CN114760205B - Self-calibration optimization method of distributed network - Google Patents

Abstract

The invention discloses a self-calibration optimization method of a distributed network. The invention simplifies the communication model of the distributed network, defines the roles of distributed network nodes, selects the calibration parameters and statistical targets of the distributed network, and automatically detects distributed The quality of the network data transmission path implements an optimization strategy that prioritizes transmission reliability/bandwidth utilization. The invention can automatically calibrate the network communication parameters of the nodes in the distributed network, so as to achieve the purpose of network optimization and improvement of communication quality. Operation and maintenance personnel can carry out self-calibration operations without manually adjusting the network parameters of each node. The present invention provides two optimization strategies of reliability and bandwidth utilization, and has stronger scene adaptability.

Description

A self-calibration optimization method for distributed networks

technical field

The invention belongs to the technical field of computer communication, and in particular relates to a self-calibration optimization method of a distributed network.

Background technique

Distributed networks are often composed of different network media, such as wired LAN, public network, wireless WIFI, VPN virtual network and other mixed networks. This networking mode leads to inconsistency in the network bandwidth of each data path in the distributed network, and inconsistency in link transmission quality. At present, when distributed network nodes communicate with each other, there are the following problems: 1. Using the default communication parameters, it cannot adapt to the bandwidth and link quality changes on different data links, resulting in packet loss; 2. There is no effective tool to calibrate the network , leading to difficulties in network optimization; 3. There is no network optimization strategy for operation and maintenance personnel to choose, and it is difficult to adapt to the scene.

The technical solution most similar to the present invention comes from the patent CN201911126014.7, "Method for Optimizing Network Status in Remote Terminals of DTU Power Distribution Automation", which proposes to add network status data forwarding and testing functions to terminal nodes in the power network , to count the periodic data, random data and burst data passing through the nodes, use the statistical algorithm to complete the network analysis, and finally obtain the purpose of the appropriate communication data interval. However, this patent does not take the communication packet loss rate and single packet data size into consideration as factors affecting the quality of network communication, but only considers the data interval; the network status evaluation of this patent needs to modify the normal communication messages of the nodes, which requires a certain amount of nodes Resources; the statistical analysis algorithm used in this patent is relatively complex, and it is difficult to implement, which is not conducive to batch implementation.

Contents of the invention

The purpose of the present invention is to provide a self-calibration optimization method of a distributed network aiming at the deficiencies of the prior art.

The purpose of the present invention is achieved through the following technical solutions: a self-calibration optimization method for a distributed network, the method comprising the following steps:

S1. Simplify the communication model of the distributed network into a star network topology model. For each distributed node, it acts as a central node, and other distributed nodes communicate with the central node point-to-point;

S2, define the role of distributed network nodes, and deploy distributed network calibration software:

Define the central node as the calibration client, and deploy the calibration client software on it; define the distributed nodes other than the central node as the calibration server, and deploy the calibration server software on it;

The calibration client is used to send a ping packet to the calibration server, specify the size of the packet, specify the packet sending cycle, calculate the calibration server return packet delay, and record the number of packet return failures;

The calibration server is used to respond to the ping packet sent by the calibration client, and receive the network parameter configuration command sent by the calibration client;

S3, selecting a distributed network calibration parameter and a statistical target, the calibration parameter includes a ping packet size and a packet sending cycle, and the statistical target is the time required to complete 1GB data transmission and the number of lost packets;

Select several files for the size of the ping packet used for calibration; record the network delay from the central node to the distributed node as T1, take several file sending cycles, and each file sending cycle is a different multiple of T1; different file ping packet size and sending cycle Form a set of calibration parameters;

S4, distributed network self-calibration, including:

The calibration client selects a set of calibration parameters, and sends ping packets continuously according to the calibration parameters. If the response is lost in the middle, the number of lost packets will be increased by 1. When the cumulative number of bytes sent by the ping packets reaches 1GB, the test of this group is completed, and the test of this group is obtained. The time T2 to complete the 1GB data transmission and the total number of packet loss N in this group; select another set of calibration parameters to repeat the test process, and obtain the T2 and N values corresponding to each set of calibration parameters; the group with the smallest T2 value is the optimal network bandwidth utilization group; the group with the smallest N value is the optimal group for network reliability;

The user selects a set of network parameters from the optimal network bandwidth utilization group and the optimal network reliability group through the calibration client and sends them to the calibration server to complete the self-calibration optimization of the distributed network.

