CN111884869A - Method, device and system for monitoring network quality - Google Patents
Method, device and system for monitoring network quality Download PDFInfo
- Publication number
- CN111884869A CN111884869A CN202010435314.XA CN202010435314A CN111884869A CN 111884869 A CN111884869 A CN 111884869A CN 202010435314 A CN202010435314 A CN 202010435314A CN 111884869 A CN111884869 A CN 111884869A
- Authority
- CN
- China
- Prior art keywords
- detection
- machine
- network
- edge computing
- monitoring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/10—Active monitoring, e.g. heartbeat, ping or trace-route
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/10—Active monitoring, e.g. heartbeat, ping or trace-route
- H04L43/103—Active monitoring, e.g. heartbeat, ping or trace-route with adaptive polling, i.e. dynamically adapting the polling rate
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/16—Threshold monitoring
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0805—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability
- H04L43/0811—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking connectivity
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0805—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability
- H04L43/0817—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking functioning
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Cardiology (AREA)
- General Health & Medical Sciences (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention relates to a method, a device and a system for monitoring network quality. The method comprises the following steps: determining a detection target of the monitor, wherein the detection target is the monitor in other edge computing nodes; receiving a detection result sent back by the monitoring machine to perform network detection on the detection target; and determining the network quality of the edge computing network according to the detection result. The method achieves the purpose of determining the network quality of each edge computing node.
Description
Technical Field
The present invention relates to the field of network communications technologies, and in particular, to a method, an apparatus, and a system for monitoring network quality.
Background
A Content Delivery Network (CDN) is a novel network content service system, and a basic idea of the CDN is to avoid bottlenecks and links on the internet that may affect data transmission speed and stability, so that content transmission is faster and more stable. By placing node servers at each position of the network, a layer of intelligent virtual network based on the existing Internet is formed, and the CDN system can redirect the request of a user to a service node closest to the user in real time according to network flow, connection of each node, load condition, distance to the user, response time and other comprehensive information. The method aims to enable the user to obtain the required content nearby, solve the network congestion condition and improve the response speed of the user for accessing the website.
In the edge computing scene, services are distributed in a large number of nodes across operators and regions, the network quality of different nodes is uneven, and in particular, in the cross-country and cross-operator scene, once various network fluctuations occur, the stability and reliability of the edge computing services are affected.
The content distribution network can be divided into a central network and an edge network, wherein the edge network is composed of a plurality of edge computing nodes, and each edge computing node comprises a plurality of node machines.
At present, common network quality monitoring methods of data centers and cloud computing products are more concentrated on central network quality monitoring, and for a service sinking type scene of edge computing, quality monitoring is generally performed in the following manner: firstly, selecting a monitor from a central network, wherein the monitor simulates a client to initiate a service request to a node machine in an edge computing node; and secondly, mutually simulating service requests among node machines in part of different edge computing nodes. In such a monitoring mode, network request monitoring data close to client services is lacked, and the quality of the node is difficult to evaluate.
Disclosure of Invention
The invention provides a method, a device and a system for monitoring network quality, which realize the effect of comprehensively evaluating the network quality among a plurality of edge computing network nodes.
In order to achieve the above object, an aspect of the present application provides a method for monitoring network quality, where a scheduling management platform has associated therewith a plurality of edge computing nodes, each edge computing node includes a monitor, and the method is applied to the scheduling management platform, and the method includes:
determining a detection target of the monitor, wherein the detection target is the monitor in other edge computing nodes;
receiving a detection result sent back by the monitoring machine for network detection of the detection target;
and determining the network quality of the edge computing network according to the detection result.
On the basis, the monitoring machine is associated with area information, operator information and label information, wherein the label information comprises a first label and a second label;
the determining of the detection target of the monitoring machine comprises:
selecting a monitor of a first tag as a probe machine, the first tag indicating that the monitor is not being a probe machine;
determining a detected machine related to the detecting machine according to the region information and the operator information, wherein the detected machine is a monitoring machine other than the detecting machine, and the detected machine is a detection target of the detecting machine;
modifying the first label to a second label;
and returning to execute the operation of selecting the monitoring machine with the first label as the detection machine until the label information of all the monitoring machines is the second label.
On the basis, the determining a detected machine associated with the detected machine according to the region information and the operator information comprises:
selecting a first number of monitors as detected machines from the monitors with the same area information and the same operator information as the detected machines;
selecting a second number of monitors as detected machines from the monitors which are different from the area information of the detected machines and have the same operator information;
and selecting a third number of monitors as detected machines from the monitors with the same area information as the detected machines and different operator information.
On this basis, still include:
determining a probing frequency of the probing machine;
taking the product of the detection frequency and the number of the detected machines as the actual monitoring capability value of the detection machine;
and adjusting the detection frequency or the number of the detected machines so that the actual monitoring capability value accords with the theoretical monitoring capability value of the detected machines.
On the basis, the detection result comprises self abnormity;
the determining the network quality of the edge computing network according to the detection result includes:
determining that the detection result is abnormal;
and determining that the network quality of the edge computing node where the monitoring machine reporting the self abnormity is located is low.
On the basis, the detection result comprises connection abnormity;
the determining the network quality of the edge computing network according to the detection result includes:
determining that the detection result is a connection difference;
determining a monitor associated with the connection anomaly;
when one monitoring machine is associated with the connection abnormity exceeding the preset abnormal quantity, the network quality of the edge computing node where the monitoring machine is located is determined to be low.
