A network communication method and device

Abstract

This application relates to communication technologies, and in particular, to a network communication method and apparatus. When a backbone network corresponding to a certain server fails, the method and apparatus can enable an acceleration instance to access the server through the public network in a timely manner, ensuring the access performance of the server. The method includes: instructing each of a plurality of acceleration instances to access a server associated with the acceleration instance through a backbone network corresponding to the acceleration instance; the backbone network corresponding to the acceleration instance is the backbone network corresponding to the server associated with the acceleration instance among a plurality of backbone networks; instructing each of the plurality of acceleration instances to detect the backbone network corresponding to the acceleration instance; when a detection result of a first acceleration instance among the plurality of acceleration instances detecting a first backbone network indicates that the first backbone network fails, instructing the first acceleration instance to access a first server through the public network, so that the first acceleration instance sends an access request to the first server through the public network.

Description

Technical Field

[0001] This application relates to communication technologies, and in particular, to a network communication method and apparatus. Background Art

[0002] With the continuous development of communication technologies, more and more users start to use services provided by a server. To ensure the access performance of users to the server, service providers provide acceleration services, such as global accelerator (GA) services. The acceleration service is implemented through a backbone network deployed by the service provider, and the backbone network can provide data transmission services with high reliability and low latency. One end of the backbone network is connected to a point of presence (POP), and the other end is connected to the server. In the acceleration service, a user accesses the backbone network through the network service providing point, and then accesses the server through the backbone network.

[0003] Generally, service providers usually provide multiple servers, and different servers are deployed in different regions. Different servers are connected to different backbone networks. The same network service providing point is connected to multiple backbone networks to connect to different servers through different backbone networks among the multiple backbone networks. The backbone network may fail, resulting in problems such as interruption of the backbone network. In related technologies, when all the backbone networks connected to the network service providing point fail, the network service providing point switches the access request to the public network (Internet) to access the server through the public network. Since all backbone networks usually do not fail simultaneously, if the backbone network connected to a certain server fails while other backbone networks do not fail for a long time, it is impossible to access the server through the public network for a long time, which causes the access performance of the server to decline and affects the user experience. Summary of the Invention

[0004] This application provides a network communication method and apparatus, which can enable an acceleration instance to access a server through the public network in a timely manner when the backbone network corresponding to the server fails, without waiting for all backbone networks to fail, thereby ensuring the access performance of the server and the user experience.

[0005] In a first aspect, a network communication method is provided. This method is applied to a network service providing point, and the system where the network service providing point is located further includes multiple backbone networks and multiple service ends. Among them, each backbone network in the multiple backbone networks corresponds to at least one service end in the multiple service ends. Multiple acceleration instances are deployed at the network service providing point, and each acceleration instance in the multiple acceleration instances is associated with at least one service end in the multiple service ends. The method includes: instructing each acceleration instance in the multiple acceleration instances to access the service end associated with the acceleration instance through the backbone network corresponding to the acceleration instance, where the backbone network corresponding to the acceleration instance is the backbone network in the multiple backbone networks corresponding to the service end associated with the acceleration instance; instructing each acceleration instance in the multiple acceleration instances to detect the backbone network corresponding to the acceleration instance; when the detection result of the first acceleration instance in the multiple acceleration instances detecting the first backbone network indicates that the first backbone network fails, instructing the first acceleration instance to access the first service end through the public network, so that the first acceleration instance sends an access request to the first service end through the public network, where the first service end is the service end in the multiple service ends associated with the first acceleration instance, and the first backbone network is the backbone network in the multiple backbone networks corresponding to the first service end.

[0006] Through this method, when the backbone network corresponding to the service end fails, the acceleration instance associated with the service end can promptly access the service end through the public network, ensuring the access performance and user experience of the service end. Moreover, there is no need to transmit the access request of the service end through the backbone network corresponding to other service ends, thus not affecting the access performance and user experience of other service ends.

[0007] In a possible implementation manner, the first acceleration instance belongs to a user. Instructing each acceleration instance in the multiple acceleration instances to detect the backbone network corresponding to the acceleration instance includes: receiving and recording the detection policy configured by the user for the first acceleration instance; instructing the first acceleration instance to detect the first backbone network according to the detection policy.

[0008] In this implementation manner, the user can configure the corresponding detection policy according to their business needs, etc., improving the fit between the detection policy and the business.

[0009] In a possible implementation, the multiple acceleration instances further include a second acceleration instance, which is associated with a second server among the multiple servers, and the second server corresponds to a second backbone network among the multiple backbone networks; when the detection result of the first acceleration instance among the multiple acceleration instances detecting the first backbone network indicates that the first backbone network fails, instructing the first acceleration instance to access the first server through the public network includes: when the detection result of the first acceleration instance detecting the first backbone network indicates that the first backbone network fails, and the detection result of the second acceleration instance detecting the second backbone network indicates that the second backbone network does not fail, instructing the first acceleration instance to access the first server through the public network, and instructing the second acceleration instance to continue to access the second server through the second backbone network.

[0010] In this implementation, only the acceleration instance corresponding to the faulty backbone network is accessed to the public network, while the acceleration instance corresponding to the non-faulty backbone network still accesses the backbone network without switching to the public network. This implementation realizes network conversion at the acceleration instance level.

[0011] In a possible implementation, after instructing the first acceleration instance to access the first server through the public network, the method further includes: instructing the first acceleration instance to continue to detect the first backbone network; when the detection result of the first acceleration instance continuing to detect the first backbone network indicates that the fault of the first backbone network is eliminated, instructing the first acceleration instance to access the first server through the first backbone network again.

[0012] In this implementation, when the fault of the backbone network is eliminated, the acceleration instance accesses the backbone network again to send an access request to the server through the backbone network, thus ensuring the access performance of the service.

[0013] In a possible implementation, instructing the first acceleration instance to access the first server through the public network includes: instructing the first acceleration instance to add the address of the network service providing point in the public network to the data packet whose destination is the first server, so that the first server sends the response packet of the data packet to the network service providing point through this address.

[0014] Among them, when the first server receives the data packet, it can obtain the address of the network service providing point in the public network. The first server sends the response packet of the data packet to the network service providing point through the address of the network service providing point in the public network. Thus, the network service providing point can deliver the response packet to the first acceleration instance, and the first acceleration instance outputs the response packet to the user. Among them, the data packet belongs to the service traffic, and the response packet of the data packet belongs to the return traffic of the service traffic. Thus, the homology and traceability of the traffic are realized.