Further, in step S3, the size of the ping packet used for calibration is 32, 64, 128, 256, 512, and 1024 bytes, a total of 6 files, and the sending period is 1.5xT1, 2.0xT1, 2.5xT1, a total of 3 files, and the 18 sets of calibration parameters, and the corresponding time T2 for completing 1GB data transmission and the number of packet loss N constitute an evaluation table.

Further, in step S4, the calibration process is initiated by the calibration client, and a set of calibration parameters is selected: a ping packet of 32 bytes, and a packet sending cycle of 1.5xT1. The calibration client first presses a 32-byte ping packet, with 1.5 times the network delay T1 is the time interval from the receiving time of the last response to the sending of the ping packet. The time difference between the sending of the ping packet and the receiving of the reply is recorded as t1, t2, t3...tm. 1. When the cumulative number of bytes sent by the ping packets reaches 1GB, the test of this group is completed. The time for this group to complete 1GB data transmission is T2=t1+t2+t3...+tm, and the total number of lost packets in this group is N; Then select the next set of calibration parameters: ping packet 64 bytes, packet sending cycle 1.5xT1 to test until all sets of calibration parameters are tested and a complete evaluation form is obtained.

Further, in step S4, the group with the smallest T2 value means that the 1GB data transmission is completed at the fastest speed, which is the group with the optimal network bandwidth utilization; the group with the smallest N value means that the 1GB data transmission is completed with the least number of packet loss , which is the optimal group for network reliability, thus setting two optional tuning strategies.

The beneficial effects of the present invention are: the method of the present invention simplifies the communication model of the distributed network, defines the roles of distributed network nodes, selects the calibration parameters and statistical targets of the distributed network, and automatically detects the quality of the data transmission path of the distributed network. Implement an optimization strategy that prioritizes transmission reliability/bandwidth utilization. The invention can automatically calibrate the network communication parameters of the nodes in the distributed network, so as to achieve the purpose of network optimization and improvement of communication quality. Operation and maintenance personnel can carry out self-calibration operations without manually adjusting the network parameters of each node. The present invention provides two optimization strategies of reliability and bandwidth utilization, and has stronger scene adaptability.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required in the embodiments. Obviously, the accompanying drawings in the following description are only some of the present invention. Embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without paying creative labor.

FIG. 1 is a flowchart of a self-calibration optimization method for a distributed network provided by an exemplary embodiment of the present invention;

FIG. 2 is a schematic diagram of a typical topology of a distributed network provided by an exemplary embodiment of the present invention;

FIG. 3 is a schematic diagram of a simplified model of a distributed network provided by an exemplary embodiment of the present invention;

Fig. 4 is a flowchart of a single-group self-calibration provided by an exemplary embodiment of the present invention.

Detailed ways

In order to better understand the technical solutions of the present application, the embodiments of the present application will be described in detail below in conjunction with the accompanying drawings.

It should be clear that the described embodiments are only some of the embodiments of the present application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of this application.

Terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "a", "said" and "the" used in the embodiments of this application and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise.

A self-calibration optimization method of a distributed network provided by the present invention, as shown in Figure 1, the specific implementation steps are as follows:

(1) Simplify the communication model of the distributed network

As shown in Figure 2, the distributed network is composed of different network media, such as wired LAN, public network, wireless WIFI, VPN virtual network and other mixed networks. From the perspective of each distributed node, other nodes communicate with it point-to-point in this network, which is a characteristic of star network topology. Therefore, as shown in FIG. 3 , the present invention simplifies the distributed network topology into a star network topology model, that is, for each distributed node, it acts as a central node, and other distributed nodes perform point-to-point communication with the central node.