In order to achieve the above object, another aspect of the present application provides a method for monitoring network quality, where a scheduling management platform has associated therewith a plurality of edge computing nodes, each edge computing node includes a monitor, and the method is applied to the monitor, and the method includes:
initiating a target request to the scheduling management platform;
receiving a detection target sent back by the dispatching management platform in response to the target request, wherein the detection target is the monitoring machine in other edge computing nodes;
initiating network probing to the probing target to obtain probing information;
determining a detection result according to the detection information;
and sending the detection result to a scheduling management platform.
On this basis, the initiating network probing to the probing target to obtain probing information includes:
determining the monitor initiating network detection as a detector;
determining the detection target as a detected machine;
the detecting machine sequentially initiates connection requests to the detected machine;
receiving feedback information sent back by the detected machine according to the connection request;
and determining response time as the detection information, wherein the response time is the time difference between the connection request and the feedback information.
On this basis, the network probe comprises at least one of the following:
initiating network probing based on a tcp _ ping protocol;
initiating network probing based on tcp _ connect;
initiating network probing based on udp _ ping protocol;
network probing is initiated based on icmp _ ping.
On this basis, the determining a detection result according to the detection information includes:
screening abnormal time exceeding preset time from the response time, wherein the abnormal time represents that the network connection between the detecting machine and the detected machine is overtime;
when the quantity of the abnormal time accords with a first condition, generating a detection result of abnormal connection, wherein the abnormal connection represents network abnormality of an edge computing node where a detected machine is located;
and when the quantity of the abnormal time accords with a second condition, generating a detection result of the self abnormality, wherein the self abnormality represents the network abnormality of the edge computing node where the detection machine is located.
In order to achieve the above object, another aspect of the present application provides an apparatus for monitoring network quality, where a scheduling management platform has a plurality of associated edge computing nodes, and each edge computing node includes a monitor, the apparatus includes:
a detection target determining module, configured to determine a detection target of the monitor, where the detection target is the monitor in the other edge computing nodes;
a detection result receiving module, configured to receive a detection result sent back by the monitoring machine to perform network detection on the detection target;
and the network quality determining module is used for determining the network quality of the edge computing network according to the detection result.
In order to achieve the above object, another aspect of the present application provides an apparatus for monitoring network quality, where a scheduling management platform has a plurality of associated edge computing nodes, and each edge computing node includes a monitor, the apparatus includes:
a target request initiating module, configured to initiate a target request to the scheduling management platform;
a detection target receiving module, configured to receive a detection target sent back by the scheduling management platform in response to the target request, where the detection target is the monitor in the other edge computing node;
a network detection initiating module, configured to initiate network detection to the detection target to obtain detection information;
the detection result determining module is used for determining a detection result according to the detection information;
and the detection result sending module is used for sending the detection result to the scheduling management platform.
In order to achieve the above object, another aspect of the present application provides a system for monitoring network quality, including: two or more edge computing nodes and a scheduling management platform;
the edge computing nodes comprise monitors, the monitors are associated with detection targets, and the detection targets are the monitors in other edge computing nodes;
each detection target is in communication connection with the monitor, the detection target sends a detection result back to the monitor, and the monitor determines the network quality of the edge computing node according to the detection result;
each monitoring machine is in communication connection with the scheduling management platform, the monitoring machines upload the detection results to the scheduling management platform, and the scheduling management platform receives the detection results and determines the network quality of the edge computing network according to the detection results. .
On the basis of the above-mentioned technical scheme,
further comprising:
the scheduling management platform determines the number of theoretical detection links according to the number of the monitoring machines;
the dispatching management platform receives the actual detection link number sent back by the monitoring machine;
determining a detection link difference value, wherein the detection link difference value is the difference value between the theoretical detection link quantity and the actual detection link quantity;
calculating link detection integrity, wherein the link detection integrity is the ratio of the link difference in the theoretical detection link quantity;
and when the link detection integrity is lower than an integrity threshold, determining that the system for monitoring the network quality of the edge computing node is abnormal.
Therefore, the technical scheme provided by the application can sink the monitoring machine from the central network to the edge network so as to obtain the network request monitoring data similar to the customer service. The monitoring machine serves as a detector to monitor the network quality of other edge computing nodes, and meanwhile, the monitoring machine (serving as a detected machine) can also be monitored by the monitoring machines of other nodes, namely, the network quality of the edge computing node where the monitoring machine is located can be monitored.
On the basis, the problems that monitoring scheduling needs to be achieved through manual deployment of monitoring tasks are solved, for example, machine conflict and data abnormity are caused by cooperation of multiple persons with different detection tasks, or automatic scheduling switching cannot be achieved when a monitoring machine is abnormal, and the like.
