WO2024113803A1 - Network optimization method, edge device and computer-readable storage medium - Google Patents

Abstract

The present application discloses a network optimization method, an edge device and a computer-readable storage medium. The network optimization method comprises: after hotspot information in a network is acquired, determining a hotspot address and a hotspot level in the hotspot information; on the basis of the hotspot address and the hotspot level, determining an equivalent multi-path group reaching the hotspot address, wherein the equivalent multi-path group comprises the shortest path and an expansion path; and according to the equivalent multi-path group, transmitting traffic data corresponding to the hotspot information.

Description

Network optimization method, edge device and computer readable storage medium

Related Applications

This application claims priority to Chinese patent application No. 202211546033.7 filed on December 1, 2022, the entire contents of which are incorporated by reference into this application.

Technical Field

The present application relates to the field of communication technology, and in particular to a network optimization method, an edge device, and a computer-readable storage medium.

Background technique

Generally, the traffic of the entire network tends to be stable, but in some emergencies, or at special times, the network traffic will gather for a short time with some hot events, that is, the phenomenon of hot traffic will occur. However, when hot traffic occurs, the network cannot be adjusted in time according to the hot traffic situation, resulting in network congestion or even traffic loss, making users feel low about the network usage. Therefore, when hot traffic occurs, how to adjust the network to improve users' perception of network usage is an urgent problem to be solved.

Summary of the invention

The main purpose of this application is to provide a network optimization method, edge device and computer-renewable storage medium, aiming to solve the technical problem of how to improve the user's perception of network usage when hot traffic occurs.

To achieve the above objectives, the present application provides a network optimization method, which is applied to edge devices and includes the following steps:

After obtaining the hotspot information in the network, determining the hotspot address and hotspot level in the hotspot information;

Determine an equal-cost multi-path group to reach the hotspot address based on the hotspot address and the hotspot level, wherein the equal-cost multi-path group includes a shortest path and an extended path;

The traffic data corresponding to the hotspot information is transmitted according to the equal-cost multi-path group.

In addition, to achieve the above purpose, the present application also provides an edge device, the edge device The invention comprises: a memory, a processor and a network optimization program stored in the memory and executable on the processor. When the network optimization program is executed by the processor, the steps of the network optimization method described above are implemented.

In addition, to achieve the above-mentioned purpose, the present application also provides a computer-readable storage medium, on which a network optimization program is stored. When the network optimization program is executed by a processor, the steps of the network optimization method as described above are implemented.

The embodiment of the present application determines the equivalent multi-path group that reaches the hotspot address according to the hotspot level and the hotspot address in the hotspot information after receiving the hotspot information in the network, and distributes the traffic data to the equivalent multi-path group for transmission. Therefore, when hotspot traffic appears in the network, the network can be timely adjusted and globally optimized according to the hotspot information. Since the equivalent multi-path group includes the shortest path and the extended path, when the traffic data is transmitted through the equivalent multi-path group, network congestion or even traffic discarding when the traffic data is transmitted only through the shortest path is avoided, thereby ensuring that the traffic data is transmitted correctly and effectively, and improving the user's perception of network use.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of a flow chart of a first embodiment of a network optimization method of the present application;

FIG2 is a schematic diagram of a detailed process of constructing an equivalent multi-path group in the network optimization method of the present application;

FIG3 is a schematic diagram of the location of a reserved field with an extended function in an LACP protocol message;

FIG4 is a schematic diagram showing an example of a first embodiment of the network optimization method of the present application;

FIG5 is a schematic diagram of a flow chart of constructing an expansion path in the network optimization method of the present application;

FIG6 is a flow chart of the memory operation in the network optimization method of the present application.

The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings.

Detailed ways

It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

Generally speaking, the traffic data transmission status of the entire network is basically stable. However, when there are some emergencies, such as natural disasters, wars, hot news, etc., or some special events and special times, such as National Day, Spring Festival, World Cup, etc., the traffic data on the above network will be stable. The transmission status will show a short-term aggregation along with the above-mentioned hot content. At this time, the network is prone to congestion and paralysis, which has a high maintenance cost and seriously affects the user's network usage perception.

