Network measurement method and device, electronic equipment, storage medium and computer program product

Abstract

The invention relates to a network measurement method and device, electronic equipment, a storage medium and a computer program product. The network measurement method comprises the following steps: receiving watching quality degradation information sent by a client; obtaining a network degradation time interval of the target network based on the watching quality degradation information; using the congestion control model to obtain network state information of the target network with the round-trip time of the data packet as the granularity, and monitoring the target network to obtain data packet transmission information of the target network with the round-trip time of the data packet as the granularity; determining the network state information and the data packet transmission information in the network degradation time interval as the original information of the weak network part of the target network; a plurality of key features with the first preset duration as the granularity are extracted from the original information of the weak network part, the key features are determined as network measurement data of the weak network part, and the key features refer to features representing the real-time quality of the weak network part.

Description

Technical Field

[0001] The present disclosure relates to the field of communications, and in particular to a network measurement method and device, an electronic device, a storage medium, and a computer program product. Background Art

[0002] In recent years, with the development of large-scale media services (such as short videos, long videos, and live broadcasts), smooth and clear video viewing experience has become increasingly important. When users watch videos, freezes, long first screens, and low clarity will cause a decline in the user's viewing experience, affecting the user's viewing interest and thus reducing the user's viewing time. The degradation of the user's network is the core factor that causes the degradation of user experience. Therefore, studying countermeasures under weak networks is a core topic that the industry is increasingly paying attention to.

[0003] Nowadays, network infrastructure has been greatly developed, and network speed has gradually increased, but the network also has large fluctuations. In order to combat network fluctuations, the industry uses adaptive multi-rate algorithms, congestion control algorithms, preloading algorithms, etc. to combat network fluctuations. However, network fluctuations are diverse. Increased latency, reduced bandwidth, and increased packet loss in the underlying network will lead to a decrease in the available bandwidth of the user application layer. If there is a quantitative expression of weak network changes and a rich online real weak network database can be established locally, this will greatly improve the research and development efficiency of weak network algorithm strategies and greatly reduce the cost of trial and error.

[0004] Currently, the industry and academia mostly use network measurement and network loss meters to record and replay weak networks. Existing network measurement methods mainly include bandwidth measurement, delay measurement, maximum transmission unit (MTU) measurement, and reachability measurement. However, more intensive network measurements will introduce huge additional bandwidth costs and may have a destructive impact on the user experience; more sparse network measurements cannot reflect the changes in the actual network. Summary of the invention

[0005] The present disclosure provides a network measurement method and device, an electronic device, a storage medium and a computer program product, so as to at least solve the problem that the bandwidth cost consumed by the related technology is high or the change of the real network cannot be reflected.

[0006] According to a first aspect of an embodiment of the present disclosure, a network measurement method is provided, comprising: receiving viewing quality degradation information sent by a client; obtaining a network degradation time interval of a target network based on the viewing quality degradation information; obtaining network status information of the target network with a round-trip time of a data packet as a granularity by using a congestion control model, and obtaining data packet transmission information of the target network with a round-trip time of a data packet as a granularity by monitoring the target network; determining the network status information and data packet transmission information within the network degradation time interval as original information of a weak network portion of the target network; extracting a plurality of key features with a first predetermined duration as a granularity from the original information of the weak network portion and determining the plurality of key features as network measurement data of the weak network portion, wherein the key features refer to features that characterize the real-time quality of the weak network portion.

[0007] Optionally, based on the viewing quality degradation information, obtaining the network degradation time interval of the target network includes: based on the viewing quality degradation information, obtaining the degradation start time when the viewing quality degradation occurs on the client and the start download time of the video part where the viewing quality degradation occurs on the client; based on the first time, the second time, the degradation start time and the start download time of the client, determining the degradation start time when the viewing quality degradation occurs on the server and the start download time of the video part on the server, wherein the first time is the time when the transport layer connected to the client receives the viewing quality degradation information, and the second time is the time when the server receives the viewing quality degradation information; with the start download time and the degradation start time of the server as the boundary, determining the network degradation time interval of the target network.

[0008] Optionally, the network status information and data packet transmission information within the network degradation time interval are determined as the original information of the weak network part of the target network, including: expanding the network degradation time interval based on a second predetermined time length; and determining the network status information and data packet transmission information within the expanded network degradation time interval as the original information of the weak network part of the target network.

[0009] Optionally, in the case where the key features include physical propagation delay, bottleneck routing bandwidth, random packet loss rate and maximum queue depth of bottleneck routing, multiple key features with a first predetermined time length as the granularity are extracted from the original information of the weak network part, including: taking the physical propagation delay and physical bandwidth in the original information as the physical propagation delay and bottleneck routing bandwidth of the weak network part, respectively; determining the correlation between the observed round-trip delay and the observed packet loss rate in the original information, and determining the random packet loss rate and the maximum queue depth of the bottleneck routing of the weak network part based on the correlation; sampling the physical propagation delay, bottleneck routing bandwidth, random packet loss rate and maximum queue depth of bottleneck routing respectively with the first predetermined time length as the granularity to obtain multiple key features of each key feature.