Drawings
FIG. 1 is a schematic diagram of a system for monitoring network quality in an embodiment of the present invention;
fig. 2 is a schematic diagram showing steps of a method for monitoring network quality in embodiment 1 of the present invention;
FIG. 3 is a schematic diagram of a procedure for a monitor to determine other monitors as detection targets in embodiment 1 of the present invention;
fig. 4 is a schematic step diagram of a method for monitoring network quality in embodiment 2 of the present invention;
fig. 5 is a schematic diagram of a step of initiating network probing to a probe target to obtain probe information in embodiment 2 of the present invention;
fig. 6 is a schematic structural diagram of an apparatus for monitoring network quality according to embodiment 3 of the present invention;
fig. 7 is a schematic structural diagram of an apparatus for monitoring network quality according to embodiment 3 of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
The present application provides a method for monitoring network quality, which can be used in a system for monitoring network quality as shown in fig. 1, and in particular, in an edge computing node of the system. Specifically, referring to fig. 1, the overall technical architecture of the system for monitoring network quality is organized and implemented in a C/S (Client/Server) mode. The Server is a scheduling management platform, and the Client is a node machine in the edge computing node. The scheduling management platform is associated with a plurality of edge computing nodes (such as an edge computing node 1, an edge computing node 2, and an edge computing node n), and the edge computing nodes include a plurality of node machines (such as an edge computing node 1 including a node machine 11, a node machine 12, a node machine 13, and a node machine 1 n). The dispatching management platform determines a monitor from node machines included in the edge computing nodes: such as node machine 12 in edge compute node 1, such as node machine 21 in edge compute node 2, such as node machine nn in edge compute node n. The dispatching management platform determines a detection target for the monitor, wherein the detection target is a monitor in other edge computing nodes, and the steps are as follows: the detection targets of the node machine 12 are a node machine 21 and a node machine nn. The monitor initiates a target request to the schedule management platform, and if the monitor (node machine 12) initiates a target request to the schedule management platform, the monitor can accept a detection target sent back by the schedule management platform. The monitoring machine (node machine 12) carries out network detection on the detection targets (node machine 21 and node machine nn) to obtain a detection result. The monitor (node machine 12) sends the detection result to the dispatching management platform, and the dispatching management platform comprehensively evaluates the network quality of each edge computing node according to the detection result.
Example 1
Referring to fig. 1 and fig. 2, in an embodiment of the present application, the method for monitoring network quality may include the following steps:
and S11, determining the detection target of the monitoring machine.
And the dispatching management platform determines a detection target for each monitoring machine.
The scheduling management platform stores the region information, the operator information and the label information of each monitoring machine. The region information is generally provinces where the monitoring machine is located, the operator information is generally telecom, Unicom and others (all operators except telecom and Unicom), the label information comprises a first label and a second label, the first label indicates that the monitoring machine is not matched as a detection machine, and the second label indicates that the monitoring machine is matched as a detection machine. The detecting machine and the detected machine are concepts added for distinguishing the identity of the monitoring machine later, and are also monitoring machines in nature. When the dispatching management platform selects one monitor to determine the detection target, the selected monitor is called a detection machine, and the matched detection target is called a detection machine.
In a specific implementation, as shown in fig. 3, other monitors can be determined as detection targets for the monitors by the following steps:
and S111, selecting the monitoring machine of the first label as a detection machine.
And the dispatching management platform selects the monitor with the label information as the first label as the detector.
And S112, determining the detected machine related to the detecting machine according to the region information and the operator information.
The dispatching management platform selects monitors with a first number (Rule _ i) as detected machines from the monitors with the same area information and the same operator information as the detected machines. That is, in the monitoring machines of the same operator and the same province, Rule _ i are selected as the detection targets, and when the number of the detection targets is less than the number of Rule _ i, all the monitoring machines are selected as the detection targets.
And selecting a second number (Rule _ j) of monitors as detected machines from the monitors which are different from the area information of the detected machines and have the same operator information. That is, in the monitoring machines of the same operator and adjacent provinces, Rule _ j are selected as detection targets, and when the number of the detection targets is less than Rule _ j, all the monitoring machines are selected as the detection targets.
A third number (Rule _ k) of monitors are selected as the detected machines from among the monitors having the same area information as the detecting machine and different operator information. Namely, if the detecting machine is a small operator, Rule _ k monitors of the telecom and Unicom nodes of the province are selected as detection targets, and when the number of the detection targets is less than the number of the Rule _ k monitors, all the monitors are selected as the detection targets.
The number of probed machines is limited such that the sum of the first number, the second number and the third number does not exceed a certain value. Wherein the first amount ranges from 10 to 100; the second amount ranges from 1 to 5; the third amount ranges from 1 to 5; the range of the specific value does not exceed 240. Of course, the first number, the second number, the third number, and the specific value are not limited in this embodiment, but only provide a feasible solution.
It should be noted that the detection between the detecting machine and the detected machine is bidirectional, that is, after one detected machine is matched with one detecting machine, the detected machine is used as the detecting machine, and the originally matched detecting machine is used as the detected machine.
On the basis of the above embodiment, the method further includes: determining the detection frequency of a detector; taking the product of the detection frequency and the number of the detected machines as the actual monitoring capability value of the detection machine; and adjusting the detection frequency or the number of the detected machines so that the actual monitoring capacity value accords with the theoretical monitoring capacity value of the detection machine.
Generally, the actual monitoring capability value (maxPingLoad) of the monitoring machine, which is related to the number of detected machines (Rule _ i + Rule _ j + Rule _ k) and the detection frequency (f), can be understood as: and if the maxPingLoad is too large, the machine resource consumption is too large, and the reliability of the detection data is reduced. Therefore, the probing frequency or the number of probed devices needs to be adjusted so that the actual monitoring capability value (maxPingLoad) matches the theoretical monitoring capability value (generally, the value range is 1000-.
The detection machine can respectively initiate continuous detection of different protocols to a detection target at a certain periodic frequency (600 times/min), so that not only can network quality monitoring data of a time period corresponding to a network problem fed back by a client be accurately matched, but also most network protocol requests of the client are covered.
In view of the fact that the detection amount is large, CPU and IO (Input/Output) of the monitor itself may run high, the node machine triggers the scheduling management platform to execute the processing of the relevant reduction amount or switching the monitor by monitoring the local resources and reporting to the scheduling management platform. When the monitoring machine runs high, or no other monitoring machine is available in the edge computing node, or the edge computing node has a monitoring machine with better selection, the edge computing node can be automatically triggered to switch the monitoring machine, and reliable guarantee and the last safety line are provided for continuous monitoring of the edge computing node.