Therefore, in order to cope with possible network congestion and paralysis when hot content appears on the network, it is possible to predict the emergence of hot content on the network and adjust the entire network equipment in advance according to the predicted hot content. For example, according to the highest value expected in history and adding a certain amount of redundant bandwidth to ensure the network bandwidth occupancy in a certain place, such as during the World Cup match time, the entire network equipment is adjusted in advance to ensure network bandwidth occupancy. Or when the emergence of hot content causes network congestion and paralysis, reconfigure the congested nodes, share the traffic data among multiple nodes, and other operations to ensure the normal transmission of traffic data without packet loss. Alternatively, the server resources in the network can be clouded, and the resources to be accessed can be distributed on multiple servers as much as possible to minimize the impact of sudden traffic data accessing hot content on the server. The above-mentioned coping strategies have the following problems: when hot content suddenly appears, network nodes cannot be adjusted automatically in real time and are basically in a passive response state; or they can only make corresponding advance adjustments to the hot spots that are predicted in advance, but there is no prepared strategy to face the variability of the situation. For example, if the traffic data for accessing hot content is too large, far exceeding the historical expected maximum value, it will still be congested and paralyzed, or if the traffic data is too small, far less than the historical expected maximum value, it will cause cost waste. Although server cloudification can partially alleviate the problem of traffic sharing when hot content appears, if the traffic data far exceeds the network bandwidth of the server where the hot content is located, then network congestion will still occur.

To solve the above problems, the embodiments of the present application provide a network optimization method, an edge device, and a computer-readable storage medium. When hot content appears in the data-bearing network, the entire network automatically adapts and adjusts to reasonably and evenly distribute the traffic data for accessing the hot content on the network as much as possible, so as to achieve the purpose of automatically adjusting the hot traffic data. At the same time, the priority of the hot traffic data in the network is improved, so that the above hot traffic data has a high priority transmission in the existing network bandwidth, ensuring the correct and effective transmission of the hot traffic data. Through this embodiment, the intelligent operation and maintenance and adaptation capabilities of the network can be improved, the overall performance of the network can be improved, and the network maintenance costs can be saved, which can significantly improve the user's perception of network use.

The embodiments of the present application are further described below in conjunction with the accompanying drawings.

Referring to FIG. 1 , the present application provides a network optimization method. In a first embodiment of the network optimization method, the network optimization method is applied to an edge device, including the following steps:

Step S10, after obtaining the hotspot information in the network, determine the hotspot in the hotspot information Address and hotspot levels;

For example, after a hot content appears in the network, the number of visits to a certain server or certain servers in the network will increase sharply, and at this time it can be considered that the hot content is in a certain server or certain servers. In one scenario, the server can detect whether the number of visits to a certain file or certain data of its own has increased sharply. If the number of visits is much greater than the preset historical number of visits, it is determined that the number of visits has increased sharply, and at this time it can be considered that the hot content appears in the server. Among them, the server can be a cloud server.

Exemplarily, after determining that hot content has appeared in the server, the hot information of the hot content can be notified to any network device in the network where the server is located, such as an edge device. Among them, the server can inform the edge device through a link connected to the switch. That is, the server can inform the switch of the hot information of the hot content appearing in the server through a specific message or message, that is, inform the switch that the IP address of this server or the IP address of a virtual machine on this server has become a hot IP address. Based on the IP address message of the hot content received from the server, the switch converts this address into hot information and notifies the network through the routing protocol. The routing protocol may be OSPF (Open Shortest Path First), BGP (Border Gateway Protocol), IGP (Interior Gateway Protocol), RIP (Routing Information Protocol), etc. The reserved field of the routing protocol may be used during notification to inform the hotspot IP address information to the neighboring devices related to the entire network in a specific message. For example, if the entire network uses the OSPF protocol, when notifying the server address LSA message, this address may be marked as a hotspot address. This allows the neighbors related to the entire network to know that the route corresponding to this IP address in the network has become a hotspot route. While notifying the above hotspot address, the reserved field of the routing protocol may be used to notify the hotspot level of the hotspot address. The hotspot content is positively correlated with the hotspot level, for example, the denser the hotspot content, the higher the hotspot level. The edge device may be a computer, a mobile phone or other smart terminal.

Among them, specific messages can be extensions of existing protocols or sent using private protocols. The link can be a link aggregation group, then the server can communicate with the network device through the switch and according to the preset protocol, such as the LACPDU (Link Aggregation Control Protocol Data Unit) message in the LACP (Link Aggregation Control Protocol), to inform the network device of the hotspot information. For example, as shown in Figure 3, the LACPDU unit is a 128-byte message, the protocol type is the field from the 25th byte to the 26th byte, the default protocol type is 0x8809, and the message subtype is the field of the 27th byte, the default The hotspot level can be informed to the network device through the expansion of the reserved field A in the 34th to 36th bytes of the LACPDU unit; the 2-byte protocol type feature field and the 1-byte message subtype feature field in the message are used to determine whether it is the LACP protocol. If so, the LACP protocol is determined to be the routing protocol corresponding to the hotspot information.