(2) Define the role of distributed network nodes and deploy distributed network calibration software

After analysis, the main factors affecting the quality of communication in the distributed network are the size of a single data packet and the time slot interval between packets.

Therefore, before performing distributed network self-calibration, define the role of distributed network nodes, define the central node as the calibration client, and deploy the calibration client software on it; define the distributed nodes other than the central node as the calibration server , deploy the calibration server software on it.

The calibration client can send a ping packet to the calibration server, specify the size of the packet, specify the packet sending cycle, calculate the delay of the calibration server’s return packet, and record the number of packet return failures.

The calibration server can respond to the ping packet sent by the calibration client, and can receive the network parameter configuration command sent by the calibration client.

(3) Select distributed network calibration parameters and statistical targets

Specifically, the distributed network calibration parameters include a ping packet size and a packet sending cycle.

Combined with common communication scenarios in distributed networks, the size of the ping packet used for calibration is 32, 64, 128, 256, 512, and 1024 bytes, a total of 6 files.

The determination of the packet sending cycle is related to the actual network conditions. The network delay from the central node to the distributed nodes is T1, and the packet sending cycle is 1.5xT1, 2.0xT1, and 2.5xT1.

Combined with the analysis of the requirements of distributed networks for communication quality, it is mainly divided into two dimensions: communication reliability and communication effective bandwidth. Therefore, the time T2_X and the number of packet loss N_X required to complete the 1GB data transmission are selected as the statistical targets. In summary, the following evaluation table is obtained:

Table 1 Distributed network calibration parameters and statistical target groups

(4) Distributed network self-calibration

As shown in Figure 4, the calibration process is initiated by the calibration client. According to the calibration parameters in Table 1, the calibration client first presses a 32-byte ping packet, with 1.5 times the network delay T1 as the time interval from the last response receiving time to the sending of this ping packet, and the ping packet is sent until the response is received ( The time difference between the time of receiving the response information from the calibration server) is recorded as t1, t2, t3...tm. If the response is lost in the middle, the packet loss count N will be increased by 1. When the cumulative number of bytes sent by the ping packet reaches 1GB, This group of tests is complete. The time for this group to complete the 1GB data transmission is T2=t1+t2+t3...+tm, and the total number of lost packets in this group is N. Repeat the test process shown in Figure 4 in turn according to the parameters in Table 1, then you can get the T2 and N values of each group of parameters.

Among them, the group with the smallest T2 value means that the 1GB data transmission is completed at the fastest speed, which is the group with the optimal network bandwidth utilization; the group with the smallest N value means that the 1GB data transmission is completed with the least number of packet loss, which is the network reliability. optimal group.

Users can select a set of network parameters from the optimal network bandwidth utilization group and the optimal network reliability group through the central node (calibration client) and send them to the distributed nodes (calibration server), which completes the tuning The final network parameters are issued to complete the distributed network optimization. The present invention provides two optimization strategies of network reliability/bandwidth utilization for users to choose, and the scene adaptability is stronger.

It should be noted that the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus comprising a set of elements includes not only those elements, but also includes none other elements specifically listed, or also include elements inherent in the process, method, commodity, or apparatus. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising said element.

The foregoing describes specific embodiments of this specification. Other implementations are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. Multitasking and parallel processing are also possible or may be advantageous in certain embodiments.

Terms used in one or more embodiments of the present specification are for the purpose of describing specific embodiments only, and are not intended to limit the one or more embodiments of the present specification. As used in one or more embodiments of this specification and the appended claims, the singular forms "a", "the", and "the" are also intended to include the plural forms unless the context clearly dictates otherwise. It should also be understood that the term "and/or" as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

It should be understood that although the terms first, second, third, etc. may be used in one or more embodiments of the present specification to describe various information, the information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of one or more embodiments of this specification, first information may also be called second information, and similarly, second information may also be called first information. Depending on the context, the word "if" as used herein may be interpreted as "at" or "when" or "in response to a determination."