S113, modifying the first label into a second label.
After the monitoring machine is matched with the detection target, the label information of the monitoring machine serving as the detection machine is modified from the first label to the second label, and the fact that the monitoring machine is matched with the detection target is shown.
And S114, returning to execute the operation of selecting the monitoring machine with the first label as the detection machine until the label information of all the monitoring machines is the second label.
And the dispatching management platform reselects the monitoring machines with the label information as the first labels as the detecting machines, and matches the detection targets until the label information of all the monitoring machines is the second labels.
And S12, receiving a detection result sent back by the monitoring machine to the detection target through network detection.
And S13, determining the network quality of the edge computing network according to the detection result.
Generally, this step can be understood as that the dispatch management platform sends the detection target to the monitor, and the monitor performs network detection on the detection target. And the dispatching management platform receives a detection result sent back by the monitoring machine after the network detection is carried out on the detection target, and comprehensively evaluates the network quality of each edge computing node according to the detection result. The dispatching management platform receives a target request initiated by a monitor, sends a detection target corresponding to the monitor, and the monitor performs network detection on the detection target after receiving the detection target.
Generally, when a probe is abnormal in a network, there may be a situation that the probe cannot report the abnormality. At this time, the network quality abnormality of the node where the current detecting machine is located is indirectly discovered by depending on the detection result of the network detection performed by other monitoring machines. Such as: a prober a, which detects targets b1, b 2.,. bi, i.e. b1, b 2.,. bi, is also detecting the prober a; in the detection results of b1, b2, b, bi, j link anomalies exist in the detection result of the probe a, and when j/i is greater than a threshold value, the network quality of the downlink of the edge computing node where the probe a is located is considered to be low.
Therefore, according to the technical scheme provided by the application, one node machine is automatically selected from the edge computing network to serve as a monitor, the detection target of the monitor is determined, and the detection result sent back after the monitor carries out network detection on the detection target is received, so that the network quality of each edge computing node is comprehensively evaluated. Thus realizing the functions of automatic monitoring, automatic scheduling, automatic identification, automatic alarm and the like.
On this basis, not only can the scheduling management platform comprehensively evaluate the network quality of each edge computing node, but also the monitor of each edge computing network can analyze all detection results in each period. The monitor can finally calculate the real network quality of sending and receiving, and further generate an alarm to prompt operators to sense and process the abnormity prior to the client, so as to avoid influencing the client service.
Example 2
Referring to fig. 1 and 4, in an embodiment of the present application, the method for monitoring network quality may include the following steps:
and S21, initiating a target request to the dispatching management platform.
The monitoring machine sends a target request to the dispatching management platform, wherein the target request comprises the identification number of the monitoring machine so as to distinguish the monitoring machine from other monitoring machines.
The monitoring machine is determined from node machines included in the edge computing nodes through the scheduling management system. In this embodiment, the edge computing node includes a plurality of node machines, and the scheduling management platform designates a node machine as a monitor for the edge computing node after acquiring attribute information of each node machine.
And S22, receiving the detection target sent back by the dispatching management platform in response to the target request.
As a specific means for determining the detection target, reference may be made to the description of step S11 in embodiment 1.
And S23, initiating network detection to the detection target to obtain detection information.
And the monitoring machine carries out network detection on the detection target after receiving the detection target.
The network probing includes at least one of: initiating network detection based on a tcp _ ping protocol, initiating network detection based on a tcp _ connect, and initiating network detection based on a udp _ ping protocol; network probing is initiated based on icmp _ ping.
By improving the coverage degree of the detection protocol, some resource bottlenecks of a monitoring machine in the edge computing network are overcome, and corresponding performance protection is also performed on a system for monitoring the network quality.
As shown in fig. 5, the monitoring machine may initiate network probing to the probe target to obtain probe information by:
s231, determining the monitor initiating the network detection as a detector.
And S232, determining the detection target as a detected machine.
And S233, the detecting machine sequentially sends connection requests to the detected machine.
When the network detection is based on the tcp _ ping protocol, the detector initiates a syn packet to detect the target; when network probing is based on tcp _ connect, initiating a tcp three-way handshake to probe a target; when the network detection is based on udp _ ping protocol, initiating udp message to detect the target; when network probing is based on the icmp _ ping protocol, an icmp message is initiated to probe the target.
The coverage of these four probing protocols allows the probing execution process to overcome some of the resource bottlenecks of the edge servers.
And S234, receiving feedback information sent back by the detected machine according to the connection request.
And S235, determining response time as detection information, wherein the response time is a time difference between the connection request and the feedback information.
The probe machine determines the time difference between sending the connection request and receiving the feedback information as the response time of the detected machine in a certain network detection situation.
And S24, determining a detection result according to the detection information.
The detection result comprises a connection abnormity and a self abnormity. The connection abnormity is network abnormity reported by the monitor and connected with other monitors, and the self abnormity is the self network abnormity presumed and reported by the monitor according to the connection abnormity.
Specifically, the probe machine determines a response time with the detected machine, and from the response time, a response time at which the network connection between the probe machine and the detected machine is over time is screened out as an abnormal time.
If the times of the abnormal time between a detector and a detected machine meet a first condition, the conditions of time delay and packet loss possibly exist between the detector and the detected machine are described. The first condition is that the number of occurrences of the anomaly time does not exceed half of the total number of response times. At this time, a detection result of the connection abnormality is generated. If the times of the abnormal time between a detector and a detected machine meet the second condition, the network abnormality of the edge computing node where the detector is located is described.