For example, after the edge device detects the existence of hotspot information in the network, it is necessary to determine the hotspot level and hotspot address contained in the hotspot information, so as to subsequently select the combined path according to the hotspot level and hotspot address to transmit the traffic data corresponding to the hotspot information.

Exemplarily, the hotspot information includes at least a hotspot address and a hotspot level. The hotspot address may include the IP (Internet Protocol) address of the server where the hotspot content appears or the IP address of a virtual machine on the server. The hotspot level may be graded based on the size of the traffic data of accessing the hotspot content in the server in the network, and the grade division method may be to divide the level according to the frequency of simultaneous access of the hotspot content in the server by the same user or multiple users. For example, if the frequency of simultaneous access to the hotspot content in the server is n times more than the preset historical access frequency, then one level is added; it may also be graded based on the average number of accesses to the hotspot content in the server within a preset time period, for example, if the average number of accesses to the hotspot content in the server within one minute is greater than or equal to the preset grade division threshold, then one level is added. Moreover, the hotspot level may be directly presented in a digital identifier. For example, 0-10 levels may be used to refer to the hotspot level, and the larger the number, the higher the level, corresponding to the larger the traffic data of accessing the hotspot content in the above server. In particular, hotspot level 0 refers to the disappearance of hotspot traffic data, which is a normal traffic data transmission situation. It should be noted that the historical access frequency, n times and preset level classification threshold can be set by the user independently and are not restricted here.

Step S20, determining an equal-cost multi-path group to reach the hotspot address based on the hotspot address and the hotspot level, wherein the equal-cost multi-path group includes a shortest path and an extended path;

Exemplarily, after the edge device obtains the hotspot address and hotspot level of the hotspot information, in order to avoid network congestion when subsequent edge devices query the hotspot content, multiple paths to the hotspot address (such as the shortest path and the extended path) can be constructed based on the hotspot address and the hotspot level, and they can be aggregated to form an equivalent multi-path group.

When constructing an equal-cost multi-path group, the shortest path from the edge device to the server can be determined as a path in the equal-cost multi-path group. For example, if the network uses the OSPF protocol, the shortest path can be calculated based on the SPF (Shortest Path First) algorithm in the OSPF protocol, such as using the number of hops in the route and the route cost (path overhead) value to calculate the shortest path. For example, If the reference bandwidth is 100Mbps (i.e., 10 ⁸ bits of data are transmitted per second) and the actual bandwidth is 10Mbps, then cost = 100/10 = 10. In the OSPF protocol path calculation process, the cost value of each interface in the link is calculated separately, and then the path cost value required to reach the target node from the current node is accumulated. The path with the smallest path cost value is the shortest path. After calculating the shortest path from the edge device to the server where the hot content is located, the shortest path can be written into the routing table of the edge device.

Exemplarily, in order to avoid network congestion, at least one other path to the above-mentioned server can be expanded in the edge device according to the above-mentioned hotspot level, that is, the expanded path. And the expanded path can be a path that determines all qualified and reachable hotspot addresses according to the range of preset cost values corresponding to the above-mentioned hotspot level. For example, the edge device first determines different preset cost value ranges according to the hotspot level, and then screens each valid path according to the preset cost value range to obtain an expanded path that meets the requirements. Among them, the effective path is the path that the edge device can effectively reach the server. And the cost value range is not fixed, and the cost value range can be increased or reduced according to the persistence of the hotspot information of the server where the hotspot content appears or the number of user connections. Correspondingly, the larger the cost value range, the more expanded paths. And after determining the expanded path, the expanded path can be written into the routing table of the edge device as a path in the equivalent multi-path group.

Step S30: Transmitting traffic data corresponding to the hotspot information according to the equal-cost multi-path group.

Exemplarily, after the equal-cost multi-path group is constructed in the edge device, the server can be accessed through the equal-cost multi-path group consisting of the shortest path and at least one extended path to load share the traffic, that is, the edge device can access the server through any one or more equal-cost paths in the equal-cost multi-path group, and the server transmits traffic data to the edge device that applies for access through any one or more equal-cost paths in the equal-cost multi-path group. For example, the messages communicated between the edge device and the server belong to the same data stream, then the equal-cost multi-path group that implements load sharing based on the data stream can evenly use all the above paths to transmit the data stream. For example, if the above message has 10 data streams and there are 2 equal-cost paths to choose from, then 5 data streams are transmitted on each side.