[0015] In a possible implementation, the first server includes a cloud management platform and infrastructure. Cloud resources associated with the first acceleration instance are deployed in the infrastructure, and the cloud management platform is used to manage the cloud resources. Among them, the infrastructure includes at least one cloud data center, each cloud data center is provided with multiple servers, and the cloud resources are deployed in at least one server of the infrastructure. Instructing each acceleration instance among the multiple acceleration instances to access the server associated with the acceleration instance through the backbone network corresponding to the acceleration instance includes: instructing the first acceleration instance to send an access request to the cloud management platform through the first backbone network, and the cloud management platform is used to forward the access request to the cloud resources so that the access request accesses the cloud resources.

[0016] In a second aspect, a network communication device is provided. The device is configured at a network service providing point, and the system where the network service providing point is located further includes multiple backbone networks and multiple servers. Among them, each backbone network among the multiple backbone networks corresponds to at least one server among the multiple servers. Multiple acceleration instances are deployed at the network service providing point, and each acceleration instance among the multiple acceleration instances is associated with at least one server among the multiple servers. The device includes: a first indication module, configured to instruct each acceleration instance among the multiple acceleration instances to access the server associated with the acceleration instance through the backbone network corresponding to the acceleration instance, where the backbone network corresponding to the acceleration instance is the backbone network corresponding to the server associated with the acceleration instance among the multiple backbone networks; a second indication module, configured to instruct each acceleration instance among the multiple acceleration instances to detect the backbone network corresponding to the acceleration instance; a third indication module, configured to, when the detection result of the first acceleration instance among the multiple acceleration instances detecting the first backbone network indicates that the first backbone network fails, instruct the first acceleration instance to access the first server through the public network, so that the first acceleration instance sends an access request to the first server through the public network, where the first server is the server associated with the first acceleration instance among the multiple servers, and the first backbone network is the backbone network corresponding to the first server among the multiple backbone networks.

[0017] In a third aspect, a computing device cluster is provided, including at least one computing device, and each computing device includes a processor and a memory. The processor of at least one computing device is configured to execute instructions stored in the memory of at least one computing device, so that the computing device cluster executes the method provided in the first aspect.

[0018] In a fourth aspect, a computer-readable storage medium is provided, including computer program instructions. When the computer program instructions are executed by the computing device cluster, the computing device cluster executes the method provided in the first aspect.

[0019] In a fifth aspect, a computer program product including instructions is provided. When the instructions are run by the computer device cluster, the computer device cluster is caused to execute the method provided in the first aspect.

[0020] For the beneficial effects of the second to fifth aspects, reference may be made to the introduction of the beneficial effects of the first aspect above, which will not be elaborated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is a schematic diagram of a system provided by an embodiment of the present application;

[0022] Figure 2 is a schematic diagram of the backbone network between the server and the acceleration instance provided by an embodiment of the present application;

[0023] Figure 3 is a schematic diagram of a server provided by an embodiment of the present application;

[0024] Figure 4 is a schematic diagram of another server provided by an embodiment of the present application;

[0025] Figure 5 is a schematic diagram of a network service providing point provided by an embodiment of the present application;

[0026] Figure 6 is a flowchart of a network communication method provided by an embodiment of the present application;

[0027] Figure 7 is a flowchart of another network communication method provided by an embodiment of the present application;

[0028] Figure 8 is a schematic structural diagram of a network communication device provided by an embodiment of the present application;

[0029] Figure 9 is a schematic structural diagram of a computing device provided by an embodiment of the present application;

[0030] Figure 10 is a schematic structural diagram of a computing device cluster provided by an embodiment of the present application;

[0031] Figure 11 is a schematic structural diagram of another computing device cluster provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] The solutions provided by the embodiments of the present application will be described below in conjunction with the accompanying drawings. Among them, in the embodiments of the present application, "a plurality of" means two or more, and "a variety of" means two or more. "First", "second", etc. are only used to distinguish similar objects and do not have to be used to describe a specific order or the number of objects.

[0033] To facilitate understanding of the solutions provided by the embodiments of the present application, the technical terms that may be involved in the embodiments of the present application will be introduced first.

[0034] Cloud technology refers to a hosting service that unifies a series of resources such as hardware, software, and networks within a wide area network or a local area network to achieve data computing, storage, processing, and sharing.

[0035] Cloud management platform: A platform provided by cloud resource providers for interacting with users. Users can register accounts on the cloud management platform and rent cloud resources in the infrastructure managed by the cloud management platform using the accounts.

[0036] Infrastructure: Facilities used to provide cloud resources. The infrastructure may include at least one cloud data center, and each cloud data center is provided with multiple servers. The servers in the infrastructure can provide cloud resources.

[0037] Cloud resources: Also known as cloud services, are software and / or hardware used to implement services such as computing and storage. Cloud resources can include virtual machines and containers for computing services, buckets for object storage services, cloud hard disks, and other cloud resources.

[0038] Point of presence (POP) of network service: A service access node, device, equipment, or cluster established by a service provider. Users access the service end of the service provider through the point of presence of network service. In some embodiments, the point of presence of network service can be one or more physical computing devices, such as servers. In some embodiments, the point of presence of network service can be one or more virtual computing instances.

[0039] Virtual computing instance: Also known as a virtual computing device, refers to a complete computer system that is simulated by software, has a complete hardware system, and runs in a completely isolated environment. Among them, the hardware system of the virtual computing instance is a virtual hardware system obtained by virtualizing physical hardware. A complete virtual computing instance has an independent virtual computing system (such as a central processing unit (CPU)) and a virtual disk. Among them, a virtual computing instance that needs to communicate with the outside world also has an independent virtual network card. Typical virtual computing instances include virtual machines (VMs), containers, elastic cloud servers (ECSs), etc.

[0040] Backbone network: Also known as the cloud backbone network, it is a network deployed by service providers to connect network service provision points and servers. Among them, the network service provision points and servers connected through the backbone network are usually located in different regions. Different service providers usually have their own backbone networks to connect their network service provision points and servers in different regions. Service providers usually establish their own global backbone networks. In some embodiments, the backbone network supports the virtual extensible local area network (VXLAN) technology. Through the backbone network, message transmission can be carried out using the VXLAN technology.

[0041] Virtual extensible local area network (VXLAN): It is an overlay network technology that enables layer 2 intercommunication through layer 3. Among them, VXLAN uses the outer tunnel in the user datagram protocol (UDP) format as the data link layer, and the data message content and the VXLAN network identifier (VNI) are transmitted as the tunnel payload. Since UDP is used as the transmission means for the outer layer, the payload data can be easily transmitted in the layer 3 network.