Generally, the detecting machine a has n detected machines, each detected machine has an equal probability event of network abnormality, the probability is denoted as p, and p is between (0,1), and is generally much smaller than 1. More than half of the detected machines have network abnormality probability of
And the probability of the network abnormality of the detector is q. Because of the fact that
That is, the probability that more than half of the detected machines are abnormal is far lower than that of the network abnormality of the detecting machines, and once the probability occurs, the uplink network abnormality of the edge computing node where the detecting machines are located can be deduced. At this time, a detection result of the self abnormality is generated.
And the dispatching management platform receives the detection result sent back by the monitoring machine and comprehensively evaluates the network quality of each edge computing node according to the detection result. And each monitor sends back a detection result, and the dispatching management platform counts the detection results sent back by each monitor.
And if the detection result sent back by the monitor is abnormal, determining that the network quality of the edge computing node where the monitor reporting the abnormality is located is low.
If the detection result sent back by the control machine is abnormal connection, the abnormal connection related monitoring machine is ensured; and when one monitoring machine is associated with the connection abnormity exceeding the preset abnormal quantity, determining that the network quality of the edge computing node where the monitoring machine is located is low.
And S25, sending the detection result to a scheduling management platform.
Therefore, according to the technical scheme provided by the application, one node machine is automatically selected from the edge computing network to serve as a monitor, the detection target of the monitor is determined, and the detection result sent back after the monitor carries out network detection on the detection target is received, so that the network quality of each edge computing node is comprehensively evaluated. Thus realizing the functions of automatic monitoring, automatic scheduling, automatic identification, automatic alarm and the like.
Example 3
Fig. 6 is a schematic structural diagram of an apparatus for monitoring network quality according to embodiment 3 of the present invention, where the apparatus includes: a detection target determination module 61, a detection result receiving module 62 and a network quality determination module 63. Wherein:
a detection target determining module 61, configured to determine a detection target of the monitor, where the detection target is the monitor in the other edge computing nodes;
a detection result receiving module 62, configured to receive a detection result sent back by the monitoring machine to perform network detection on the detection target;
and a network quality determining module 63, configured to determine the network quality of the edge computing network according to the detection result.
Therefore, according to the technical scheme provided by the application, one node machine is automatically selected from the edge computing network to serve as a monitor, the detection target of the monitor is determined, and the detection result sent back after the monitor carries out network detection on the detection target is received, so that the network quality of each edge computing node is comprehensively evaluated. Thus realizing the functions of automatic monitoring, automatic scheduling, automatic identification, automatic alarm and the like.
On this basis, the detection target determination module 61 includes:
a prober selection submodule for selecting a monitor of a first tag as a prober, the first tag indicating that the monitor is not to be a prober;
a detected machine selection submodule, configured to determine, according to the region information and the operator information, a detected machine associated with the detected machine, where the detected machine is a monitor other than a detecting machine, and the detected machine is a detection target of the detecting machine;
the label modifying submodule is used for modifying the first label into a second label;
and the execution returning submodule is used for returning and executing the operation of selecting the monitoring machine with the first label as the detection machine until the label information of all the monitoring machines is the second label.
On the basis, the detected machine selection submodule comprises:
a first determining unit, configured to select a first number of monitors as detected machines from the monitors having the same area information and the same operator information as the detected machines;
a second determining unit, configured to select a second number of monitors from monitors different from the probe area information and having the same operator information as the probed equipment;
and the third determining unit is used for selecting a third number of monitors as detected machines from the monitors which have the same area information and different operator information with the detected machines.
On this basis, still include:
the frequency determination module is used for determining the detection frequency of the detector;
the numerical value calculation module is used for taking the product of the detection frequency and the number of the detected machines as the actual monitoring capability value of the detection machine;
and the numerical value adjusting module is used for adjusting the detection frequency or the number of the detected machines so as to enable the actual monitoring capability value to accord with the theoretical monitoring capability value of the detected machines.
On the basis, the detection result comprises self abnormity; the network quality determination module 63 includes:
the first abnormity determining submodule is used for determining that the detection result is self abnormity;
and the first quality reporting submodule is used for determining that the network quality of the edge computing node where the monitoring machine reporting the self abnormity is located is low.
On the basis, the detection result comprises connection abnormity; the network quality determination module 63 includes:
the second abnormity determining submodule is used for determining that the detection result is a connection abnormity;
the association information determining module is used for determining the monitor associated with the abnormal connection;
and the second quality reporting submodule is used for determining that the network quality of the edge computing node where the monitoring machine is located is low when the monitoring machine is associated with the connection abnormity exceeding the preset abnormal quantity.
Example 4
Fig. 7 is a schematic structural diagram of an apparatus for monitoring network quality according to embodiment 3 of the present invention, where the apparatus includes: a target request initiating module 71, a probe target receiving module 72, a network probe initiating module 73, a probe result determining module 74 and a probe result sending module 75. Wherein:
a target request initiating module 71, configured to initiate a target request to the scheduling management platform;
a detection target receiving module 72, configured to receive a detection target sent back by the scheduling management platform in response to the target request, where the detection target is the monitor in the other edge computing nodes;
a network probe initiating module 73, configured to initiate network probe to the probe target to obtain probe information;
a detection result determining module 74, configured to determine a detection result according to the detection information;
and a detection result sending module 75, configured to send the detection result to the scheduling management platform.
Therefore, according to the technical scheme provided by the application, one node machine is automatically selected from the edge computing network to serve as a monitor, the detection target of the monitor is determined, and the detection result sent back after the monitor carries out network detection on the detection target is received, so that the network quality of each edge computing node is comprehensively evaluated. Thus realizing the functions of automatic monitoring, automatic scheduling, automatic identification, automatic alarm and the like.