Exemplarily, after transmitting the traffic data corresponding to the above hotspot information according to the equal-cost multi-path group, if hotspot upgrade information in the network is also received, the hotspot upgrade information is used as the hotspot information to continue to perform the above step S10 of determining the hotspot address and hotspot level in the hotspot information. When the edge device receives the hotspot disappearance information in the network, the edge device can directly continue to transmit the untransmitted traffic data in the traffic data according to the shortest path.

Moreover, in this embodiment, all network resources that are physically connected to the server where the hotspot traffic data (i.e., the traffic data corresponding to the hotspot information) is located can be scheduled and adjusted, and the transmission input of all network resources that are physically connected to it can be gradually increased according to the demand for the hotspot traffic data, until all network resources that are physically connected to it are fully invested in the transmission of the hotspot traffic data or until the hotspot content disappears and the hotspot traffic data is reduced. Among them, the method of adjusting the transmission can be expanded using a preset routing protocol, which can be an OSPF protocol. When the edge network device (such as a router) performs routing calculations, the cost (path overhead) value of the path to the upper server is expanded according to the hotspot level of the server, so as to calculate more paths and invest them in the load balancing transmission process of the hotspot traffic data to achieve the adjustment effect of traffic sharing.

In addition, to assist in understanding the network optimization method flow in this embodiment, an example is given below.

For example, as shown in FIG4 , there is hotspot content in the network, and the network uses the OSPF protocol for routing calculation. Assuming that the hotspot information is stored on the hotspot server 10.1.1.1, after the hotspot server detects that a file with hotspot access content appears on itself, it informs the network device 2 directly connected to the hotspot server of the message that it has become a hotspot through a message, and the network device 2 receives the message that the hotspot server has become a hotspot and the hotspot level of the corresponding hotspot. Network device 2 calculates the LSA packet containing the hotspot address and the hotspot level through the OSPF protocol, and broadcasts it in the above-mentioned bearer network. After the intermediate node (network device A, network device B and network device C) receives the above-mentioned LSA packet, it only calculates the above-mentioned hotspot address 10.1.1.1 through the OSPF protocol, and then continues to broadcast. After the edge device (network device 1) of the network receives the LSA packet, it calculates the hotspot address 10.1.1.1 and the hotspot level is hotspot level 1, and then sets the route calculation cost range to 10. The edge device starts to calculate the route and finds that the shortest path to the hotspot server IP address 10.1.1.1 is cost 50, that is, it passes through network device A to reach network device 2, and the next hop is 20.1.1.1. Then, according to the cost range cost10, the calculated path cost is expanded and it is found that there is another path to 10.1.1.1 with a cost of 60, that is, it passes through network device C, and the next hop is 30.1.1.1. When calculating on the edge device, according to the cost range set to 10, the cost of the second path is changed from 60 to 50, and two equal-cost paths in the ECMP group with a cost of 50 are calculated.

The edge device writes the path to 10.1.1.1 into the routing table as an ECMP path, and the next hop exits are 20.1.1.1/30.1.1.1. At this time, the traffic data sent to the hotspot server 10.1.1.1 will be load-shared by the ECMP group to the above two paths for forwarding, thereby achieving load sharing. As the hot content on the hotspot server 10.1.1.1 continues to heat up and the number of visits continues to increase, the hotspot server will be ranked The upgraded information is notified to the directly connected network device 2. When the network device resolves that the hotspot level of the hotspot server is level 3, network device 2 generates an LSA packet with the 10.1.1.1 address of the OSPF protocol, and adds the hotspot level 3 to the reserved field of the above protocol, and then broadcasts the LSA packet in the bearer network. After receiving the LSA packet, the intermediate nodes (network device A, network device B, and network device C) resolve the hotspot address for broadcasting. After receiving the LSA packet, the edge device resolves the hotspot address and hotspot level 3, and sets the route calculation cost range to 30. The edge device starts the route calculation and obtains that the shortest path to the hotspot server with an IP address of 10.1.1.1 is cost 50, that is, it passes through network device A, and the next hop is 20.1.1.1. Then, the cost of the path is calculated based on the cost range. It is found that the cost of the second path to 10.1.1.1 is 60, that is, it passes through network device C, and the next hop is 30.1.1.1, and the cost of the third path to 10.1.1.1 is 80, that is, it passes through network device B and network device C, and the next hop is 40.1.1.1; when the edge device calculates, the cost of the second and third paths is modified to 50 according to the cost range, and three ECMP group equivalent paths with a cost of 50 are calculated. The edge device writes the path to the hotspot server 10.1.1.1 into the routing table as an ECMP path, and the next hop exits are 20.1.1.1/30.1.1.1/40.1.1.1 respectively. At this time, the traffic data sent to the hotspot server is load-shared by the ECMP group to the above three equal-cost paths for forwarding, realizing load sharing.