[0042] Public network: Also known as the Internet, it is a globally interconnected internet that enables interconnection and interoperability between different internet service providers (ISPs). The core characteristics of the public network are global and open, and any user with the corresponding permissions and devices can access and use the public network. Each device connected to the public network has a unique public internet protocol (IP) address, and devices can directly access each other through the public IP address.

[0043] Global accelerator (GA) service: It refers to providing application acceleration services for global business users. By providing a unified public IP address and a highly reliable, low-latency, easy-to-manage, secure and compliant network service, users can quickly access cloud services globally and obtain a good experience. The global accelerator service generally uses anycast technology to publish the same IP address in multiple regions, and this address can also be called an anycast address, anycast IP address, etc.

[0044] Acceleration instance: Also known as an acceleration service instance, it is a network acceleration service instance purchased or leased by a user to access a server. The acceleration instance can be used by the user to access the server through the backbone network to improve access performance. Among them, the acceleration instance purchased or leased by the user to access the server is associated with the server. Exemplarily, the acceleration instance can be an application program deployed at a network service providing point.

[0045] Internet Control Message Protocol (ICMP): It is a sub-protocol of the Transmission Control Protocol (TCP) / IP protocol suite, used to transfer control messages between communication nodes or devices such as IP hosts and routers. Among them, control messages refer to messages about the network itself, such as whether the network is accessible, whether the host is reachable, and whether the route is available.

[0046] Anycast: It is a communication mode in the IP network, and can also be called geocast, unicast, etc. Among them, an anycast address corresponds to multiple target nodes, and anycast can be used to send the data packet of the source node to the target node that is the closest in terms of the network topology. Different from broadcast / multicast, multicast / broadcast allows the source node to send data to a group of target nodes, while anycast allows the source node to send data to one of a group of target nodes, and this node is selected by the routing system and is transparent to the source node. At the same time, the routing system selects the "nearest" node to provide services for the source node, which provides better services for the source node to a certain extent and also reduces the network load.

[0047] Same source and same trace: It means that the service traffic is sent out from a certain device point, and the return traffic of the service traffic is also received by the same device.

[0048] Server: It refers to the end where the service is deployed, and the user can rent or purchase one or more services deployed on the server. In some embodiments, the server can be a cloud server or a server in the form of a cloud. The cloud server can include a cloud management platform and infrastructure. In some embodiments, the server can also be a non-cloud form of server, which can be called an out-of-cloud server.

[0049] In the related art, when the backbone network connecting a certain server fails, the network service providing point sends the access request for the server to other backbone networks through the network between different backbone networks, so as to forward it to the server through other backbone networks. The transmission performance of transmitting the access request through the network between different backbone networks and other backbone networks is often lower than that of transmitting the access request through the public network, and it increases the network overhead of other backbone networks, affecting the access performance of the server corresponding to other backbone networks.

[0050] In addition, in this related art, when all the backbone networks connected to a network service provider point are interrupted, the network service provider point connects to the server through the public network to send the access requests received by the network service provider point to the server through the public network. Since all backbone networks usually do not fail simultaneously, if the backbone network connected to a certain server is interrupted while other backbone networks have not been interrupted for a long time, it will be difficult to access the server through the public network for a long time, which results in a decline in the access performance of the server and affects the user experience. Moreover, in this related art, if some of the multiple backbone networks connected to a network service provider point fail while the other part does not, the non-failed backbone networks have to transmit all the access requests received by the network service provider point, which leads to a decline in transmission performance and affects the access performance of each server corresponding to the network service provider point and the user experience of each user.

[0051] An embodiment of the present application provides a network communication method. This method can detect each backbone network among multiple backbone networks connected to a network service provider point. When it detects that the backbone network corresponding to any acceleration instance fails, it instructs the acceleration instance to access the public network to send the access requests that should have been transmitted by this backbone network to the server through the public network. In this method, when there is a backbone network failure, there is no need to transmit access requests through other backbone networks, thus not affecting the access performance of the servers corresponding to other backbone networks and the user experience. Moreover, when there is a backbone network failure, access requests can be sent through the public network in a timely manner, ensuring the access performance of the server corresponding to this backbone network and the user experience.

[0052] Next, the network communication method provided by the embodiment of the present application will be described.

[0053] Figure 1 A system that can be used to implement this method is shown. This system may include a network service provider point 100, multiple backbone networks, and multiple servers. Among them, the multiple backbone networks may include a first backbone network, a second backbone network, and a third backbone network, and the multiple servers may include a first server, a second server, a third server, etc. Each backbone network among the multiple backbone networks corresponds to at least one server among the multiple servers. For example, the first backbone network corresponds to the first server, the second backbone network corresponds to the second server, and the third backbone network corresponds to the third server. One end of the backbone network is connected to the network service provider point 100, and the other end is connected to the server corresponding to this backbone network. That is to say, there is a connection between the backbone network and the network service provider point 100, and there is a connection between the backbone network and the server corresponding to this backbone network. In addition, there is a connection between the network service provider point 100 and the public network, and there is also a connection between the server and the public network.

[0054] In some embodiments, the backbone network corresponding to the server refers to the backbone network among the multiple backbone networks whose transmission performance from the network service providing point 100 to the server is greater than that of the public network. Among them, the metrics for measuring the transmission performance may include latency, throughput, etc.

[0055] Without considering the transmission performance, for any server among the multiple servers, any backbone network among the multiple backbone networks may be reachable. That is to say, any backbone network among the multiple backbone networks can transmit the access request from the network service providing point 100 to the server. Among them, the transmission performance of some backbone networks for transmitting the access request of the server is greater than that of the public network for transmitting the access request, and the transmission performance of some backbone networks for transmitting the access request of the server is less than that of the public network for transmitting the access request. For example, some backbone networks need to combine the networks between different backbone networks to send the access request to the server, which results in the transmission performance of transmitting the access request of the server being less than that of the public network for transmitting the access request. For another example, for a certain server, due to factors such as long backbone network paths and the need for forwarding devices, the transmission performance of transmitting the access request of the server is less than that of the public network for transmitting the access request. While some backbone networks have short paths or do not need to combine the networks between different backbone networks and can send the access request to the server. Therefore, the transmission performance of this backbone network for transmitting the access request of the server is greater than that of the public network for transmitting the access request.

[0056] The backbone network whose transmission performance is greater than that of the public network can be called the backbone network corresponding to the server. Exemplarily, as Figure 2 shown, it can be set that the above multiple backbone networks further include a fourth backbone network. Among them, the fourth backbone network can transmit the access request (such as the access request corresponding to the first acceleration instance) from the network service providing point 100 to the first server. Due to factors such as the long path of the fourth backbone network and the need for forwarding devices to forward, the transmission performance of the fourth backbone network for transmitting the access request from the network service providing point 100 to the first server is less than the transmission performance of transmitting the access request from the network service providing point 100 to the first server through the public network. Therefore, the fourth backbone network does not belong to the backbone network corresponding to the first server. In addition, the transmission performance of the first backbone network for transmitting the access request (such as the access request corresponding to the first acceleration instance) from the network service providing point 100 to the first server is greater than the transmission performance of transmitting the access request from the network service providing point 100 to the first server through the public network. Therefore, the first backbone network belongs to the backbone network corresponding to the first server.