The above descriptions are only preferred embodiments of one or more embodiments of this specification, and are not intended to limit one or more embodiments of this specification. Within the spirit and principles of one or more embodiments of this specification, Any modification, equivalent replacement, improvement, etc. should be included in the scope of protection of one or more embodiments of this specification.

Claims (4)

1. A self-calibration optimization method of a distributed network, characterized in that, comprising the following steps: S1. Simplify the communication model of the distributed network into a star network topology model. For each distributed node, it acts as a central node, and other distributed nodes communicate with the central node point-to-point; S2, define the role of distributed network nodes, and deploy distributed network calibration software: Define the central node as the calibration client, and deploy the calibration client software on it; define the distributed nodes other than the central node as the calibration server, and deploy the calibration server software on it; The calibration client is used to send a ping packet to the calibration server, specify the size of the packet, specify the packet sending cycle, calculate the calibration server return packet delay, and record the number of packet return failures; The calibration server is used to respond to the ping packet sent by the calibration client, and receive the network parameter configuration command sent by the calibration client; S3, selecting a distributed network calibration parameter and a statistical target, the calibration parameter includes a ping packet size and a packet sending cycle, and the statistical target is the time required to complete 1GB data transmission and the number of lost packets; Select several files for the size of the ping packet used for calibration; record the network delay from the central node to the distributed node as T1, take several file sending cycles, and each file sending cycle is a different multiple of T1; different file ping packet size and sending cycle Form a set of calibration parameters; S4, distributed network self-calibration, including: The calibration client selects a set of calibration parameters, and sends ping packets continuously according to the calibration parameters. If the response is lost in the middle, the number of lost packets will be increased by 1. When the cumulative number of bytes sent by the ping packets reaches 1GB, the test of this group is completed, and the test of this group is obtained. The time T2 to complete the 1GB data transmission and the total number of packet loss N in this group; select another set of calibration parameters to repeat the test process, and obtain the T2 and N values corresponding to each set of calibration parameters; the group with the smallest T2 value is the optimal network bandwidth utilization group; the group with the smallest N value is the optimal group for network reliability; The user selects a set of network parameters from the optimal network bandwidth utilization group and the optimal network reliability group through the calibration client and sends them to the calibration server to complete the self-calibration optimization of the distributed network. 2. The self-calibration optimization method of a distributed network according to claim 1, wherein in step S3, the size of the ping packet used for calibration is 32, 64, 128, 256, 512, 1024 bytes There are 6 files in total, and the packet sending cycle is 1.5xT1, 2.0xT1, and 2.5xT1. There are 3 files in total. 18 sets of calibration parameters are constructed, and an evaluation table is formed with the corresponding time T2 for completing 1GB data transmission and the number of packet loss N. 3. The self-calibration optimization method of a distributed network according to claim 2, characterized in that, in step S4, the calibration process is initiated by the calibration client, and a set of calibration parameters is selected: ping packet 32 bytes, packet sending cycle 1.5xT1, the calibration client first presses a 32-byte ping packet, and takes 1.5 times the network delay T1 as the time interval from the last response receiving time to the sending of this ping packet, and the time difference between sending the ping packet and receiving the reply is recorded as t1 , t2, t3...tm, if the response is lost in the middle, the packet loss count N will be increased by 1. When the cumulative number of bytes sent by the ping packet reaches 1GB, the test of this group is completed, and the time for this group to complete the 1GB data transmission is T2 =t1+t2+t3...+tm, the total number of lost packets in this group is N; then select the next set of calibration parameters: ping packet 64 bytes, packet sending cycle 1.5xT1 for testing, until all sets of calibration parameters are tested, Get the full evaluation form. 4. A self-calibration optimization method for a distributed network according to any one of claims 1-3, characterized in that, in step S4, the group with the smallest T2 value indicates that the 1GB data transmission is completed at the fastest speed , is the optimal group for network bandwidth utilization; the group with the smallest N value means that the 1GB data transmission is completed with the least number of packet loss, and it is the optimal group for network reliability, so two optional tuning strategies are set.

Images