When the hotspot content on the hotspot server 10.1.1.1 disappears, the server notifies the directly connected bearer network device 2 of the hotspot disappearance information. Network device 2 finds that the server sends a message that the hotspot disappears, and the hotspot level is 0, then network device 2 generates an LSA packet for the 10.1.1.1 address of the OSPF protocol, and adds the hotspot level to 0 in the protocol reserved field, and then broadcasts it in the bearer network. After receiving the LSA packet, the intermediate node (network device A, network device B, and network device C) parses the hotspot address for broadcasting. After receiving the above LSA packet, the edge device parses and obtains the hotspot level as 0, then starts routing calculation, deletes the original ECMP group route to 10.1.1.1, and calculates that the shortest path to the server IP address 10.1.1.1 is cost 50, and the next hop is 20.1.1.1. The edge device writes the path to 10.1.1.1 into the routing table as a normal route, and the next hop exit is 20.1.1.1. At this time, the traffic data sent to the server is forwarded through the shortest path.

In this embodiment, after receiving the hotspot information in the network, an equal-cost multi-path group to the hotspot address is determined according to the hotspot level and the hotspot address in the hotspot information, and the traffic data is shared to the equal-cost multi-path group for transmission. Therefore, when hotspot traffic appears in the network, the network can be adjusted and globally optimized according to the hotspot information in a timely manner. Since the equal-cost multi-path group includes the shortest path and the extended path, when the traffic data is transmitted through the equal-cost multi-path group, it is avoided that only the shortest path is used. When transmitting traffic data over short paths, network congestion or even traffic discarding may occur. This ensures that traffic data is transmitted correctly and effectively, improving users' perception of network usage.

Based on the first embodiment of the present application, a second embodiment of the network optimization method of the present application is proposed. In this embodiment, referring to FIG. 2 , the above step S20 determines an equivalent multi-path group to reach the hotspot address based on the hotspot address and the hotspot level, including:

Step a, determining the shortest path to the hotspot address, and obtaining a first path cost value of the shortest path;

Exemplarily, after the edge device in the network obtains the hotspot address and hotspot level in the hotspot information of the server through the preset communication protocol analysis and calculation, the shortest path to the hotspot address and its corresponding cost value are calculated through the preset routing protocol, and the cost value of the shortest path is used as the first path cost value. For example, if the network uses the OSPF protocol, the network device calculates the hotspot address and hotspot level in the LSA group through the OSPF protocol, and then calculates the shortest path and its corresponding cost value according to the route of the edge device and the hotspot address, and uses the cost value as the first path cost value of the shortest path.

Step b, determining a path cost value range according to the hotspot level and the first path cost value;

Exemplarily, after the edge device parses and obtains the hotspot level, the cost value range is determined according to the correspondence between the hotspot level and the first path cost value. For example, each hotspot level may correspond to a cost value range, such as hotspot level 5, corresponding to a cost value range of 50; when the first path cost value is 50, if the hotspot level is hotspot level 1, the cost value range is 10.

Step c, determining a path whose path cost satisfies the path cost value range as an extended path;

Exemplarily, the path cost of each path is first obtained, and the path whose path cost satisfies the path cost value range is used as the extended path. For example, an extended path to the hotspot address is constructed based on the cost value 50 of the shortest path and the cost value range 10 corresponding to the hotspot level 1, that is, the path corresponding to the cost value greater than 50 but less than or equal to 10 is used as the extended path to the hotspot address, and accordingly, the cost value of the extended address is modified to the cost value 50 corresponding to the shortest path, so as to add more extended paths to the ECMP group.

Step d: forming the equal-cost multi-path group according to the shortest path and the extended path.

Exemplarily, after the shortest path and the extended path group are obtained, an equal-cost multi-path group can be directly constructed, that is, the equal-cost multi-path group at least includes the shortest path and the extended path group.

In this embodiment, it is only necessary to modify the routing protocol to enable it to transmit hotspot information, which may be the hotspot address and hotspot level, and adjust the cost range of the path calculation for the received hotspot information on the edge device, so as to increase the load sharing path of the bearer network reaching the server, thereby achieving the effect of automatically sharing the hotspot traffic data. This embodiment can improve the intelligent operation and maintenance and adaptation capabilities of the network, improve the overall performance of the network, save network maintenance costs, and improve users' perception of network use.

