systems 941, such as Windows Server 2000. TM , Mac OS X TM , Unix TM ,Linux TM , FreeBSD TM etc.

[0149] The CPU 922 is used to execute the following steps:

[0150] According to the detection stream sent by the source server in the target data center to the server to be tested, the network quality data corresponding to the detection stream is obtained; the target data center includes multiple business servers, and the target data center provides network services through data interaction between the multiple business servers; one detection stream corresponds to two different business servers as the source server and the server to be tested among the multiple business servers;

[0151] According to the packet loss index screening condition, identifying the target network quality data whose actual packet loss index satisfies the packet loss index screening condition from the acquired network quality data;

[0152] Determine the target detection flow corresponding to the target network quality data according to the corresponding relationship between the network quality data and the detection flow;

[0153] Based on the determined target detection flow, the acquired network quality data is denoised.

[0154] The present application embodiment also provides another network quality data noise reduction device, which can be a terminal device. Fig.14 For the sake of convenience, only the parts related to the embodiments of the present application are shown. For specific technical details not disclosed, please refer to the method part of the embodiments of the present application. Take the terminal device as a mobile phone as an example:

[0155] Fig.14 The block diagram shows a partial structure of a mobile phone provided in an embodiment of the present application. Fig.14 The mobile phone includes: a radio frequency (RF) circuit 1010, a memory 1020, an input unit 1030, a display unit 1040, a sensor 1050, an audio circuit 1060, a wireless fidelity (WiFi) module 1070, a processor 1080, and a power supply 1090. Those skilled in the art can understand that Fig.14 The mobile phone structure shown in the figure does not constitute a limitation on the mobile phone, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0156] Combine the following Fig.14 A detailed introduction to the various components of the mobile phone:

[0157] The RF circuit 1010 can be used for receiving and sending signals during information transmission or calls. In particular, after receiving the downlink information of the base station, it is sent to the processor 1080 for processing; in addition, the designed uplink data is sent to the base station. Generally, the RF circuit 1010 includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (full name: LowNoiseAmplifier, English abbreviation: LNA), a duplexer, etc. In addition, the RF circuit 1010 can also communicate with the network and other devices through wireless communication. The above-mentioned wireless communications may use any communication standard or protocol, including but not limited to Global System of Mobile Communications (Global System of Mobile communication, English abbreviation: GSM), General Packet Radio Service (General Packet Radio Service, GPRS), Code Division Multiple Access (Code Division Multiple Access, English abbreviation: CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (Long Term Evolution, English abbreviation: LTE), e-mail, Short Messaging Service (SMS), etc.

[0158] The memory 1020 can be used to store software programs and modules. The processor 1080 executes various functional applications and data processing of the mobile phone by running the software programs and modules stored in the memory 1020. The memory 1020 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory 1020 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0159] The input unit 1030 can be used to receive input digital or character information, and to generate key signal input related to the user settings and function control of the mobile phone. Specifically, the input unit 1030 may include a touch panel 1031 and other input devices 1032. The touch panel 1031, also known as a touch screen, can collect the user's touch operation on or near it (such as the user's operation on the touch panel 1031 or near the touch panel 1031 using any suitable object or accessory such as a finger, stylus, etc.), and drive the corresponding connection device according to a pre-set program. Optionally, the touch panel 1031 may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into contact coordinates, and then sends it to the processor 1080, and can receive and execute commands sent by the processor 1080. In addition, the touch panel 1031 can be implemented in various types such as resistive, capacitive, infrared, and surface acoustic waves. In addition to the touch panel 1031, the input unit 1030 may also include other input devices 1032. Specifically, the other input devices 1032 may include but are not limited to one or more of a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, etc.

[0160] The display unit 1040 can be used to display information input by the user or information provided to the user and various menus of the mobile phone. The display unit 1040 may include a display panel 1041. Optionally, the display panel 1041 may be configured in the form of a liquid crystal display (full name in English: Liquid Crystal Display, English abbreviation: LCD), an organic light-emitting diode (full name in English: Organic Light-Emitting Diode, English abbreviation: OLED), etc. Further, the touch panel 1031 may cover the display panel 1041. When the touch panel 1031 detects a touch operation on or near it, it is transmitted to the processor 1080 to determine the type of touch event. Subsequently, the processor 1080 provides a corresponding visual output on the display panel 1041 according to the type of touch event. Although in Fig.14 In the embodiment, the touch panel 1031 and the display panel 1041 are used as two independent components to realize the input and output functions of the mobile phone, but in some embodiments, the touch panel 1031 and the display panel 1041 can be integrated to realize the input and output functions of the mobile phone.