[0057] Among them, the transmission performance of the access request of the transmission server refers to the transmission performance of transmitting the access request from the network service providing point 100 to the server. For example, the transmission performance of the first backbone network transmitting the access request from the network service providing point 100 to the first server is greater than that of the public network transmitting the access request from the network service providing point 100 to the first server, while the transmission performance of the second backbone network transmitting the access request from the network service providing point 100 to the first server is less than that of the public network transmitting the access request from the network service providing point 100 to the first server, and the transmission performance of the third backbone network transmitting the access request from the network service providing point 100 to the first server is less than that of the public network transmitting the access request from the network service providing point 100 to the first server.

[0058] In some embodiments, the backbone network corresponding to the server can be one.

[0059] In some embodiments, the backbone network corresponding to the server can be at least two. The at least two backbone networks are both connected to the server and connected to the network service providing point 100. That is to say, any backbone network among the at least two backbone networks can transmit the access request from the network service providing point 100 to the server, and the transmission performance is greater than that of the public network transmitting the access request from the network service providing point 100 to the server.

[0060] In some embodiments, the servers among the above-mentioned multiple servers can be deployed in the cloud. Taking the first server as an example, as Figure 3 shown, the first server may include a cloud management platform and infrastructure. Among them, cloud resources associated with the first acceleration instance are deployed in the infrastructure, and the cloud management platform is used to manage the cloud resources. Exemplarily, the cloud resources have an elastic public IP address (EIP). Among them, the infrastructure includes at least one cloud data center, each cloud data center is provided with multiple servers, and the cloud resources are deployed in at least one server of the infrastructure.

[0061] In this embodiment, the first backbone network is a backbone network deployed by the service provider of the first server, and the first server is directly connected to the first backbone network. That is to say, the first server can be directly connected to the first backbone network without passing through the public network. Among them, the cloud management platform in the first server is directly connected to the first backbone network.

[0062] In some embodiments, the servers among the above-mentioned multiple servers are connected to the corresponding backbone network through the public network. That is to say, in this embodiment, the server is not directly connected to its corresponding backbone network, but is connected to the public network, and the public network is connected to its corresponding backbone network to connect the server to the backbone network. Taking the second server as an example, as Figure 4As shown, there is a public network between the second server and the second backbone network. Exemplarily, the second server has an IP outside the cloud or a domain name.

[0063] Multiple acceleration instances are deployed in the network service providing point 100, such as a first acceleration instance, a second acceleration instance, and a third acceleration instance. Each of the multiple acceleration instances is associated with at least one of the multiple servers. For example, the first acceleration instance is associated with the first server, the second acceleration instance is associated with the second server, and the third acceleration instance is associated with the third server.

[0064] In some embodiments, as Figure 1 shown, different acceleration instances can belong to different users respectively. For example, the first acceleration instance belongs to user 1, the second acceleration instance belongs to user 2, and the third acceleration instance belongs to user 3. Among them, if an acceleration instance is purchased or leased by a certain user, then the acceleration instance belongs to that user. The acceleration instance can be purchased or leased by a user to access a certain server, and then that server is associated with the acceleration instance.

[0065] As Figure 5 shown, the network service providing point 100 can include a management module. The management module can instruct the acceleration instances deployed in the network service providing point 100 to access the network. For example, it instructs the acceleration instances to access the backbone network or the public network. Among them, the acceleration instances deployed in the network service providing point 100 can access the backbone network through the connection between the network service providing point 100 and the backbone network. The acceleration instances deployed in the network service providing point 100 can access the public network through the connection between the network service providing point 100 and the public network. The management module can also instruct the acceleration instances deployed in the network service providing point 100 to detect the corresponding backbone network.

[0066] The above examples introduce a system provided by the embodiments of the present application. Next, in combination with this system, the process of the network communication method provided by the embodiments of the present application will be described. Among them, this method can be executed by the network service providing point 100. Exemplarily, this method can be executed by the management module in the network service providing point 100. As Figure 6 shown, this method can include the following steps.

[0067] First, in step 601, instruct each of the multiple acceleration instances to access the server associated with the acceleration instance through the backbone network corresponding to the acceleration instance. Among them, the backbone network corresponding to the acceleration instance is the backbone network corresponding to the server associated with the acceleration instance among the multiple backbone networks.

[0068] For example, if the first acceleration instance is associated with the first server, the first backbone network corresponding to the first server is also the backbone network corresponding to the first acceleration instance. The network service providing point 100 may instruct the first acceleration instance to access the first server through the first backbone network.

[0069] Similarly, the second acceleration instance corresponds to the second backbone network, and the network service providing point 100 may instruct the second acceleration instance to access the second server through the second backbone network. The third acceleration instance corresponds to the third backbone network, and the network service providing point 100 may instruct the third acceleration instance to access the third server through the third backbone network.

[0070] Among them, an acceleration instance accessing a server through a backbone network means that the acceleration instance sends an access request to the server through the backbone network. For example, if the first acceleration instance accesses the first server through the first backbone network, the first acceleration instance sends an access request to the first server through the first backbone network.

[0071] In some embodiments, as introduced above, the backbone network corresponding to the server may be at least two backbone networks, and the acceleration instances associated with the server may access the at least two backbone networks simultaneously.

[0072] In some embodiments, as introduced above, the first server includes a cloud management platform and an infrastructure, and cloud resources associated with the first acceleration instance are deployed in the infrastructure. The cloud management platform is used to manage the cloud resources; among them, the infrastructure includes at least one cloud data center, each cloud data center is provided with multiple servers, and the cloud resources are deployed in at least one server of the infrastructure. In step 601, it may be instructed that the first acceleration instance sends an access request to the cloud management platform through the first backbone network, and the cloud management platform is used to forward the access request to the cloud resources so that the access request accesses the cloud resources.

[0073] And, in step 602, it is instructed that each of the multiple acceleration instances probes the backbone network corresponding to the acceleration instance.

[0074] Each acceleration instance probing the backbone network corresponding to the acceleration instance realizes the probing at the acceleration instance level, and can detect whether the backbone network corresponding to each acceleration instance fails.