In one embodiment, step b, determining a path cost value range according to the hotspot level and the first path cost value, includes:

Step b1, obtaining a preset path cost value corresponding to the hotspot level;

Step b2, using the first path cost value as a lower limit path cost value;

Step b3, taking the sum of the preset cost value and the first path cost value as the upper limit path cost value;

Step b4: taking the range between the lower limit path cost value and the upper limit path cost value as the path cost value range.

Exemplarily, the hotspot level can be set according to the needs of the user. For example, the hotspot level can be 0-10. When the hotspot corresponding to hotspot level 0 disappears, the preset path cost value corresponding to each increase in the hotspot level increases by a preset amount. For example, if the preset path cost value corresponding to hotspot level 1 is 10, then the preset path cost value corresponding to hotspot level 2 is 20. When the edge device obtains the hotspot level corresponding to the server, it can determine the preset path cost value corresponding to the above hotspot level. It should be noted that the above 0-10 levels and the corresponding preset path cost values are only for the convenience of understanding of this embodiment, and do not mean that it can only be set in this way. The setting of the hotspot level and the corresponding preset path cost value can be arbitrary by the user and will not be repeated here.

Exemplarily, after determining the preset path cost value corresponding to the hotspot level, the sum of the preset cost value and the first path cost value corresponding to the shortest path can be directly used as the upper limit path cost value in the path cost value range, and the first path cost value can be used as the lower limit path cost value in the path cost value range. For example, if the first path cost value corresponding to the shortest path is 50, the hotspot level is level three, and the preset path cost value corresponding to the hotspot level is 30, then the path cost value range is [50, 80].

In this embodiment, the first path cost value corresponding to the shortest path is used as the lower limit path cost value in the path cost value range, and the sum of the preset path cost value corresponding to the hotspot level and the first path cost value is used as the upper limit path cost value in the path cost value range, thereby ensuring that the obtained The validity of the obtained path cost value range.

In one embodiment, referring to FIG. 5 , step c, determining a path whose path cost satisfies the path cost value range as an extended path, includes:

Step c1, determining all paths to the hotspot address, and determining other paths among all the paths except the shortest path;

Step c2, determining the second path cost values corresponding to the other paths, and determining the matching path cost values that meet the path cost value range among the second path cost values, and using the other paths corresponding to the matching path cost values as extended paths.

Exemplarily, after determining the shortest path to the hotspot address, all other paths except the shortest path can be determined. The second path cost value includes the path costs of other paths. Then compare each second path cost value with the path cost value range to determine the second path cost value (i.e., the matching path cost value) that is exactly within the path cost value range, and use the corresponding other paths as extended paths. For example, the cost value corresponding to the shortest path is cost50, and the path cost values corresponding to the other determined paths are cost60, cost70, and cost55. The determined hotspot level is hotspot level 1, and the corresponding preset path cost value is cost10. Then the path with a path cost value less than or equal to cost60 in other paths is a matching path, that is, the path cost value of cost55 and other paths corresponding to cost60 are matched. The shortest path and the other matched paths are used as equivalent exports of the ECMP group.

In this embodiment, by determining the matching path cost values satisfying the path cost value range among the second path cost values corresponding to other paths, and using the corresponding other paths as the extended paths, the validity of the acquired extended paths is ensured.

In one embodiment, step d, forming the equal-cost multi-path group according to the shortest path and the extended path, comprises:

Step d1, at the edge device, adjusting the path cost value of the extended path to be the same as the first path cost value of the shortest path, and constructing an equal-cost multi-path group according to the adjusted extended path and the shortest path.

For example, in the edge device, after the extended path is determined, the path cost value of the extended path can be adjusted, for example, to be the same as the first path cost value of the shortest path. For example, if the first path cost value is 50 and the path cost value of the extended path is 60, the path cost value of the extended path can be modified to 50 in the edge device. Then, the extended path with the adjusted path cost value can be The paths and shortest paths are filled in the equal-cost multi-path group to obtain the equal-cost multi-path group for practical application. It should be noted that when adjusting the extended path, it is only performed in the edge device and not on the server side. Therefore, on the server side, the path cost value of the extended path is still the actual path cost value. For example, if the path cost value of the extended path is 60 and becomes 50 after adjustment in the edge device, then the path cost value of the extended path is displayed as 50 on the edge device side, and the path cost value of the extended path is displayed as 60 on the server side.