[0010] Optionally, based on the correlation, the random packet loss rate of the weak network part and the maximum queue depth of the bottleneck route are determined, including: in response to the correlation being less than a first threshold, taking the average of the observed packet loss rates of all observed round-trip delay intervals as the random packet loss rate of the weak network part; in response to the correlation being greater than a second threshold, taking the average observed packet loss rate corresponding to a predetermined observed round-trip delay interval as the random packet loss rate of the weak network part, wherein the predetermined observed round-trip delay interval is an interval with the smallest average observed round-trip delay; in response to the correlation being greater than a third threshold, determining the maximum queue depth of the bottleneck route based on the maximum observed round-trip delay, the bottleneck route bandwidth and the physical propagation delay.

[0011] Optionally, determining the correlation between the observed round-trip delay and the observed packet loss rate in the original information includes: partitioning the observed round-trip delay in the original information to obtain multiple observed round-trip delay intervals; for each observed round-trip delay interval, obtaining an average value of the observed packet loss rate and an average value of the observed round-trip delay; and determining the correlation between the observed round-trip delay and the observed packet loss rate based on the Spearman correlation coefficient of the average value of the observed packet loss rate and the average value of the observed round-trip delay in each observed round-trip delay interval.

[0012] According to a second aspect of an embodiment of the present disclosure, a network measurement device is provided, comprising: a receiving unit, configured to receive viewing quality degradation information sent by a client; a first acquisition unit, configured to acquire a network degradation time interval of a target network based on the viewing quality degradation information; a second acquisition unit, configured to acquire network status information of the target network with a round-trip time of a data packet as a granularity by using a congestion control model, and acquire data packet transmission information of the target network with a round-trip time of a data packet as a granularity by monitoring the target network; a determination unit, configured to determine the network status information and the data packet transmission information within the network degradation time interval as original information of a weak network portion of the target network; an extraction unit, configured to extract a plurality of key features with a first predetermined duration as a granularity from the original information of the weak network portion and determine the plurality of key features as network measurement data of the weak network portion, wherein the key features refer to features characterizing the real-time quality of the weak network portion.

[0013] Optionally, the first acquisition unit is further configured to obtain, based on the viewing quality degradation information, the degradation start time of the viewing quality degradation occurring on the client and the start download time of the video part where the viewing quality degradation occurs on the client; determine the degradation start time of the viewing quality degradation occurring on the server and the start download time of the video part on the server based on the first time, the second time, the degradation start time and the start download time of the client, wherein the first time is the time when the transport layer connected to the client receives the viewing quality degradation information, and the second time is the time when the server receives the viewing quality degradation information; determine the network degradation time interval of the target network with the start download time and the degradation start time of the server as boundaries.

[0014] Optionally, the determination unit is further configured to expand the network degradation time interval based on a second predetermined time length; and determine the network status information and data packet transmission information within the expanded network degradation time interval as original information of the weak network part of the target network.

[0015] Optionally, when the key features include physical propagation delay, bottleneck routing bandwidth, random packet loss rate and maximum queue depth of bottleneck routing, the extraction unit is further configured to use the physical propagation delay and physical bandwidth in the original information as the physical propagation delay and bottleneck routing bandwidth of the weak network part, respectively; determine the correlation between the observed round-trip delay and the observed packet loss rate in the original information, and based on the correlation, determine the random packet loss rate and the maximum queue depth of the bottleneck routing of the weak network part; sample the physical propagation delay, bottleneck routing bandwidth, random packet loss rate and maximum queue depth of bottleneck routing respectively with the first predetermined time length as the granularity to obtain multiple key features of each key feature.

[0016] Optionally, the extraction unit is further configured to, in response to the correlation being less than a first threshold, use the average value of the observed packet loss rates of all observed round-trip delay intervals as the random packet loss rate of the weak network part; in response to the correlation being greater than a second threshold, use the average observed packet loss rate corresponding to a predetermined observed round-trip delay interval as the random packet loss rate of the weak network part, wherein the predetermined observed round-trip delay interval is an interval with the smallest average observed round-trip delay; in response to the correlation being greater than a third threshold, determine the maximum queue depth of the bottleneck route based on the maximum observed round-trip delay, the bottleneck route bandwidth and the physical propagation delay.

[0017] Optionally, the extraction unit is further configured to partition the observed round-trip delay in the original information to obtain multiple observed round-trip delay intervals; for each observed round-trip delay interval, obtain the observed packet loss rate average value and the observed round-trip delay average value; based on the Spearman correlation coefficient of the observed packet loss rate average value and the observed round-trip delay average value of each observed round-trip delay interval, determine the correlation between the observed round-trip delay and the observed packet loss rate.

[0018] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement the network measurement method of the present disclosure.

[0019] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided. When instructions in the computer-readable storage medium are executed by at least one processor, the at least one processor is prompted to execute the network measurement method of the present disclosure as described above.

[0020] According to a fifth aspect of an embodiment of the present disclosure, a computer program product is provided, including computer instructions, which implement the network measurement method of the present disclosure when executed by a processor.

[0021] The technical solution provided by the embodiments of the present disclosure brings at least the following beneficial effects:

[0022] According to the network measurement method and device, electronic device, storage medium and computer program product disclosed in the present invention, when the viewing quality deteriorates, the client sends viewing quality degradation information to the server, that is, the client disclosed in the present invention can feedback the viewing quality degradation information in real time, and after receiving the information, the server can know the network degradation time interval based on the viewing quality degradation information, and then combine the monitored data packet transmission information and the network status information output by the congestion control model to obtain the network measurement data of the weak network part of the real network, which is used to accelerate the iteration of the local algorithm and improve the quality of online user use. Therefore, the present invention solves the problem that the bandwidth cost consumed by the related technology is high or cannot reflect the changes in the real network.