[0075] In some embodiments, the server associated with the acceleration instance corresponds to at least two backbone networks, and the acceleration instance may access the at least two backbone networks simultaneously. It may be instructed that the acceleration instance probes each of the at least two backbone networks.

[0076] In some embodiments, the acceleration instance's detection of the backbone network corresponding to the acceleration instance may be an ICMP detection. In some embodiments, the acceleration instance's detection of the backbone network may be to detect the transmission delay of the backbone network. Exemplarily, in each detection, the acceleration instance may send a detection message to the server through the backbone network corresponding to the acceleration instance. After receiving the detection message, the server may feedback a response message to the acceleration instance through the backbone network. The acceleration instance may calculate the time difference between the sending time of the detection message it sent and the receiving time of the response message it received to obtain the transmission delay of the backbone network.

[0077] In some embodiments, the acceleration instance may detect the backbone network corresponding to the acceleration instance according to the detection strategy of the acceleration instance. Among them, the detection strategy includes the time interval between two adjacent detections, that is to say, the detection strategy determines how often a detection is performed.

[0078] In an example of this embodiment, the detection strategies of different acceleration instances among the multiple acceleration instances are independent of each other. That is to say, the detection strategy of an acceleration instance and the detection strategy of another acceleration instance are independent of each other and have no influence on each other.

[0079] As introduced above, the acceleration instance belongs to the user, and the user to which the acceleration instance belongs can configure the detection strategy for the acceleration instance. That is to say, the detection strategy for the acceleration instance to detect the backbone network corresponding to the acceleration instance is configured by the user to which the acceleration instance belongs. For example, as introduced above, the first acceleration instance belongs to user 1, and user 1 can configure a detection strategy for the first acceleration instance. The network service provider point 100 can receive and record the detection strategy configured by user 1 for the first acceleration instance, and instruct the first acceleration instance to detect the first backbone network according to the detection strategy.

[0080] Thus, the user can configure the corresponding detection strategy according to its service needs, etc., improving the fit between the detection strategy and the service.

[0081] Next, in step 603, when the detection result of the first acceleration instance among the multiple acceleration instances detecting the first backbone network indicates that the first backbone network fails, instruct the first acceleration instance to access the first server through the public network, so that the first acceleration instance sends an access request to the first server through the public network; wherein, the first server is the server associated with the first acceleration instance among the multiple servers, and the first backbone network is the backbone network corresponding to the first server among the multiple backbone networks.

[0082] In some embodiments, the detection by the first acceleration instance of the first backbone network may be to detect the transmission delay of the first backbone network. If the transmission delay of the first backbone network is greater than the delay threshold, the detection result of this detection indicates that the first backbone network has a fault. If the transmission delay of the first backbone network is less than or equal to the delay threshold, the detection result of this detection indicates that the first backbone network has no fault. Among them, the delay threshold may be preset, such as 1 millisecond, 2 milliseconds or 3 milliseconds, etc. Exemplarily, the delay threshold can be configured by User 1. Exemplarily, the delay threshold may be included in the detection strategy of the first acceleration instance.

[0083] In some embodiments, the detection result of the first acceleration instance's detection of the first backbone network may be the detection result of one detection. If the detection result of this detection indicates that the first backbone network has a fault, it can be determined that the first backbone network has a fault.

[0084] In some embodiments, the detection result of the first acceleration instance's detection of the first backbone network may be the detection results of multiple consecutive detections. If the detection results of these multiple consecutive detections all indicate that the first backbone network has a fault, and the number of detections in these multiple consecutive detections is greater than the fault threshold, it can be determined that the first backbone network has a fault. Otherwise, it can be determined that the first backbone network has no fault. For example, it can be set that the fault threshold is N, and N is an integer greater than or equal to 1. If the detection results of N + 1 consecutive detections or more all indicate that the first backbone network has a fault, then it can be determined that the first backbone network has a fault. Otherwise, it can be determined that the first backbone network has no fault.

[0085] The fault threshold may be preset. Exemplarily, N may be 3, 5 or 7, etc. Exemplarily, the fault threshold can be configured by User 1. Exemplarily, the fault threshold may be included in the detection strategy of the first acceleration instance.

[0086] When the first backbone network has a fault, the first acceleration instance accesses the public network and sends an access request to the first server through the public network, thereby ensuring the access performance of the first server.

[0087] Among them, as introduced above, the backbone network corresponding to the acceleration instance refers to the backbone network with a transmission performance greater than that of the public network. The fault of the backbone network corresponding to the acceleration instance means the fault of the backbone network with a transmission performance greater than that of the public network. The transmission performance of other backbone networks transmitting the access request of this acceleration instance is less than the transmission performance of the public network transmitting the access request of this acceleration instance. Therefore, when this acceleration instance sends an access request to the server through the public network, the access performance of this server can be ensured. Among them, other backbone networks refer to backbone networks other than the backbone network corresponding to this acceleration instance.

[0088] In some embodiments, as introduced above, the server corresponding to the acceleration instance is connected to at least two backbone networks, and the acceleration instance probes each of the at least two backbone networks. When all of the at least two backbone networks fail, the acceleration instance is instructed to access the public network. If some of the at least two backbone networks fail and the other part does not fail, the acceleration instance is instructed to continue accessing the non-failed backbone network.

[0089] In some embodiments, instructing the first acceleration instance to access the first server through the public network may include: instructing the first acceleration instance to add the address of the network service providing point 100 in the public network to the data packet whose destination is the first server, so that the first server sends the response packet of the data packet to the network service providing point 100 through this address. Wherein, when the first server receives the data packet, it can obtain the address of the network service providing point 100 in the public network. The first server sends the response packet of the data packet to the network service providing point 100 through the address of the network service providing point 100 in the public network. Thus, the network service providing point 100 can deliver the response packet to the first acceleration instance, and the first acceleration instance outputs the response packet to the user. Wherein, the data packet belongs to service traffic, and the response packet of the data packet belongs to the return traffic of the service traffic. Through this embodiment, the same-source and same-traceability of traffic can be realized.

[0090] In an example of this embodiment, when accessing the server through the public network, the packet is no longer encapsulated with VXLAN, and the address of the network service providing point 100 in the public network is used to replace the address of the user (or the user's client) in the public network, which can ensure the same-source and same-traceability of traffic.

[0091] In an example of this embodiment, the address of the network service providing point 100 in the public network may be the public IP address of the network service providing point 100. The public IP address of the network service providing point 100 may be the local IP segment of the network service providing point 100.