In this embodiment, a path cost value that is the same as the first path cost value is assigned to the extended path in the edge device so that an equal-cost multi-path group is constructed based on the extended path and the shortest path. This makes the path cost values of all paths in the equal-cost multi-path group in the edge device consistent, thereby avoiding the phenomenon of accessing the server from one path during subsequent hot spot information access, which causes network congestion.

In one embodiment, determining the hotspot address and the hotspot level in the hotspot information includes:

Step e, determining the routing protocol corresponding to the hotspot information according to the characteristic field in the hotspot information;

Step f, extracting the hotspot address and hotspot level in the hotspot information according to the routing protocol.

Exemplarily, the switch notifies the network of the IP address message of the hotspot information received from the server side through the routing protocol (it can be notified through routing protocols such as IGP, BGP, etc.), and the reserved field of the routing protocol can be used when notifying to inform the hotspot IP address information to the neighboring devices related to the whole network, such as edge devices. After detecting the hotspot information, the edge device can extract the characteristic field carried in the hotspot information to determine the routing protocol (such as BGP, IGP, etc.) used by the message corresponding to the hotspot information. For example, if the LACPDU unit is a 128-byte message, the protocol type is the field from the 25th byte to the 26th byte, the default protocol type is 0x8809, and the message subtype is the field of the 27th byte, which defaults to 0x01 (i.e., LACP message). The hotspot level can be extended to inform the above network devices through the reserved field A in the 34th byte to the 36th byte in the LACPDU unit; whether it is the LACP protocol can be determined by the 2-byte protocol type characteristic field and the 1-byte message subtype characteristic field in the above message. After the routing protocol is determined, the hotspot address and hotspot level in the hotspot information can be directly extracted according to the routing protocol.

In this embodiment, the routing protocol is determined according to the characteristic field in the hotspot information, and then the hotspot address and hotspot level in the hotspot information are extracted according to the routing protocol, thereby ensuring that the edge device can effectively obtain the hotspot address and hotspot level.

In one embodiment, the traffic volume corresponding to the hotspot information transmitted by the equal-cost multi-path group is After that, including:

Step g: after receiving the hotspot upgrade information in the network, continue to execute the step of determining the hotspot address and hotspot level in the hotspot information by using the hotspot upgrade information as the hotspot information.

Exemplarily, after detecting that the hotspot content in the network continues to heat up, that is, after receiving the hotspot upgrade information in the network, the equivalent-cost multi-path group continues to be updated, and then the traffic data corresponding to the hotspot information is transmitted according to the updated equivalent-cost multi-path group. Among them, the method of updating the equivalent-cost multi-path group can be to first determine the new hotspot level corresponding to the hotspot upgrade information, and then determine the new expansion path according to the new hotspot level, and then update the equivalent-cost multi-path group according to the new expansion path.

Exemplarily, after detecting that the popularity of hot spot information in the network has decreased, that is, the hot spot level of the hot spot information has decreased, the equivalent-cost multi-path group can also be updated, and when updating the equivalent-cost multi-path group, the latest extended path can be determined based on the downgraded hot spot level, and the equivalent-cost multi-path group can be updated based on the latest extended path. After obtaining the updated equivalent-cost multi-path group, the traffic data corresponding to the hot spot information is transmitted based on the updated equivalent-cost multi-path group.

In this embodiment, after detecting the hotspot upgrade message, the step of determining the hotspot address and hotspot level in the hotspot information is continued, thereby avoiding the phenomenon of network congestion or even traffic discard caused by the hotspot level being too high.

In one embodiment, after the traffic data corresponding to the hotspot information is transmitted according to the equal-cost multi-path group, the method includes:

Step h, after receiving the hotspot disappearance information in the network, deleting the equal-cost multi-path group;

Step x: Continue to transmit the untransmitted flow data in the flow data according to the shortest path.

For example, after detecting that the hotspot in the network has disappeared, that is, the hotspot information has been restored to normal information, that is, after receiving the hotspot disappearance information, the equivalent multi-path group can be deleted at this time, and the untransmitted traffic data in the traffic data can be directly transmitted according to the shortest path to save resources.

In this embodiment, after detecting that the hotspot disappears, the equal-cost multi-path group is directly deleted and transmission is performed according to the shortest path, thereby saving costs and ensuring normal transmission of traffic data in the network.

In addition, the present application also provides an edge device, which includes a memory, a processor, and a network optimization program stored in the memory and executable on the processor. When the network optimization program is executed by the processor, the steps of the network optimization method described above are implemented.