[0023] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute improper limitations on the present disclosure.

[0025] Figure 1 is a schematic diagram of an implementation scenario of a network measurement method according to an exemplary embodiment of the present disclosure;

[0026] Figure 2 is a schematic diagram of a weak network measurement and full-link optimization framework according to an exemplary embodiment of the present disclosure;

[0027] Figure 3 is a flow chart of a network measurement method according to an exemplary embodiment of the present disclosure;

[0028] Figure 4 is a block diagram of a network measurement device according to an exemplary embodiment of the present disclosure;

[0029] Figure 5 is a block diagram of an electronic device 500 according to an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0030] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings.

[0031] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The implementation methods described in the following examples do not represent all implementation methods consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the attached claims.

[0032] It should be noted that the phrase "at least one of the items" in the present disclosure includes three types of parallel situations: "any one of the items", "a combination of any number of the items", and "all of the items". For example, "including at least one of A and B" includes the following three parallel situations: (1) including A; (2) including B; (3) including A and B. Another example is "executing at least one of step 1 and step 2" which means the following three parallel situations: (1) executing step 1; (2) executing step 2; (3) executing step 1 and step 2.

[0033] Currently, network measurement mainly includes bandwidth measurement, delay measurement, maximum transmission unit (MTU) measurement and reachability measurement. The following briefly introduces two network measurement methods:

[0034] 1. The PathLoad-based network bandwidth measurement method is a typical method of active network measurement, that is, actively sending a packet sequence at a fixed rate, recording the arrival time and reception time of the packet, and analyzing the arrival trend of the packet interval. If the packet arrival interval has an increasing trend, it means that the sending rate is greater than the link bandwidth, and the receiving rate at this time is calculated as the link bandwidth; if the packet arrival interval has no increasing trend, the binary method is used to increase the sending rate until the packet arrival interval is observed to have an increasing trend.

[0035] However, the network bandwidth measurement method based on PathLoad has the following three disadvantages: 1) PathLoad needs to actively send packets for network measurement, which will introduce huge bandwidth costs when actually deployed in online streaming applications; 2) PathLoad uses a binary method to detect bandwidth. The binary boundary selection greatly affects the detection speed, and the bandwidth measurement efficiency is low; 3) PathLoad can only reflect the network bandwidth level when it is detecting, and cannot reflect network characteristics in other dimensions, such as delay, random packet loss, queue depth and other factors. It also cannot reflect the network characteristics when the quality of streaming applications deteriorates.

[0036] In particular, in recent years, many variant algorithms that use packet pairs and packet sequences to measure bandwidth have emerged. However, they will introduce more bandwidth costs and cannot simultaneously observe multi-dimensional network characteristics. Here, only PathLoad is listed as a typical algorithm.

[0037] 2. The network measurement method based on ICMP packets is a typical method for measuring reachability, packet loss and delay. That is, the ping command is used on the client to periodically send ICMP packets, measure the average network round-trip delay and jitter level, and observe the packet loss rate and network reachability over a period of time.

[0038] However, the network measurement method based on ICMP packets has the following two disadvantages: 1) Delay measurement using IMCP packets is generally sparse, with a typical interval of 1000ms. If it is more dense, it cannot reflect the true level of physical delay; 2) Using ICMP packets to measure cannot reflect the bandwidth level of the network, and it cannot reflect the network characteristics when the quality of streaming media applications deteriorates.

[0039] In order to solve the problems of high bandwidth cost, slow measurement speed and incomplete reflection of network characteristics of related network measurement methods, the present invention obtains network measurement data of the weak network part that conforms to the real user network through real-time feedback of application software (such as streaming media application on the client) combined with congestion control model and information required for real-time statistics of network monitoring, so as to accelerate the iteration of local algorithm and improve the quality of online user use.

[0040] Hereinafter, a network measurement method and apparatus, an electronic device, a storage medium, and a computer program product according to exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.

[0041] Figure 1 is a schematic diagram of an implementation scenario of a network measurement method according to an exemplary embodiment of the present disclosure, such as Figure 1The implementation scenario includes a server 100, a user terminal 110 and a user terminal 120, wherein the user terminals are not limited to two, and include but are not limited to mobile phones, personal computers and other devices. The user terminals can install streaming media application software. The server can be a single server, or a server cluster composed of several servers, or a cloud computing platform or a virtualization center.

[0042] After the user terminals 110 and 120 access the target network, the user can watch the video through the streaming media application software on the user terminal 110 or 120. If the viewing quality deteriorates during the video viewing process, the viewing quality degradation information is sent to the server 100. After receiving the viewing quality degradation information sent by the user terminals 110 and 120, the server 100 obtains the network degradation time interval of the target network based on the viewing quality degradation information, and obtains the network status information of the target network with the round-trip time of the data packet as the granularity by using the congestion control model, and obtains the data packet transmission information of the target network with the round-trip time of the data packet as the granularity by monitoring the target network; then, the network status information and data packet transmission information within the network degradation time interval are determined as the original information of the weak network part of the target network; then, multiple key features with the first predetermined time length as the granularity are extracted from the original information of the weak network part and the multiple key features are determined as the network measurement data of the weak network part, wherein the key feature refers to the feature that characterizes the real-time quality of the weak network part.