[0092] In an example of this example, the network service providing point 100 corresponds to two computer rooms, and the relevant devices in the two computer rooms can be used as the network service providing point 100 at different time periods to improve the disaster tolerance and reliability of the network service providing point 100. Wherein, the two computer rooms correspond to different local IP segments. Different local IP segments are different subnet segments. When the relevant devices in one of the computer rooms are used as the network service providing point 100, the local IP segment corresponding to this computer room is used as the public IP address of the network service providing point 100. That is to say, which computer room's relevant devices are used as the network service providing point 100, then the local IP segment corresponding to this computer room is used as the public IP address of the network service providing point 100.

[0093] Through the network communication method provided by the embodiments of the present application, only the acceleration instances corresponding to the faulty backbone network can be connected to the public network, while the acceleration instances corresponding to the non-faulty backbone network are still connected to the backbone network without the need to switch to the public network.

[0094] For example, it can be set that the multiple acceleration instances introduced above further include a second acceleration instance. The second acceleration instance is associated with a second server among the multiple servers, and the second server corresponds to a second backbone network among the multiple backbone networks. Step 603 (that is, when the detection result of the first acceleration instance among the multiple acceleration instances detecting the first backbone network indicates that the first backbone network is faulty, instructing the first acceleration instance to access the first server through the public network) includes: when the detection result of the first acceleration instance detecting the first backbone network indicates that the first backbone network is faulty, and the detection result of the second acceleration instance detecting the second backbone network indicates that the second backbone network is not faulty, instructing the first acceleration instance to access the first server through the public network, and instructing the second acceleration instance to continue to access the second server through the second backbone network.

[0095] The related implementation method of the second acceleration instance detecting the backbone network can be implemented with reference to the introduction of step 602 above, and will not be elaborated here.

[0096] In this embodiment, when the detection result of the first acceleration instance detecting the first backbone network indicates that the first backbone network is faulty, while the detection result of the second acceleration instance detecting the second backbone network indicates that the second backbone network is not faulty, the first acceleration instance sends an access request to the first server through the public network, while the second acceleration instance still sends an access request to the second server through the second backbone network. That is to say, which network the second acceleration instance uses to send an access request to the second server is not affected by the network accessed by the first acceleration instance or other acceleration instances. Herein, the network accessed by the acceleration instance refers to the network used by the acceleration instance to send an access request to the associated server.

[0097] This embodiment realizes network conversion at the acceleration instance level and reduces the scope affected by a certain backbone network failure.

[0098] In some embodiments, after step 603, the network service providing point 100 can instruct the first acceleration instance to continue to detect the first backbone network; when the detection result of the first acceleration instance continuing to detect the first backbone network indicates that the fault of the first backbone network has been eliminated, the network service providing point 100 can instruct the first acceleration instance to access the first server through the first backbone network again.

[0099] Among them, the related implementation method of the detection can be implemented with reference to the introduction of step 602 above, and will not be elaborated here.

[0100] In some embodiments, when the detection result of a detection of the first backbone network by the first acceleration instance indicates that the fault of the first backbone network has been eliminated (for example, the detection result indicates that the transmission delay of the first backbone network is less than the delay threshold introduced above), it can be determined that the fault of the first backbone network has been eliminated.

[0101] In some embodiments, when the detection results of consecutive multiple detections of the first backbone network by the first acceleration instance all indicate that the fault of the first backbone network has been eliminated, and the number of detections in the consecutive multiple detections is greater than the elimination threshold, it can be determined that the fault of the first backbone network has been eliminated. Otherwise, it can be determined that the fault of the first backbone network has not been eliminated yet. For example, the elimination threshold can be set to M, where M is an integer greater than or equal to 1. If the detection results of consecutive M + 1 or more detections all indicate that the first backbone network fault has been eliminated, it can be determined that the first backbone network fault has been eliminated. Otherwise, it is determined that the first backbone network fault has not been eliminated yet.

[0102] The elimination threshold can be set in advance. Exemplarily, M can be 3, 5, or 7, etc. Exemplarily, the elimination threshold can be configured by User 1. Exemplarily, the elimination threshold can be included in the detection strategy of the first acceleration instance.

[0103] In this embodiment, when the fault of the first backbone network has been eliminated, the first acceleration instance can be instructed to access the first backbone network again to send an access request to the first server through the first backbone network, thereby ensuring the access performance of the service.

[0104] In summary, through the network communication method provided by the embodiments of the present application, when the backbone network corresponding to the server fails, the acceleration instance associated with the server can be timely instructed to access the server through the public network, ensuring the access performance and user experience of the server. Moreover, there is no need to transmit the access request of the server through the backbone network corresponding to other servers, thus not affecting the access performance and user experience of other servers.

[0105] In some embodiments, another network communication method is proposed. As Figure 7As shown, in this method, user 1 can purchase or rent a corresponding acceleration instance, such as the first acceleration instance. After user 1 purchases or rents the first acceleration instance, the detection switch of the first acceleration instance can be turned on, and the detection policy of the first acceleration instance can be configured. Exemplarily, the cloud management platform of the first server can provide a relevant interface for user 1 to purchase or rent the first acceleration instance, turn on the detection switch of the first acceleration instance, and configure the acceleration policy. Among them, when the detection switch of the first acceleration instance is turned on, the cloud management platform can provide a configuration interface for the detection policy. This configuration interface can include configuration boxes for the detection interval and thresholds, for user 1 to configure the detection interval (i.e., how often to perform a detection) and relevant thresholds (such as a fault threshold, a latency threshold, etc.).

[0106] After the detection policy of the first acceleration instance is configured, the first acceleration instance can be deployed. Among them, the cloud management platform of the first server can deploy the first acceleration instance at the network service providing point 100 to complete the deployment of the first acceleration instance.

[0107] The first acceleration instance can actively detect the health of the backbone network corresponding to the first server (i.e., the first backbone network) according to the detection policy to determine whether there is a fault.

[0108] When the detection result of detecting the first backbone network indicates that the first backbone network fails, the first acceleration instance accesses the public network to send subsequent access requests to the first server through the public network. Here, the subsequent access requests refer to the access requests initiated by user 1 after the first acceleration instance accesses the public network.

[0109] When the detection result of continuously detecting the first backbone network indicates that the fault of the first backbone network is eliminated, the first acceleration instance accesses the first backbone network again to send subsequent access requests to the first server through the first backbone network. Here, the subsequent access requests refer to the access requests initiated by user 1 after the first acceleration instance accesses the first backbone network again.