On this basis, the network probing initiating module 73 includes:
the detection machine determination submodule is used for determining the monitoring machine which initiates network detection as a detection machine;
the detected machine determining submodule is used for determining that the detection target is a detected machine;
the connection request initiating submodule is used for the detecting machine to sequentially initiate connection requests to the detected machine;
the feedback information determining submodule is used for receiving feedback information sent back by the detected machine according to the connection request;
and the detection information determining submodule is used for determining response time as the detection information, wherein the response time is the time difference between the connection request and the feedback information.
On this basis, the network probe comprises at least one of the following:
initiating network probing based on a tcp _ ping protocol;
initiating network probing based on tcp _ connect;
initiating network probing based on udp _ ping protocol;
network probing is initiated based on icmp _ ping.
On this basis, the detection result determination module 74 includes:
the abnormal time determining submodule is used for screening the abnormal time which exceeds the preset time from the response time, and the abnormal time represents that the network connection between the detecting machine and the detected machine is overtime;
the connection abnormity generation submodule is used for generating a detection result of connection abnormity when the quantity of the abnormal time accords with a first condition, and the connection abnormity represents the network abnormity of the edge computing node where the detected machine is located;
and the self-abnormity generation submodule is used for generating a self-abnormity detection result when the quantity of the abnormity time accords with a second condition, wherein the self-abnormity represents the network abnormity of the edge computing node where the detection machine is positioned.
Example 5
Referring to fig. 1, the present application further provides a system for monitoring network quality. A plurality of edge computing nodes are associated with the dispatch management platform, the edge computing nodes including a monitor, the system comprising: two or more edge computing nodes and a scheduling management platform;
the edge computing nodes comprise monitors, the monitors are associated with detection targets, and the detection targets are monitors in other edge computing nodes;
each detection target is in communication connection with the monitor, the detection targets send back detection results to the monitor, and the monitor determines the network quality of the edge computing nodes according to the detection results;
each monitor is in communication connection with the scheduling management platform, the monitors upload detection results to the scheduling management platform, and the scheduling management platform receives the detection results and determines the network quality of the edge computing network according to the detection results.
On the basis, the scheduling management platform determines the number of theoretical detection links according to the number of the monitoring machines; the dispatching management platform receives the actual detection link quantity sent back by the monitoring machine; determining a detection link difference value which is the difference value between the theoretical detection link quantity and the actual detection link quantity; calculating the link detection integrity, wherein the link detection integrity is the ratio of the link difference in the theoretical detection link quantity; and when the link detection integrity is lower than the integrity threshold, determining that the system for monitoring the network quality of the edge computing node is abnormal.
Generally, when generating a probe task, the scheduling management platform counts the theoretical number of probes and probe targets, that is, the theoretical probe link data. When the dispatching management platform receives the detection result sent back by the monitor, the actual number of the detectors and the detection targets, that is, the actual detection link data, is counted from the detection result. According to the theoretical detection link data and the actual detection link data, the calculation of the integrity deviation of the detection link is completed as follows:
the integrity deviation is 100% (theoretical number of probing links-actual number of probing links/theoretical number of probing links).
Through the calculation of the integrity deviation of the detection link, the detection link can be identified from generation, transmission and validation to the final storage and warehousing of the network quality monitoring data, and whether the system for monitoring the network quality is abnormal or not can be identified, so that the purpose of monitoring the closed loop is achieved. For example: such as: and for the A monitoring machine, m detection links exist, the number of the links actually reported by the detection machine is n, and when n/m is smaller than a threshold value, the system judges that the integrity is abnormal, and triggers an alarm to notify operators to pay attention to.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions to enable a computer device (which may be a robot, a personal computer, a server, or a network device) to perform the method for monitoring network quality according to any embodiment of the present invention.
It should be noted that, in the above apparatus for monitoring network quality, each unit and each module included in the apparatus are only divided according to functional logic, but are not limited to the above division, as long as the corresponding function can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
In the description herein, references to the description of the term "in an embodiment," "in another embodiment," "exemplary" or "in a particular embodiment," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although the invention has been described in detail hereinabove by way of general description, specific embodiments and experiments, it will be apparent to those skilled in the art that many modifications and improvements can be made thereto based on the invention. Accordingly, such modifications and improvements are intended to be within the scope of the invention as claimed.
Claims (14)
1. A method for monitoring network quality, wherein a scheduling management platform has associated therewith a plurality of edge computing nodes, an edge computing node comprising a monitor, the method being applied to the scheduling management platform, the method comprising:
determining a detection target of the monitor, wherein the detection target is the monitor in other edge computing nodes;
receiving a detection result sent back by the monitoring machine for network detection of the detection target;
and determining the network quality of the edge computing network according to the detection result.
2. The method of claim 1, wherein the monitor is associated with regional information, operator information, and tag information, the tag information comprising a first tag and a second tag;
the determining of the detection target of the monitoring machine comprises:
selecting a monitor of a first tag as a probe machine, the first tag indicating that the monitor is not being a probe machine;
determining a detected machine related to the detecting machine according to the region information and the operator information, wherein the detected machine is a monitoring machine other than the detecting machine, and the detected machine is a detection target of the detecting machine;
modifying the first label to a second label;
and returning to execute the operation of selecting the monitoring machine with the first label as the detection machine until the label information of all the monitoring machines is the second label.
3. The method of claim 2, wherein said determining a probed equipment associated with the probe machine based on the zone information and the operator information comprises:
selecting a first number of monitors as detected machines from the monitors with the same area information and the same operator information as the detected machines;
selecting a second number of monitors as detected machines from the monitors which are different from the area information of the detected machines and have the same operator information;
and selecting a third number of monitors as detected machines from the monitors with the same area information as the detected machines and different operator information.