[0043] After obtaining the network measurement data of the weak network part, the network measurement data can be used to establish a large-scale network database of weak networks of real online users, and perform accelerated simulation playback locally, greatly improving the research and development efficiency of weak network confrontation. Figure 2 The weak network measurement and full-link optimization framework in the present disclosure is demonstrated. When a user uses streaming media application software to watch a video and the viewing quality deteriorates, the network measurement unit will record the network status in real time, that is, the network measurement data, and then perform weak network playback and evaluation offline to form a weak network database to provide support for offline strategy optimization.

[0044] Figure 3 is a flow chart of a network measurement method according to an exemplary embodiment of the present disclosure, such as Figure 3 As shown, the network measurement method includes the following steps:

[0045] In step S301, viewing quality degradation information sent by a client is received.

[0046] As an example, the client runs application software (such as streaming media application software) using the target network. When it is detected that the viewing quality of the application software is degraded, the viewing quality degradation information can be sent to the server.

[0047] As an example, the viewing quality degradation is a video freeze. At this time, the viewing quality degradation information (QoS info ) can include but is not limited to: video freeze duration (block dur ), video viewing freeze times (block count ), watch the video freeze moment (block ts ), application layer video bitrate, and the start time of downloading the video part where the freeze occurs. start ).

[0048] It should be noted that the degradation of viewing quality is not limited to video freezes, but may also be the degradation of viewing quality of various types of streaming media, such as decreased video clarity, length of time spent on the first screen of a video, etc. This disclosure only lists video freezes for exemplary explanation.

[0049] In step S302, based on the viewing quality degradation information, the network degradation time interval of the target network is obtained.

[0050] As an example, the corresponding time information of the server can be obtained based on the time information corresponding to the freeze in the viewing quality degradation information sent by the client, and then the network degradation time interval can be determined based on the corresponding time information of the server.

[0051] According to an exemplary embodiment of the present disclosure, the network degradation time interval of the target network can be determined through the following specific steps: based on the viewing quality degradation information, the degradation start time of the viewing quality degradation occurring on the client and the start download time of the video part where the viewing quality degradation occurs on the client are obtained; based on the first time, the second time, the degradation start time and the start download time of the client, the degradation start time of the viewing quality degradation occurring on the server and the start download time of the video part on the server are determined, wherein the first time is the time when the transport layer connected to the client receives the viewing quality degradation information, and the second time is the time when the server receives the viewing quality degradation information; with the start download time and the degradation start time of the server as the boundary, the network degradation time interval of the target network is determined.

[0052] Through this embodiment, the time of the server and the client is aligned based on the time when the transport layer connected to the client receives the viewing quality degradation information and the time when the server receives the viewing quality degradation information, so that based on the aligned time, the server can determine a relatively accurate network degradation time interval.

[0053] As an example, the present disclosure connects the transport layer and the client application software. When the viewing quality of the application software degrades, the application software transparently transmits the viewing quality degradation information to the connected transport layer, and the transport layer transmits the viewing quality degradation information to the server through the streaming media information frame. The server receives the viewing quality degradation information, starts the network measurement process, and records the measured information locally on the server. During the non-network peak period, the network measurement recorded information is transmitted to the weak network database for storage.

[0054] As an example, if the viewing quality of the application software deteriorates, the viewing quality degradation information QoS info Passed to the transport layer, when the client transport layer receives the viewing quality degradation information QoS info , record this time as t c , the transport layer will c and QoS info The server receives the QoS information of the degraded viewing quality. info , record this time as t s , then, analyze QoS info The client's timestamp block ts and download start , which can then be converted into the corresponding timestamp on the server according to the following formula:

[0055] block tsserver =block ts -t c +t s -srtt

[0056] =download start -t c +t s -srtt

[0057] Among them, srtt is the smoothed rtt calculated by the server, which is not discussed in detail in this disclosure.

[0058] Get the timestamp block corresponding to the server ts_server and download start_server After that, we can get the network degradation time interval, that is, [download start_server ,block ts_server ].

[0059] In step S303, the congestion control model is used to obtain network status information of the target network with the round-trip time of the data packet as the granularity, and the data packet transmission information of the target network with the round-trip time of the data packet as the granularity is obtained by monitoring the target network.

[0060] As an example, a confirmation message of a data packet transmitted through the target network can be obtained, and the confirmation message can be input into the congestion control model, thereby obtaining the network status information of the target network corresponding to the round-trip time of the data packet, such as physical propagation delay and physical bandwidth, which is not limited in the present disclosure.

[0061] As an example, during the operation of the target network, by monitoring the target network, corresponding data packet transmission information can be observed, such as observing the round-trip delay and observing the packet loss rate, which is not limited in the present disclosure.

[0062] As an example, the server can record network status information and data packet transmission information for up to T seconds (such as 600 seconds) with round-trip delay (RTT) as the granularity, where the network status information may include: the physical bandwidth bw output by the congestion control model i , the physical propagation delay d output by the congestion control model i ; Data packet transmission information may include: observed packet loss rate l i , the current observed round trip delay rtt i , where i represents the i-th RTT.

[0063] In step S304, the network state information and data packet transmission information within the network degradation time interval are determined as original information of the weak network part of the target network.