In addition, in one embodiment, FIG6 is a schematic diagram of the structure of an edge device of an embodiment of the present application. As shown in FIG6, at the hardware level, the edge device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. Among them, the memory may include a memory, such as a high-speed random access memory (Random-Access Memory, RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage, etc. Of course, the edge device may also include hardware required for other services. The processor, the network interface, and the memory may be interconnected through an internal bus, and the internal bus may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one bidirectional arrow is used in FIG6, but it does not mean that there is only one bus or one type of bus. The memory is used to store programs. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a shared resource access control device at the logical level. The processor executes the program stored in the memory and is specifically used to execute the steps of the above-mentioned anti-channelling control method.

The specific implementation methods of the edge device of the present application are basically the same as the embodiments of the above-mentioned network optimization method, and will not be repeated here.

In addition, to achieve the above-mentioned purpose, the present application also provides a computer-readable storage medium, on which a network optimization program is stored, and when the network optimization program is executed by a processor, the steps of the network optimization method as described above are implemented.

The specific implementation of the computer-readable storage medium of the present application is basically the same as the various embodiments of the above-mentioned network optimization method, and will not be repeated here.

A person of ordinary skill in the art will appreciate that all or some of the steps, systems, and functional modules/units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between the functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium. Medium (or non-transitory medium) and communication medium (or temporary medium). As known to those of ordinary skill in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. In addition, it is known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and may include any information delivery medium.

The above are only optional embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims (10)

A network optimization method, applied to an edge device, comprising: After obtaining the hotspot information in the network, determining the hotspot address and hotspot level in the hotspot information; Determine an equal-cost multi-path group to reach the hotspot address based on the hotspot address and the hotspot level, wherein the equal-cost multi-path group includes a shortest path and an extended path; The traffic data corresponding to the hotspot information is transmitted according to the equal-cost multi-path group. The network optimization method according to claim 1, wherein the determining of an equal-cost multi-path group to reach the hotspot address based on the hotspot address and the hotspot level, wherein the equal-cost multi-path group includes a shortest path and an extended path, comprises: Determine the shortest path to the hotspot address, and obtain a first path cost value of the shortest path; determining a path cost value range according to the hotspot level and the first path cost value; Determine a path whose path cost satisfies the path cost value range as the extended path; The equal-cost multi-path group is formed according to the shortest path and the extended path. The network optimization method according to claim 2, wherein the determining the path cost value range according to the hotspot level and the first path cost value comprises: Obtaining a preset path cost value corresponding to the hotspot level; Using the first path cost value as a lower limit path cost value; Taking the sum of the preset cost value and the first path cost value as the upper limit path cost value; The range between the lower limit path cost value and the upper limit path cost value is taken as the path cost value range. The network optimization method according to claim 2, wherein determining a path whose path cost satisfies the path cost value range as an extended path comprises: Determine all paths to the hotspot address, and determine the shortest path among all the paths. Other paths outside the path; Determine the second path cost values corresponding to the other paths, determine the matching path cost values that meet the path cost value range among the second path cost values, and use the other paths corresponding to the matching path cost values as extended paths. The network optimization method according to claim 2, wherein the forming of the equal-cost multi-path group according to the shortest path and the extended path comprises: The path cost value of the extended path is adjusted at the edge device to be the same as the first path cost value of the shortest path, and the equal-cost multi-path group is constructed according to the adjusted extended path and the shortest path. The network optimization method according to claim 1, wherein the determining the hotspot address and hotspot level in the hotspot information comprises: Determine the routing protocol corresponding to the hotspot information according to the characteristic field in the hotspot information; The hotspot address and the hotspot level in the hotspot information are extracted according to the routing protocol. The network optimization method according to any one of claims 1 to 6, wherein after transmitting the traffic data corresponding to the hotspot information according to the equal-cost multi-path group, the method further comprises: After receiving the hotspot upgrade information in the network, the hotspot upgrade information is used as the hotspot information to continue to perform the step of determining the hotspot address and hotspot level in the hotspot information. The network optimization method according to any one of claims 1 to 6, wherein after transmitting the traffic data corresponding to the hotspot information according to the equal-cost multi-path group, the method further comprises: After receiving the hotspot disappearance information in the network, deleting the equal-cost multi-path group; The untransmitted flow data in the flow data continues to be transmitted according to the shortest path. An edge device, wherein the edge device comprises: a memory, a processor, and a network optimization program stored in the memory and executable on the processor, wherein the network optimization program, when executed by the processor, implements the steps of the network optimization method according to any one of claims 1 to 8. A computer-readable storage medium, wherein a network optimization program is stored on the computer-readable storage medium, and when the network optimization program is executed by a processor, the steps of the network optimization method according to any one of claims 1 to 8 are implemented.