[0161] The mobile phone may also include at least one sensor 1050, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display panel 1041 according to the brightness of the ambient light, and the proximity sensor may turn off the display panel 1041 and / or the backlight when the mobile phone is moved to the ear. As a type of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in all directions (generally three axes), and can detect the magnitude and direction of gravity when stationary. It can be used for applications that identify the posture of the mobile phone (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc.; as for other sensors that can be configured in the mobile phone, such as gyroscopes, barometers, hygrometers, thermometers, infrared sensors, etc., they will not be repeated here.

[0162] The audio circuit 1060, the speaker 1061, and the microphone 1062 can provide an audio interface between the user and the mobile phone. The audio circuit 1060 can transmit the received audio data to the speaker 1061 after converting the received audio data into an electrical signal, which is converted into a sound signal for output; on the other hand, the microphone 1062 converts the collected sound signal into an electrical signal, which is received by the audio circuit 1060 and converted into audio data, and then the audio data is output to the processor 1080 for processing, and then sent to another mobile phone through the RF circuit 1010, or the audio data is output to the memory 1020 for further processing.

[0163] WiFi is a short-range wireless transmission technology. The mobile phone can help users send and receive emails, browse web pages and access streaming media through the WiFi module 1070. It provides users with wireless broadband Internet access. Fig.14 A WiFi module 1070 is shown, but it is understandable that it is not an essential component of the mobile phone and can be omitted as needed without changing the essence of the invention.

[0164] The processor 1080 is the control center of the mobile phone. It uses various interfaces and lines to connect various parts of the entire mobile phone. By running or executing software programs and / or modules stored in the memory 1020, and calling data stored in the memory 1020, it executes various functions of the mobile phone and processes data, thereby collecting overall data and information of the mobile phone. Optionally, the processor 1080 may include one or more processing units; preferably, the processor 1080 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 1080.

[0165] The mobile phone also includes a power supply 1090 (such as a battery) for supplying power to various components. Preferably, the power supply can be logically connected to the processor 1080 through a power management system, so that the power management system can manage functions such as charging, discharging, and power consumption.

[0166] Although not shown, the mobile phone may also include a camera, a Bluetooth module, etc., which will not be described in detail here.

[0167] In the embodiment of the present application, the processor 1080 included in the mobile phone also has the following functions:

[0168] According to the detection stream sent by the source server in the target data center to the server to be tested, the network quality data corresponding to the detection stream is obtained; the target data center includes multiple business servers, and the target data center provides network services through data interaction between the multiple business servers; one detection stream corresponds to two different business servers as the source server and the server to be tested among the multiple business servers;

[0169] According to the packet loss index screening condition, identifying the target network quality data whose actual packet loss index satisfies the packet loss index screening condition from the acquired network quality data;

[0170] Determine the target detection flow corresponding to the target network quality data according to the corresponding relationship between the network quality data and the detection flow;

[0171] Based on the determined target detection flow, the acquired network quality data is denoised.

[0172] An embodiment of the present application also provides a computer-readable storage medium for storing a computer program. When the computer program runs on a network quality data denoising device, the network quality data denoising device executes any one of the network quality data denoising methods described in the aforementioned embodiments.

[0173] An embodiment of the present application also provides a computer program product including a computer program. When the computer program product is run on a network quality data denoising device, the computer program product enables the network quality data denoising device to perform any one of the implementation methods of the network quality data denoising method described in the aforementioned embodiments.

[0174] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems and devices described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0175] In the several embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the system is only a logical function division. There may be other division methods in actual implementation, such as multiple systems can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0176] The systems described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0177] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0178] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (full name in English: Read-Only Memory, English abbreviation: ROM), random access memory (full name in English: Random Access Memory, English abbreviation: RAM), disk or optical disk and other media that can store computer programs.

[0179] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A network quality data denoising method, characterized in that: The method comprises: According to the detection stream sent by the source server in the target data center to the server to be tested, network quality data corresponding to the detection stream is obtained; the target data center includes multiple business servers, and the target data center provides network services through data interaction between the multiple business servers; one detection stream corresponds to two different business servers as the source server and the server to be tested among the multiple business servers; According to the packet loss index screening condition, identifying the target network quality data whose actual packet loss index satisfies the packet loss index screening condition from the acquired network quality data; Determining a target detection flow corresponding to the target network quality data according to a correspondence between the network quality data and the detection flow; Based on the determined target detection flow, the acquired network quality data is denoised.

2. The method according to claim 1, characterized in that The packet loss index screening condition is that the actual packet loss rate in the network quality data is greater than a first packet loss rate threshold; The step of reducing noise on the acquired network quality data based on the determined target detection flow specifically includes: According to the correspondence between the detection stream and the service server, a plurality of service servers corresponding to the target detection stream are determined; the plurality of service servers corresponding to the target detection stream include a source server sending the target detection stream and a server to be tested receiving the target detection stream; Calculate the number of packet losses of each service server corresponding to the target detection stream; the number of packet losses of each service server is the sum of the number of packet losses when each service server sends the detection stream and the number of packet losses when receiving the detection stream; Based on the calculated number of packet losses of the service server, the acquired network quality data is denoised.