[0064] As an example, the network status information and data packet transmission information within the above-mentioned network degradation time interval can be obtained, and these network status information and data packet transmission information can be stored in an independent local file on the server. The corresponding file name can use the md5 code of the information, and the file can be used as the original file of the weak network part, which is equivalent to the original information of the above-mentioned weak network part.

[0065] According to an exemplary embodiment of the present disclosure, the network status information and data packet transmission information within the network degradation time interval can be determined as the original information of the weak network part of the target network through the following specific steps: the network degradation time interval is expanded based on the second predetermined time length; the network status information and data packet transmission information within the expanded network degradation time interval are determined as the original information of the weak network part of the target network. Through this embodiment, the network degradation time interval is expanded to avoid not fully covering the network degradation time.

[0066] As an example, when the network degradation time interval [download start_server ,block ts_server ], the network degradation time interval can be expanded according to the second predetermined time length to obtain the expanded network degradation time interval t i :

[0067] t i ∈[download start_server -α×srtt,block ts_server +α×srtt]

[0068] Among them, α×srtt is the second predetermined time length, which can be between 30 seconds and 40 seconds, and the present disclosure does not limit this.

[0069] Then, we can get the i The network status information and data packet transmission information in the data packet are stored in a separate local file on the server. The corresponding file name can use the md5 code of the information, and the file is used as the original file of the weak network part, which is equivalent to the original information of the weak network part mentioned above.

[0070] In step S305, multiple key features with a first predetermined duration as granularity are extracted from the original information of the weak network part and the multiple key features are determined as network measurement data of the weak network part, wherein the key features refer to features characterizing the real-time quality of the weak network part.

[0071] As an example, this embodiment can extract network measurement data that can be used for local weak network restoration from the above original information with a first predetermined time length as the granularity to improve the iteration efficiency of the local algorithm strategy. The above first predetermined time length can be set as needed, such as 1 millisecond.

[0072] As an example, in order to obtain relatively accurate network measurement data of the weak network part, the above key features include but are not limited to: random packet loss rate l r , physical propagation delay d p , bottleneck route bandwidth b, bottleneck route maximum queue depth d q This embodiment uses 1 millisecond as the granularity to extract the bottleneck route bandwidth b and the bottleneck route maximum queue depth d q , and the random packet loss rate l r , physical propagation delay d p The network measurement data within an RTT can still be recorded in the following form with RTT as the granularity, as the corresponding weak network trace for subsequent local weak network simulation and restoration:

[0073] <l r ,d p , <d q1 ,b1>, <d q2 ,b2>… <d qn ,b n >>

[0074] It should be noted that, assuming there are multiple network degradation time intervals, if the interval between two network degradation time intervals does not exceed a predetermined time (such as 30 seconds), the network measurement data of the two network degradation time intervals can be merged together as a weak network trajectory.

[0075] According to an exemplary embodiment of the present disclosure, when the key features include physical propagation delay, bottleneck routing bandwidth, random packet loss rate and maximum queue depth of bottleneck routing, extracting multiple key features with a first predetermined time length as the granularity from the original information of the weak network part can include: using the physical propagation delay and physical bandwidth in the original information as the physical propagation delay and bottleneck routing bandwidth of the weak network part, respectively; determining the correlation between the observed round-trip delay and the observed packet loss rate in the original information, and determining the random packet loss rate and the maximum queue depth of the bottleneck routing of the weak network part based on the correlation; sampling the physical propagation delay, bottleneck routing bandwidth, random packet loss rate and maximum queue depth of bottleneck routing respectively with the first predetermined time length as the granularity to obtain multiple key features of each key feature.

[0076] Through this embodiment, the above four key features are adopted to better describe the actual quality of the weak network part, so that the network measurement data of the weak network part can be used to improve the iteration efficiency of the local algorithm strategy in the future.

[0077] As an example, this embodiment directly uses the physical propagation delay d output in real time by the congestion control model as the physical propagation delay d p , and directly use the physical bandwidth bw output in real time by the congestion control model i As the bottleneck route bandwidth b. The random packet loss rate and the maximum queue depth of the bottleneck route can be determined based on the correlation between the observed round trip delay and the observed packet loss rate in the original information. After obtaining these four key features, sampling can be performed according to the first predetermined time length as needed to obtain multiple key features that meet the needs.

[0078] As an example, the above-mentioned congestion control model can adopt BBR (Bottleneck Bandwidth and Round-trip Propagation Time) V3, and other congestion control models can also be adopted. For example, the bandwidth detection speed can be improved by improving the bandwidth detection mechanism of the congestion control model. When the network bandwidth changes significantly and the application layer data is limited, the bandwidth can still be measured more accurately, and more accurate network measurement raw data will be obtained. This is not limited in the present disclosure.

[0079] It should be noted that the key features are not limited to physical propagation delay, bottleneck routing bandwidth, random packet loss rate and bottleneck routing maximum queue depth, but may also be other features, which are not limited in the present disclosure.

[0080] According to an exemplary embodiment of the present disclosure, based on the correlation, determining the random packet loss rate of the weak network part and the maximum queue depth of the bottleneck route may include: in response to the correlation being less than the first threshold, taking the average value of the observed packet loss rates of all observed round-trip delay intervals as the random packet loss rate of the weak network part; in response to the correlation being greater than the second threshold, taking the average observed packet loss rate corresponding to the predetermined observed round-trip delay interval as the random packet loss rate of the weak network part, wherein the predetermined observed round-trip delay interval is the interval with the smallest average observed round-trip delay; in response to the correlation being greater than the third threshold, determining the maximum queue depth of the bottleneck route based on the maximum observed round-trip delay, the bottleneck route bandwidth, and the physical propagation delay. Through this embodiment, based on the similarity information, a relatively accurate random packet loss rate and the maximum queue depth of the bottleneck route can be determined.