[0110] Based on the content described above, an embodiment of the present application provides a network communication device 800. The device 800 is configured at a network service providing point, and the system where the network service providing point is located further includes multiple backbone networks and multiple servers; among them, each backbone network in the multiple backbone networks corresponds to at least one server in the multiple servers; multiple acceleration instances are deployed at the network service providing point, and each acceleration instance in the multiple acceleration instances is associated with at least one server in the multiple servers. As Figure 8 shown, the device 800 includes:

[0111] The first indication module 810 is configured to indicate that each of the multiple acceleration instances accesses a server associated with the acceleration instance through a backbone network corresponding to the acceleration instance; wherein, the backbone network corresponding to the acceleration instance is the backbone network corresponding to the server associated with the acceleration instance among the multiple backbone networks.

[0112] The second indication module 820 is configured to indicate that each of the multiple acceleration instances probes the backbone network corresponding to the acceleration instance.

[0113] The third indication module 830 is configured to, when a detection result of a first acceleration instance among the multiple acceleration instances probing a first backbone network indicates that the first backbone network fails, indicate that the first acceleration instance accesses a first server through a public network, so that the first acceleration instance sends an access request to the first server through the public network; wherein, the first server is the server associated with the first acceleration instance among the multiple servers, and the first backbone network is the backbone network corresponding to the first server among the multiple backbone networks.

[0114] Among them, the first indication module 810, the second indication module 820, and the third indication module 830 can all be implemented by software or by hardware. Exemplarily, next, taking the first indication module 810 as an example, the implementation manner of the first indication module 810 will be introduced. Similarly, the implementation manners of the second indication module 820 and the third indication module 830 can refer to the implementation manner of the first indication module 810.

[0115] As an example of a software functional unit, the first indication module 810 may include code running on a computing instance. Wherein, the computing instance may include at least one of a physical host (computing device), a virtual machine, and a container. Further, the above computing instance may be one or more. For example, the first indication module 810 may include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers for running the code may be distributed in the same region or in different regions. Further, the multiple hosts / virtual machines / containers for running the code may be distributed in the same availability zone (AZ) or in different AZs, and each AZ includes one data center or multiple geographically proximate data centers. Usually, one region may include multiple AZs.

[0116] Similarly, multiple hosts / virtual machines / containers used to run the code can be distributed within the same virtual private cloud (VPC) or across multiple VPCs. Usually, one VPC is set up within one region. For cross-region communication between two VPCs within the same region and between VPCs in different regions, a communication gateway needs to be set up within each VPC, and the interconnection between VPCs is achieved through the communication gateway.

[0117] As an example of a hardware functional unit, the first indication module 810 may include at least one computing device, such as a server. Alternatively, the first indication module 810 may also be a device implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD). Among them, the above PLD may be implemented by a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0118] The multiple computing devices included in the first indication module 810 can be distributed within the same region or across different regions. The multiple computing devices included in the first indication module 810 can be distributed within the same availability zone (AZ) or across different AZs. Similarly, the multiple computing devices included in the first indication module 810 can be distributed within the same VPC or across multiple VPCs. Among them, the multiple computing devices can be any combination of computing devices such as servers, ASICs, PLDs, CPLDs, FPGAs, and GALs.

[0119] It should be noted that in other embodiments, the first indication module 810 may be used to execute Figure 6 any step in the method shown, the second indication module 820 may be used to execute Figure 6 any step in the method shown, and the third indication module 830 may be used to execute Figure 6 any step in the method shown. The steps to be implemented by the first indication module 810, the second indication module 820, and the third indication module 830 can be specified as needed, and all functions of the apparatus 800 are implemented by respectively implementing Figure 6 different steps in the method shown.

[0120] The present application also provides a computing device 900. As Figure 9 shown, the computing device 900 includes: a bus 902, a processor 904, a memory 906, and a communication interface 908. The processor 904, the memory 906, and the communication interface 908 communicate with each other via the bus 902. The computing device 900 can be a server or a terminal device. It should be understood that the present application does not limit the number of processors and memories in the computing device 900.

[0121] The bus 902 can be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity of representation, Figure 9 only one line is shown in the figure, but it does not mean that there is only one bus or one type of bus. The bus 902 can include a path for transmitting information between various components of the computing device 900 (for example, the memory 906, the processor 904, and the communication interface 908).

[0122] The processor 904 can include any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0123] The memory 906 can include a volatile memory, such as a random access memory (RAM). The memory 906 can also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0124] The memory 906 stores executable program codes, and the processor 904 executes the executable program codes to respectively implement the functions of the foregoing first indication module 810, second indication module 820, and third indication module 830, thereby implementing Figure 6 the method shown. That is, the memory 906 stores programs for executionFigure 6 Instructions of the method shown

[0125] The communication interface 908 uses a transceiver module such as, but not limited to, a network interface card or a transceiver to implement communication between the computing device 900 and other devices or communication networks.

[0126] The embodiments of the present application also provide a cluster of computing devices. The cluster of computing devices includes at least one computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. In some embodiments, the computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smart phone.

[0127] As Figure 10 shown, the cluster of computing devices includes at least one computing device 900. Instructions for executing the method shown can be stored in the same manner in the memory 906 of one or more computing devices 900 in the cluster of computing devices. Figure 6 Instructions of the method shown

[0128] In some possible implementation manners, instructions for executing parts of the method shown can also be stored separately in the memory 906 of one or more computing devices 900 in the cluster of computing devices. In other words, a combination of one or more computing devices 900 can jointly execute the instructions for executing the method shown. Figure 6 Instructions of the method shown Figure 6 Instructions of the method shown

[0129] It should be noted that memories 906 in different computing devices 900 in the cluster of computing devices can store different instructions, respectively for executing partial functions of the apparatus 800. That is, instructions stored in memories 906 in different computing devices 900 can implement the functions of one or more of the first indication module 810, the second indication module 820, and the third indication module 830.

[0130] In some possible implementation manners, one or more computing devices in the cluster of computing devices can be connected through a network. Among them, the network can be a wide area network or a local area network, etc. Figure 11 shows a possible implementation manner. As Figure 11 shown, the first computing device and the second computing device are connected through a network. Specifically, they are connected to the network through the communication interfaces in each computing device. In this type of possible implementation manners, instructions for executing the function of the first indication module 810 are stored in the memory 906 of the first computing device. At the same time, instructions for executing the functions of the second indication module 820 and the third indication module 830 are stored in the memory 906 of the second computing device.

[0131] It should be understood that Figure 11The functions of the first computing device shown can also be completed by multiple computing devices 900. Similarly, the functions of the second computing device can also be completed by multiple computing devices 900.

[0132] The embodiments of the present application also provide another computing device cluster. The connection relationship between the computing devices in the computing device cluster can be similarly referred to Figure 10 and Figure 11 the connection manner of the computing device cluster shown. The difference is that the memory 906 in one or more of the computing devices 900 in the computing device cluster may store the same instructions for executing Figure 6 the method shown.