4. The method of claim 3, further comprising:
determining a probing frequency of the probing machine;
taking the product of the detection frequency and the number of the detected machines as the actual monitoring capability value of the detection machine;
and adjusting the detection frequency or the number of the detected machines so that the actual monitoring capability value accords with the theoretical monitoring capability value of the detected machines.
5. The method of claim 1, wherein the probe result comprises a self-anomaly;
the determining the network quality of the edge computing network according to the detection result includes:
determining that the detection result is abnormal;
and determining that the network quality of the edge computing node where the monitoring machine reporting the self abnormity is located is low.
6. The method of claim 1, wherein the probe result comprises a connectivity anomaly;
the determining the network quality of the edge computing network according to the detection result includes:
determining that the detection result is a connection difference;
determining a monitor associated with the connection anomaly;
when one monitoring machine is associated with the connection abnormity exceeding the preset abnormal quantity, the network quality of the edge computing node where the monitoring machine is located is determined to be low.
7. A method for monitoring network quality, wherein a scheduling management platform has associated therewith a plurality of edge computing nodes, and wherein an edge computing node comprises a monitor, and wherein the method is applied to the monitor, and wherein the method comprises:
initiating a target request to the scheduling management platform;
receiving a detection target sent back by the dispatching management platform in response to the target request, wherein the detection target is the monitoring machine in other edge computing nodes;
initiating network probing to the probing target to obtain probing information;
determining a detection result according to the detection information;
and sending the detection result to a scheduling management platform.
8. The method of claim 7, wherein the initiating network probing towards the probe target to obtain probing information comprises:
determining the monitor initiating network detection as a detector;
determining the detection target as a detected machine;
the detecting machine sequentially initiates connection requests to the detected machine;
receiving feedback information sent back by the detected machine according to the connection request;
and determining response time as the detection information, wherein the response time is the time difference between the connection request and the feedback information.
9. The method according to claim 7 or 8, wherein the network probing comprises at least one of:
initiating network probing based on a tcp _ ping protocol;
initiating network probing based on tcp _ connect;
initiating network probing based on udp _ ping protocol;
network probing is initiated based on icmp _ ping.
10. The method of claim 8, wherein the determining a probing result according to the probing information comprises:
screening abnormal time exceeding preset time from the response time, wherein the abnormal time represents that the network connection between the detecting machine and the detected machine is overtime;
when the quantity of the abnormal time accords with a first condition, generating a detection result of abnormal connection, wherein the abnormal connection represents network abnormality of an edge computing node where a detected machine is located;
and when the quantity of the abnormal time accords with a second condition, generating a detection result of the self abnormality, wherein the self abnormality represents the network abnormality of the edge computing node where the detection machine is located.
11. An apparatus for monitoring network quality, wherein a dispatch management platform has associated therewith a plurality of edge computing nodes, the edge computing nodes including a monitor, the apparatus comprising:
a detection target determining module, configured to determine a detection target of the monitor, where the detection target is the monitor in the other edge computing nodes;
a detection result receiving module, configured to receive a detection result sent back by the monitoring machine to perform network detection on the detection target;
and the network quality determining module is used for determining the network quality of the edge computing network according to the detection result.
12. An apparatus for monitoring network quality, wherein a dispatch management platform has associated therewith a plurality of edge computing nodes, the edge computing nodes including a monitor, the apparatus comprising:
a target request initiating module, configured to initiate a target request to the scheduling management platform;
a detection target receiving module, configured to receive a detection target sent back by the scheduling management platform in response to the target request, where the detection target is the monitor in the other edge computing node;
a network detection initiating module, configured to initiate network detection to the detection target to obtain detection information;
the detection result determining module is used for determining a detection result according to the detection information;
and the detection result sending module is used for sending the detection result to the scheduling management platform.
13. A system for monitoring network quality, comprising: two or more edge computing nodes and a scheduling management platform;
the edge computing nodes comprise monitors, the monitors are associated with detection targets, and the detection targets are the monitors in other edge computing nodes;
each detection target is in communication connection with the monitor, the detection target sends a detection result back to the monitor, and the monitor determines the network quality of the edge computing node according to the detection result;
each monitoring machine is in communication connection with the scheduling management platform, the monitoring machines upload the detection results to the scheduling management platform, and the scheduling management platform receives the detection results and determines the network quality of the edge computing network according to the detection results.