[0081] As an example, the first threshold, the second threshold and the third threshold may be set as required, which is not limited in the present disclosure.

[0082] As an example, taking the first threshold as 0.25 and the second threshold as 0.75, if the similarity is less than 0.25, the average value of all observed packet loss rates is calculated as the random packet loss rate l of the weak network part. r If the correlation is greater than 0.75, the interval with the smallest average value of the observed round-trip delay is determined, and the average value of the observed packet loss rate in this interval is used as the random packet loss rate of the weak network part l r .

[0083] As an example, taking the third threshold value as 0.75, if the similarity is greater than 0.75, the maximum round trip delay rtt is determined. max , the maximum queue depth d of the bottleneck route can be obtained as follows q :

[0084] d q =(rtt max -d p )×b

[0085] According to an exemplary embodiment of the present disclosure, determining the correlation between the observed round-trip delay and the observed packet loss rate in the original information may include: partitioning the observed round-trip delay in the original information to obtain multiple observed round-trip delay intervals; for each observed round-trip delay interval, obtaining the observed packet loss rate average value and the observed round-trip delay average value; based on the Spearman correlation coefficient of the observed packet loss rate average value and the observed round-trip delay average value in each observed round-trip delay interval, determining the correlation between the observed round-trip delay and the observed packet loss rate. Through this embodiment, by partitioning and calculating the observed packet loss rate average value and the observed delay average value in each interval, the similarity can reduce the computational complexity and ensure accuracy.

[0086] As an example, taking the network status information including physical propagation delay and physical bandwidth, and the data packet transmission information including observed round-trip delay and observed packet loss rate as an example, for each piece of original information, the observed packet loss rate l in the original information can be i and observed round trip delay rtt i Forming an observation pair <rtt i ,l i >, statistics rtt i The maximum value of rtt max and minimum value rtt min , the observed round-trip delay is evenly divided into n intervals (k∈[1,n]) according to the following rules.

[0087] [rtt min +(k-1)*(rtt max -rtt min ) / n,rtt min +k*(rtt max -rtt min ) / n]

[0088] For each observed round-trip delay interval, the average value of the internal observed packet loss rate and the average value of the observed round-trip delay can be calculated to obtain <l k ,rtt k >, where k represents the kth observed round trip delay interval. Based on the average value of the observed packet loss rate and the average value of the observed round trip delay in each interval, the Spearman correlation coefficient is calculated to determine the correlation between the observed round trip delay and the observed packet loss rate.

[0089] In summary, the present invention obtains the original information of the weak network part in the network through real-time feedback from the application software (i.e., the client) in combination with the congestion control model and the information required for real-time statistics of network monitoring, and effectively converts the original data into weak network traces while removing noise data, so as to obtain a large database of weak networks that conform to the real user network, so that multi-dimensional network characteristics can be measured simultaneously without introducing additional bandwidth, thereby making large-scale industrial implementation possible.

[0090] The present invention obtains about 5 million weak network traces by performing network measurements in live streaming services in some occasions of application software. By using the weak network traces, targeted optimization is performed locally, reducing the 100-second freeze duration of live streaming in these occasions by about 10%.

[0091] Figure 4 is a block diagram of a network measurement device according to an exemplary embodiment of the present disclosure. Figure 4 The device includes a receiving unit 40, a first acquiring unit 42, a second acquiring unit 44, a determining unit 46 and an extracting unit 48.

[0092] A receiving unit 40 is configured to receive viewing quality degradation information sent by a client; a first obtaining unit 42 is configured to obtain a network degradation time interval of a target network based on the viewing quality degradation information; a second obtaining unit 44 is configured to obtain network status information of the target network with a round-trip time of a data packet as a granularity by using a congestion control model, and obtain data packet transmission information of the target network with a round-trip time of a data packet as a granularity by monitoring the target network; a determining unit 46 is configured to determine the network status information and the data packet transmission information within the network degradation time interval as original information of a weak network portion of the target network; an extracting unit 48 is configured to extract a plurality of key features with a first predetermined duration as a granularity from the original information of the weak network portion and determine the plurality of key features as network measurement data of the weak network portion, wherein the key features refer to features characterizing the real-time quality of the weak network portion.

[0093] According to an exemplary embodiment of the present disclosure, the first acquisition unit 42 is further configured to obtain, based on the viewing quality degradation information, the degradation start time of the viewing quality degradation occurring on the client and the start download time of the video part where the viewing quality degradation occurs on the client; determine the degradation start time of the viewing quality degradation occurring on the server and the start download time of the video part on the server based on the first time, the second time, the degradation start time and the start download time of the client, wherein the first time is the time when the transport layer connected to the client receives the viewing quality degradation information, and the second time is the time when the server receives the viewing quality degradation information; determine the network degradation time interval of the target network with the start download time and the degradation start time of the server as the boundary.

[0094] According to an exemplary embodiment of the present disclosure, the determination unit 46 is further configured to expand the network degradation time interval based on a second predetermined time length; and determine the network status information and data packet transmission information within the expanded network degradation time interval as the original information of the weak network part of the target network.