[0133] In some possible implementation manners, the memory 906 in one or more of the computing devices 900 in the computing device cluster may also separately store partial instructions for executing Figure 6 the method shown. In other words, the combination of one or more computing devices 900 can jointly execute the instructions for executing Figure 6 the method shown.

[0134] The embodiments of the present application also provide a computer program product including instructions. The computer program product may be software or a program product including instructions that can run on a computing device or be stored in any available medium. When the computer program product runs on at least one computing device, it causes at least one computing device to execute Figure 6 the method shown.

[0135] The embodiments of the present application also provide a computer-readable storage medium. The computer-readable storage medium may be any available medium that a computing device can store or a host migration device such as a data center including one or more available media. The available medium may be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid-state drive), etc. The computer-readable storage medium includes instructions that instruct a computing device to execute Figure 6 the method shown.

[0136] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the protection scope of the technical solutions of the embodiments of the present application.

Claims

1. A network communication method, characterized in that, The method is applied to a network service providing point, and the system where the network service providing point is located further includes a plurality of backbone networks and a plurality of service servers; wherein, for any service server among the plurality of service servers, any backbone network among the plurality of backbone networks is reachable, and each backbone network among the plurality of backbone networks corresponds to at least one service server among the plurality of service servers; a plurality of acceleration instances are deployed at the network service providing point, and each acceleration instance among the plurality of acceleration instances is associated with at least one service server among the plurality of service servers; the method includes: Instructing each acceleration instance among the plurality of acceleration instances to access the service server associated with the acceleration instance through the backbone network corresponding to the acceleration instance; wherein, the backbone network corresponding to the acceleration instance is the backbone network among the plurality of backbone networks corresponding to the service server associated with the acceleration instance; Instructing each acceleration instance among the plurality of acceleration instances to detect the backbone network corresponding to the acceleration instance; When the detection result of the first acceleration instance among the plurality of acceleration instances detecting the first backbone network indicates that the first backbone network among the plurality of backbone networks fails and the second backbone network among the plurality of backbone networks does not fail, instructing the first acceleration instance to access the first service server through the public network, so that the first acceleration instance sends an access request to the first service server through the public network; wherein, the first service server is the service server among the plurality of service servers associated with the first acceleration instance, and the first backbone network is the backbone network among the plurality of backbone networks corresponding to the first service server.

2. The method according to claim 1, wherein The first acceleration instance belongs to a user; the instructing each acceleration instance among the plurality of acceleration instances to detect the backbone network corresponding to the acceleration instance includes: Receiving and recording the detection policy configured by the user for the first acceleration instance; Instructing the first acceleration instance to detect the first backbone network according to the detection policy.

3. The method according to claim 1, characterized in that, The plurality of acceleration instances further includes a second acceleration instance, the second acceleration instance is associated with a second service server among the plurality of service servers, and the second service server corresponds to a second backbone network among the plurality of backbone networks; The when the detection result of the first acceleration instance among the plurality of acceleration instances detecting the first backbone network indicates that the first backbone network fails, instructing the first acceleration instance to access the first service server through the public network includes: When the detection result of the first acceleration instance detecting the first backbone network indicates that the first backbone network fails and the detection result of the second acceleration instance detecting the second backbone network indicates that the second backbone network does not fail, instructing the second acceleration instance to continue to access the second service server through the second backbone network.

4. The method according to claim 1, wherein After the instructing the first acceleration instance to access the first service server through the public network, the method further includes: Instructing the first acceleration instance to continue to detect the first backbone network; When the detection result of the continuous detection of the first backbone network by the first acceleration instance indicates that the fault of the first backbone network has been eliminated, it is indicated that the first acceleration instance accesses the first server through the first backbone network again.

5. The method according to claim 1, characterized in that The indication that the first acceleration instance accesses the first server through the public network includes: Indicating that the first acceleration instance adds the address of the network service providing point in the public network to the data packet whose destination is the first server, so that the first server sends the response packet of the data packet to the network service providing point through the address.

6. The method according to claim 1, characterized in that, The first server includes a cloud management platform and an infrastructure, and cloud resources associated with the first acceleration instance are deployed in the infrastructure. The cloud management platform is used to manage the cloud resources; wherein, the infrastructure includes at least one cloud data center, each cloud data center is provided with a plurality of servers, and the cloud resources are deployed in at least one server of the infrastructure; The indication that each acceleration instance in the multiple acceleration instances accesses the server associated with the acceleration instance through the backbone network corresponding to the acceleration instance includes: Indicating that the first acceleration instance sends an access request to the cloud management platform through the first backbone network, and the cloud management platform is used to forward the access request to the cloud resources, so that the access request accesses the cloud resources.

7. A network communication device, characterized in that, The device is configured at a network service providing point. The system where the network service providing point is located further includes a plurality of backbone networks and a plurality of servers; wherein, for any server in the plurality of servers, any backbone network in the plurality of backbone networks is reachable, and each backbone network in the plurality of backbone networks corresponds to at least one server in the plurality of servers; a plurality of acceleration instances are deployed at the network service providing point, and each acceleration instance in the plurality of acceleration instances is associated with at least one server in the plurality of servers; the device includes: A first indication module, configured to indicate that each acceleration instance in the plurality of acceleration instances accesses the server associated with the acceleration instance through the backbone network corresponding to the acceleration instance; wherein, the backbone network corresponding to the acceleration instance is the backbone network corresponding to the server associated with the acceleration instance among the plurality of backbone networks; A second indication module, configured to indicate that each acceleration instance in the plurality of acceleration instances detects the backbone network corresponding to the acceleration instance; A third indication module, configured to, when the detection result of the detection of the first backbone network by the first acceleration instance among the plurality of acceleration instances indicates that the first backbone network among the plurality of backbone networks fails and the second backbone network among the plurality of backbone networks does not fail, indicate that the first acceleration instance accesses the first server through the public network, so that the first acceleration instance sends an access request to the first server through the public network; wherein, the first server is the server associated with the first acceleration instance among the plurality of servers, and the first backbone network is the backbone network corresponding to the first server among the plurality of backbone networks.

8. A cluster of computing devices, characterized in that, comprising at least one computing device, each computing device including a processor and a memory; the processor of the at least one computing device is configured to execute instructions stored in the memory of the at least one computing device, so that the computing device cluster executes the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, comprising computer program instructions which, when executed by a computing device cluster, cause the computing device cluster to execute the method according to any one of claims 1 to 6.

10. A computer program product comprising instructions, characterized in that, when the instructions are run by a computer device cluster, causing the computer device cluster to execute the method according to any one of claims 1 to 6.

Images