14. The system of claim 13, further comprising:
the scheduling management platform determines the number of theoretical detection links according to the number of the monitoring machines;
the dispatching management platform receives the actual detection link number sent back by the monitoring machine;
determining a detection link difference value, wherein the detection link difference value is the difference value between the theoretical detection link quantity and the actual detection link quantity;
calculating link detection integrity, wherein the link detection integrity is the ratio of the link difference in the theoretical detection link quantity;
and when the link detection integrity is lower than an integrity threshold, determining that the system for monitoring the network quality of the edge computing node is abnormal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010435314.XA CN111884869A (en) | 2020-05-21 | 2020-05-21 | Method, device and system for monitoring network quality |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010435314.XA CN111884869A (en) | 2020-05-21 | 2020-05-21 | Method, device and system for monitoring network quality |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111884869A true CN111884869A (en) | 2020-11-03 |
Family
ID=73154139
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010435314.XA Pending CN111884869A (en) | 2020-05-21 | 2020-05-21 | Method, device and system for monitoring network quality |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111884869A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114285770A (en) * | 2021-12-25 | 2022-04-05 | 即刻雾联科技(北京)有限公司 | Full-link-detection-based pcdn node evaluation method and terminal |
| CN114584485A (en) * | 2022-01-30 | 2022-06-03 | 阿里巴巴（中国）有限公司 | Method, device, equipment and computer readable storage medium for detecting edge network quality |
| CN114615337A (en) * | 2022-01-27 | 2022-06-10 | 网宿科技股份有限公司 | Equipment scheduling method, system, server and storage medium |
| CN114679443A (en) * | 2022-03-03 | 2022-06-28 | 北京百度网讯科技有限公司 | Method and device for downloading video data, electronic equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102136969A (en) * | 2011-02-17 | 2011-07-27 | 北京蓝汛通信技术有限责任公司 | Method, device and system for probing link quality |
| CN102347864A (en) * | 2011-11-02 | 2012-02-08 | 网宿科技股份有限公司 | System and method for monitoring service quality of content distribution networks |
| CN102624584A (en) * | 2012-03-01 | 2012-08-01 | 中兴通讯股份有限公司 | Link detection method and link detection device |
| CN108768792A (en) * | 2018-07-06 | 2018-11-06 | 网宿科技股份有限公司 | Network detecting systems |
| CN109787827A (en) * | 2019-01-18 | 2019-05-21 | 网宿科技股份有限公司 | A kind of method and device of CDN network monitoring |
| CN109889569A (en) * | 2019-01-03 | 2019-06-14 | 网宿科技股份有限公司 | CDN service dispatching method and system |
-
2020
- 2020-05-21 CN CN202010435314.XA patent/CN111884869A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102136969A (en) * | 2011-02-17 | 2011-07-27 | 北京蓝汛通信技术有限责任公司 | Method, device and system for probing link quality |
| CN102347864A (en) * | 2011-11-02 | 2012-02-08 | 网宿科技股份有限公司 | System and method for monitoring service quality of content distribution networks |
| CN102624584A (en) * | 2012-03-01 | 2012-08-01 | 中兴通讯股份有限公司 | Link detection method and link detection device |
| CN108768792A (en) * | 2018-07-06 | 2018-11-06 | 网宿科技股份有限公司 | Network detecting systems |
| CN109889569A (en) * | 2019-01-03 | 2019-06-14 | 网宿科技股份有限公司 | CDN service dispatching method and system |
| CN109787827A (en) * | 2019-01-18 | 2019-05-21 | 网宿科技股份有限公司 | A kind of method and device of CDN network monitoring |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114285770A (en) * | 2021-12-25 | 2022-04-05 | 即刻雾联科技(北京)有限公司 | Full-link-detection-based pcdn node evaluation method and terminal |
| CN114615337A (en) * | 2022-01-27 | 2022-06-10 | 网宿科技股份有限公司 | Equipment scheduling method, system, server and storage medium |
| CN114615337B (en) * | 2022-01-27 | 2024-04-12 | 网宿科技股份有限公司 | Equipment scheduling method, system, server and storage medium |
| CN114584485A (en) * | 2022-01-30 | 2022-06-03 | 阿里巴巴（中国）有限公司 | Method, device, equipment and computer readable storage medium for detecting edge network quality |
| CN114584485B (en) * | 2022-01-30 | 2023-10-31 | 阿里巴巴（中国）有限公司 | Method, apparatus, device and computer readable storage medium for detecting edge network quality |
| CN114679443A (en) * | 2022-03-03 | 2022-06-28 | 北京百度网讯科技有限公司 | Method and device for downloading video data, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111884869A (en) | Method, device and system for monitoring network quality | |
| US7574502B2 (en) | Early warning of potential service level agreement violations | |
| EP2815538B1 (en) | Method and apparatus for rapid disaster recovery preparation in a cloud network | |
| US7412448B2 (en) | Performance degradation root cause prediction in a distributed computing system | |
| US20130024568A1 (en) | Management communication | |
| US20160226714A1 (en) | Method and device for monitoring network link and storage medium therefor | |
| US20200327045A1 (en) | Test System and Test Method | |
| JP6220625B2 (en) | Delay monitoring system and delay monitoring method | |
| Cao et al. | {ENVI}: elastic resource flexing for network function virtualization | |
| WO2017000536A1 (en) | Bfd method and apparatus | |
| Xu et al. | Lightweight and adaptive service api performance monitoring in highly dynamic cloud environment | |
| US10225371B2 (en) | Method and network components and self-organizing network | |
| CN109617758B (en) | Node network quality calculation method and device, server and computer storage medium | |
| CN110266741B (en) | Method and device for automatically scheduling client service in content distribution network | |
| CN117242754A (en) | Methods, systems, and computer readable media for autonomic network test case generation | |
| CN113422696A (en) | Monitoring data updating method, system, equipment and readable storage medium | |
| EP3252995A1 (en) | Method for detecting network failures | |
| CN117478582A (en) | Oil field production equipment state monitoring and optimizing method based on intelligent link management system | |
| CN111786846B (en) | Method, device, equipment and storage medium for determining monitoring machine | |
| CN113055291B (en) | Data packet sending method, router and data packet transmission system | |
| US20100153543A1 (en) | Method and System for Intelligent Management of Performance Measurements In Communication Networks | |
| KR101490316B1 (en) | Fault Detection System For Network Device And Fault Detection Method Using The Same | |
| Hossfeld et al. | Comparing the scalability of communication networks and systems | |
| JP2014137635A (en) | Response time monitoring program, method and device | |
| Abderrahim et al. | Dependability integration in cloud-hosted telecommunication services |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201103 |