[0095] According to an exemplary embodiment of the present disclosure, when the key features include physical propagation delay, bottleneck routing bandwidth, random packet loss rate and maximum queue depth of bottleneck routing, the extraction unit 48 is further configured to use the physical propagation delay and physical bandwidth in the original information as the physical propagation delay and bottleneck routing bandwidth of the weak network part, respectively; determine the correlation between the observed round-trip delay and the observed packet loss rate in the original information, and based on the correlation, determine the random packet loss rate and the maximum queue depth of the bottleneck routing of the weak network part; sample the physical propagation delay, bottleneck routing bandwidth, random packet loss rate and maximum queue depth of bottleneck routing respectively with a first predetermined time length as the granularity to obtain multiple key features of each key feature.

[0096] According to an exemplary embodiment of the present disclosure, the extraction unit 48 is further configured to, in response to the correlation being less than a first threshold, use the average value of the observed packet loss rates of all observed round-trip delay intervals as the random packet loss rate of the weak network part; in response to the correlation being greater than a second threshold, use the average observed packet loss rate corresponding to a predetermined observed round-trip delay interval as the random packet loss rate of the weak network part, wherein the predetermined observed round-trip delay interval is an interval with the smallest average observed round-trip delay; in response to the correlation being greater than a third threshold, determine the maximum queue depth of the bottleneck route based on the maximum observed round-trip delay, the bottleneck route bandwidth and the physical propagation delay.

[0097] According to an exemplary embodiment of the present disclosure, the extraction unit 48 is further configured to partition the observed round-trip delay in the original information to obtain multiple observed round-trip delay intervals; for each observed round-trip delay interval, obtain the observed packet loss rate average value and the observed round-trip delay average value; based on the Spearman correlation coefficient of the observed packet loss rate average value and the observed round-trip delay average value of each observed round-trip delay interval, determine the correlation between the observed round-trip delay and the observed packet loss rate.

[0098] According to an embodiment of the present disclosure, an electronic device may be provided. Figure 5 It is a block diagram of an electronic device 500 shown according to an exemplary embodiment of the present disclosure, the electronic device includes at least one memory 501 and at least one processor 502, the at least one memory stores a set of computer executable instructions, when the computer executable instruction set is executed by the at least one processor, the network measurement method according to the embodiment of the present disclosure is executed.

[0099] As an example, the electronic device 500 may be a PC, a tablet device, a personal digital assistant, a smart phone, or other device capable of executing the above instruction set. Here, the electronic device 1000 is not necessarily a single electronic device, but may also be any device or circuit capable of executing the above instruction (or instruction set) individually or in combination. The electronic device 500 may also be part of an integrated control system or system manager, or may be configured as a portable electronic device interconnected with a local or remote (e.g., via wireless transmission) interface.

[0100] In the electronic device 500, the processor 502 may include a central processing unit (CPU), a graphics processing unit (GPU), a programmable logic device, a dedicated processor system, a microcontroller or a microprocessor. As an example and not limitation, the processor 502 may also include an analog processor, a digital processor, a microprocessor, a multi-core processor, a processor array, a network processor, etc.

[0101] The processor 502 can execute instructions or codes stored in the memory, wherein the memory 501 can also store data. Instructions and data can also be sent and received over the network via the network interface device, wherein the network interface device can use any known transmission protocol.

[0102] The memory 501 may be integrated with the processor 502, for example, by placing RAM or flash memory within an integrated circuit microprocessor or the like. In addition, the memory 501 may include a separate device, such as an external disk drive, a storage array, or any other storage device that can be used by a database system. The memory 501 and the processor 502 may be operationally coupled, or may communicate with each other, such as through an I / O port, a network connection, etc., so that the processor 502 can read files stored in the memory 501.

[0103] In addition, the electronic device 500 may also include a video display (such as a liquid crystal display) and a user interaction interface (such as a keyboard, a mouse, a touch input device, etc.) All components of the electronic device may be connected to each other via a bus and / or a network.

[0104] According to an embodiment of the present disclosure, a computer-readable storage medium may also be provided, wherein when the instructions in the computer-readable storage medium are executed by at least one processor, the at least one processor is prompted to execute the network measurement method of the embodiment of the present disclosure. Examples of computer-readable storage media here include: read-only memory (ROM), random access programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), flash memory, non-volatile memory, CD-ROM, CD-R, CD+R, CD-RW, CD+RW, DVD-ROM, DVD-R, DVD+R, DVD-RW, DVD+RW, DVD-RAM, BD-ROM, BD-R, BD-R LTH, BD-RE, Blu-ray or optical disk storage, hard disk drive (HDD), solid state drive (SSD), card storage (such as, multimedia card, secure digital (SD) card or extreme digital (XD) card), magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk and any other device, any other device is configured to store computer programs and any associated data, data files and data structures in a non-transitory manner and provide the computer programs and any associated data, data files and data structures to a processor or computer so that the processor or computer can execute the computer program. The computer program in the above-mentioned computer-readable storage medium can be run in an environment deployed in a computer device such as a client, a host, an agent device, a server, etc. In addition, in one example, the computer program and any associated data, data files and data structures are distributed on a networked computer system, so that the computer program and any associated data, data files and data structures are stored, accessed and executed in a distributed manner by one or more processors or computers.