3. The method according to claim 2, characterized in that The denoising of the acquired network quality data based on the calculated number of packet losses of the service server specifically includes: Based on the number of packet losses of the service server corresponding to the target detection flow, determining the service server corresponding to the number of packet losses greater than the packet loss number threshold as the target server; Determine the detection flow to be denoised corresponding to the target server according to the corresponding relationship between the detection flow and the service server; According to the correspondence between the network quality data and the detection flow, determining the network quality data to be denoised corresponding to the detection flow to be denoised from the acquired network quality data; The network quality data to be de-noised is excluded from the acquired network quality data.

4. The method according to claim 2, characterized in that: The denoising of the acquired network quality data based on the calculated number of packet losses of the service server specifically includes: Sorting the number of packet losses of the service servers corresponding to the target detection flow, and determining the service server corresponding to the maximum number of packet losses as the target server; Determine the detection flow to be denoised corresponding to the target server according to the corresponding relationship between the detection flow and the service server; According to the correspondence between the network quality data and the detection flow, determining the network quality data to be denoised corresponding to the detection flow to be denoised from the acquired network quality data; The network quality data to be de-noised is excluded from the acquired network quality data.

5. The method according to claim 3 or 4, characterized in that: After excluding the network quality data to be de-noised, the method further includes: From the remaining network quality data after exclusion, network quality data whose actual packet loss rate is the maximum packet loss rate is excluded.

6. The method according to claim 3 or 4, characterized in that: After excluding the network quality data to be de-noised, the method further includes: From the remaining network quality data after the exclusion, the network quality data whose actual packet loss rate is greater than the second packet loss rate threshold is excluded.

7. The method according to any one of claims 1 to 4, characterized in that: The source server or the server to be tested is determined in the following way: According to the server screening conditions, the business servers that meet the server screening conditions are screened out from the multiple business servers in the target data center, and the remaining business servers are obtained as source servers or servers to be tested; the server screening conditions are screening conditions regarding the hardware settings, software settings or usage performance of the multiple business servers in the target data center.

8. The method according to claim 7, characterized in that The server screening condition includes at least one of a first server screening condition, a second server screening condition, a third server screening condition and a fourth server screening condition; The first server screening condition is that the usage rate of the central processing unit of the business server is greater than the usage rate threshold, or the data transmission volume of the business server is greater than the transmission volume threshold; The second server screening condition is that the identifier of the business server is in the server blacklist; The third server screening condition is that the uplink switch of the service server does not contain correct topology information; The fourth server screening condition is that a virtual machine is deployed on the business server.

9. The method according to claim 1, characterized in that: The method further comprises: Identify the network quality data to be processed whose actual packet loss rate is greater than 0 for m consecutive detection moments within a first preset time period from the acquired network quality data; wherein m is any integer greater than 1; Determining the detection flow to be processed corresponding to the network quality data to be processed according to the correspondence between the network quality data and the detection flow; According to the correspondence between the detection stream and the service server, determining a server to be tested that receives the detection stream to be processed from the multiple service servers; The identifier of the server to be tested that receives the detection flow to be processed is added to a server blacklist.

10. The method according to claim 1, characterized in that The method further comprises: At multiple detection moments, the network quality data after noise reduction is detected to be at an actual packet loss rate at each detection moment; If the actual packet loss rate of the network quality data after noise reduction is greater than 0 for n consecutive detection moments within a second preset time period, an alarm prompt about the network quality data after noise reduction is issued.

11. A network quality data noise reduction device, characterized in that: The device comprises: A quality data acquisition module is used to acquire network quality data corresponding to a detection flow according to a detection flow sent by a source server in a target data center to a server to be tested; the target data center includes a plurality of service servers, and the target data center provides network services through data interaction between the plurality of service servers; one detection flow corresponds to two different service servers as a source server and a server to be tested among the plurality of service servers; A quality data identification module is used to identify target network quality data whose actual packet loss index satisfies the packet loss index screening condition from the acquired network quality data according to the packet loss index screening condition; A target detection flow determination module, used to determine the target detection flow corresponding to the target network quality data according to the corresponding relationship between the network quality data and the detection flow; A data denoising module is used to denoise the acquired network quality data based on the determined target detection flow.

12. A network quality data noise reduction device, characterized in that: The device comprises a processor and a memory: The memory is used to store a computer program and transmit the computer program to the processor; The processor is configured to execute the steps of the network quality data denoising method according to any one of claims 1 to 10 according to the instructions in the computer program.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store a computer program, and when the computer program is executed by a network quality data denoising device, the steps of the network quality data denoising method according to any one of claims 1 to 10 are implemented.

14. A computer program product, characterized in that The invention comprises a computer program, which, when executed by a network quality data denoising device, implements the steps of the network quality data denoising method according to any one of claims 1 to 10.

Images