[0105] According to an embodiment of the present disclosure, a computer program product is provided, including computer instructions, and when the computer instructions are executed by a processor, the network measurement method of the embodiment of the present disclosure is implemented.

[0106] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the appended claims.

[0107] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A network measurement method, characterized in that: include: Receive viewing quality degradation information sent by the client; Based on the viewing quality degradation information, obtaining a network degradation time interval of the target network; Using a congestion control model, obtaining network status information of the target network with a round-trip time of a data packet as a granularity, and obtaining data packet transmission information of the target network with a round-trip time of a data packet as a granularity by monitoring the target network; Determine the network status information and the data packet transmission information within the network degradation time interval as original information of the weak network part of the target network; A plurality of key features with a first predetermined duration as a granularity are extracted from the original information of the weak network portion and the plurality of key features are determined as network measurement data of the weak network portion, wherein the key features refer to features characterizing the real-time quality of the weak network portion.

2. The network measurement method according to claim 1, characterized in that: The acquiring, based on the viewing quality degradation information, a network degradation time interval of the target network includes: Based on the viewing quality degradation information, acquiring a degradation start time when viewing quality degradation occurs on the client and a start time when the video portion where viewing quality degradation occurs occurs on the client; Determine the degradation start time of the viewing quality degradation at the server and the start download time of the video part at the server based on the first time, the second time, the degradation start time of the client and the start download time, wherein the first time is the time when the transport layer connected to the client receives the viewing quality degradation information, and the second time is the time when the server receives the viewing quality degradation information; The network degradation time interval of the target network is determined by taking the download start time of the server and the degradation start time as the boundary.

3. The network measurement method according to claim 1, characterized in that: Determining the network state information and the data packet transmission information within the network degradation time interval as original information of the weak network part of the target network includes: Expanding the network degradation time interval based on a second predetermined time period; The network status information and the data packet transmission information within the expanded network degradation time interval are determined as original information of the weak network part of the target network.

4. The network measurement method according to claim 1, characterized in that: In the case where the key characteristics include physical propagation delay, bottleneck routing bandwidth, random packet loss rate and bottleneck routing maximum queue depth, The extracting of multiple key features with a first predetermined duration as a granularity from the original information of the weak network portion includes: The physical propagation delay and the physical bandwidth in the original information are used as the physical propagation delay and the bottleneck routing bandwidth of the weak network part respectively; Determine the correlation between the observed round-trip delay and the observed packet loss rate in the original information, and based on the correlation, determine the random packet loss rate of the weak network part and the maximum queue depth of the bottleneck route; The physical propagation delay, the bottleneck routing bandwidth, the random packet loss rate and the maximum queue depth of the bottleneck routing are sampled respectively with the first predetermined time length as the granularity to obtain multiple key features of each key feature.

5. The network measurement method according to claim 4, characterized in that: The determining, based on the correlation, the random packet loss rate of the weak network portion and the maximum queue depth of the bottleneck route includes: In response to the correlation being less than a first threshold, taking an average of observed packet loss rates of all observed round-trip delay intervals as the random packet loss rate of the weak network portion; in response to the correlation being greater than a second threshold, taking an average observed packet loss rate corresponding to a predetermined observed round-trip delay interval as the random packet loss rate of the weak network portion, wherein the predetermined observed round-trip delay interval is an interval with a minimum average observed round-trip delay; In response to the correlation being greater than a third threshold, a bottleneck route maximum queue depth is determined based on a maximum observed round trip delay, the bottleneck route bandwidth, and the physical propagation delay.

6. The network measurement method according to claim 4, characterized in that: Determining the correlation between the observed round-trip delay and the observed packet loss rate in the original information includes: Partitioning the observed round-trip delay in the original information to obtain a plurality of observed round-trip delay intervals; For each observed round-trip delay interval, obtain the average observed packet loss rate and the average observed round-trip delay; Based on the Spearman correlation coefficient between the average value of the observed packet loss rate in each observed round-trip delay interval and the average value of the observed round-trip delay, the correlation between the observed round-trip delay and the observed packet loss rate is determined.

7. A network measurement device, characterized in that: include: A receiving unit, configured to receive viewing quality degradation information sent by a client; A first acquisition unit is configured to acquire a network degradation time interval of a target network based on the viewing quality degradation information; A second acquisition unit is configured to acquire network status information of the target network with a round-trip time of a data packet as a granularity by using a congestion control model, and acquire data packet transmission information of the target network with a round-trip time of a data packet as a granularity by monitoring the target network; a determining unit configured to determine the network state information and the data packet transmission information within the network degradation time interval as original information of the weak network part of the target network; The extraction unit is configured to extract multiple key features with a first predetermined time length as a granularity from the original information of the weak network part and determine the multiple key features as network measurement data of the weak network part, wherein the key features refer to features that characterize the real-time quality of the weak network part.

8. An electronic device, characterized in that: include: at least one processor; at least one memory storing computer executable instructions, When the computer executable instructions are executed by the at least one processor, the at least one processor is prompted to perform the network measurement method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: When the instructions in the computer-readable storage medium are executed by at least one processor, the at least one processor is prompted to perform the network measurement method according to any one of claims 1 to 6.

10. A computer program product comprising computer instructions, characterized in that When the computer instructions are executed by a processor, the network measurement method according to any one of claims 1 to 6 is implemented.

Images