Traffic scheduling method and system, electronic equipment and storage medium

Abstract

The invention provides a traffic scheduling method and system, electronic equipment and a storage medium. The method comprises the following steps: acquiring VPC traffic and a current cluster traffic utilization rate based on a traffic scheduling request; if it is determined that the programmable chip switch cluster can bear the VPC traffic, performing routing configuration on the programmable chip switch cluster according to a routing table generated by VPC instance information, and performing routing configuration on a boundary switch according to a scheduled VPC and a traffic scheduling request; modifying the next-hop VIP address into the VIP address of the programmable chip switch cluster; sending the VPC flow to an IDC (Internet Data Center) through a programmable chip switch cluster and a boundary switch; the VPC network segment is issued to the boundary switch through the programmable chip switch cluster based on the received signaling, so that the boundary switch completes routing convergence; and the traffic generated by the IDC is sent to the host machine through the private line access switch, the boundary switch, the NFV server gateway cluster and the programmable chip switch cluster.

Description

Technical Field

[0001] The present invention relates to the field of computers, and more specifically, to a traffic scheduling method, system, electronic device and storage medium. Background Art

[0002] In recent years, more and more companies have chosen to build their own enterprise-level cloud environments as IT infrastructure for application deployment. As application services gradually migrate to the cloud, existing data center infrastructure, private clouds, and newly built enterprise-level public clouds will inevitably coexist. At the same time, due to actual needs such as supervision and data security, companies still need to continue to maintain the above hybrid cloud status for a long time. In order to meet the high throughput, low latency, and financial-grade high availability of LAN networks between various technology stacks, the current cloud network gateway architecture is gradually evolving from server gateways based on NFV technology to hardware technology based on programmable chip switches, such as Figure 1 As shown, specifically, in the cross-region and cross-stack network interoperability scenarios in the cloud environment, DC products based on NFV technology and offload products based on programmable switching chip switches should provide network support.

[0003] However, in the process of evolution from server gateways based on NFV technology to hardware technology based on programmable chip switches, the physical network architecture of the two products is quite different, and traffic scheduling involves changes in cloud routing and physical network routing, which adds a certain degree of difficulty to cloud network traffic scheduling. In addition, traffic packet loss is prone to occur during the change process, which has a certain impact on the continuity of cloud services.

[0004] Therefore, how to provide a traffic scheduling method for network interconnection channels across instances and stacks between clouds to achieve traffic scheduling between traffic network interconnection channels and avoid traffic packet loss during the traffic scheduling process to ensure business continuity on the cloud is a problem that urgently needs to be solved in this application. Summary of the invention

[0005] In view of this, the present invention provides a traffic scheduling method, system, electronic device and storage medium to achieve traffic scheduling of a network architecture with a network interconnection channel, and to avoid traffic packet loss during the traffic scheduling process, so as to ensure the continuity of cloud services.

[0006] In a first aspect, the present application provides a traffic scheduling method, which is applicable to a network architecture with a network interconnection channel, wherein the network architecture at least includes a virtual network, an NFV server gateway cluster, a programmable chip switch cluster, a physical network, an edge switch, a dedicated line access switch, and an interconnected data center, and the method includes:

[0007] Receive a traffic scheduling request, and based on the traffic scheduling request, obtain the VPC traffic of the scheduled VPC from the NFV server gateway cluster and obtain the current cluster traffic usage rate from the programmable chip switch cluster;

[0008] Determine whether the programmable chip switch cluster can bear the VPC traffic according to the VPC traffic and the current cluster traffic usage rate;

[0009] If the programmable chip switch cluster can bear the VPC traffic, performing routing configuration on the programmable chip switch cluster according to the routing table of the scheduled VPC generated according to the VPC instance information of the scheduled VPC;

[0010] Performing routing configuration on the border switch according to the traffic scheduling request of the scheduled VPC;

[0011] Modify the first traffic scheduling direction corresponding to the scheduled VPC and the next-hop VIP address of the host machine to the VIP address of the programmable chip switch cluster to complete the configuration synchronization of the host machine; the first traffic scheduling direction is the VPC->IDC traffic scheduling direction;

[0012] Sending the VPC traffic of the scheduled VPC in the host to the IDC through the programmable chip switch cluster and the edge switch to complete the VPC traffic scheduling in the first traffic scheduling direction;

[0013] Sending a signaling to the programmable chip switch cluster, so that the programmable chip switch cluster publishes the VPC network segment of the scheduled VPC to the edge switch based on the signaling, so that the edge switch completes routing convergence;

[0014] The traffic generated by the IDC is sent to the host machine through the dedicated access switch, the edge switch, the NFV server gateway cluster, and the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction; the second traffic scheduling direction is the IDC->VPC traffic scheduling direction.

[0015] Optionally, the VPC traffic includes the VPC traffic statistics of each gateway in the NFV server gateway cluster, and the current cluster traffic usage includes the current traffic usage of each switch in the programmable chip switch cluster;

[0016] The determining, according to the VPC traffic and the current cluster traffic usage rate, whether the programmable chip switch cluster can withstand the VPC traffic includes:

[0017] For each of the switches, determining a statistical value of the remaining capacity of the switch according to the current traffic usage rate of the switch;

[0018] According to the VPC traffic statistics of each gateway and the remaining capacity statistics of each switch, it is determined whether the programmable chip switch cluster can withstand the VPC traffic.

[0019] Optionally, performing routing configuration on the programmable chip switch cluster according to the routing table of the scheduled VPC generated according to the VPC instance information of the scheduled VPC includes:

[0020] Obtain VPC instance information of the scheduled VPC;

[0021] Acquire all routing details related to the scheduled VPC according to the VPC instance information of the scheduled VPC, and generate a routing table of the scheduled VPC according to all routing details related to the scheduled VPC;

[0022] The routing table of the scheduled VPC is synchronized to the programmable chip switch cluster, so that when the programmable chip switch cluster receives the routing table of the scheduled VPC, it performs an integrity check on the routing table of the scheduled VPC through each switch, and uses the routing table of the scheduled VPC to update its own stored routing table when the integrity check passes, so as to complete the routing configuration of the programmable chip switch.

[0023] Optionally, the performing routing configuration on the edge switch according to the scheduled VPC and the traffic scheduling request includes:

[0024] Obtain the VPC network segment of the scheduled VPC;

[0025] Generate routing configuration information of the border switch according to the VPC network segment and the border switch model and routing type indicated in the traffic scheduling request;

[0026] The routing configuration information of the border switch is written into the border switch to complete the routing configuration of the border switch.

[0027] Optionally, sending the traffic generated by the IDC to the host machine through the dedicated line access switch, the edge switch, the NFV server gateway cluster, and the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction includes:

[0028] The traffic generated by the IDC is sent to the dedicated line access switch and the edge switch in a balanced manner through a hash rule;

[0029] The traffic received by the dedicated access switch is forwarded to the host machine through the NFV server gateway cluster, and the traffic received by the edge switch is forwarded to the host machine through the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction.

[0030] Optionally, after the border switch completes routing convergence, the method further includes:

[0031] Stop the session between the dedicated access switch and the IDC, or stop the session between the NFV server gateway cluster and the IDC, send the traffic generated by the IDC to the border switch, and forward the traffic received by the border switch to the host machine through the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction.

[0032] Optionally, the method further includes:

[0033] After completing the VPC traffic scheduling in the first traffic scheduling direction and the traffic scheduling in the second traffic scheduling direction, the dedicated line access switch collects the channel traffic of the scheduled VPC;

[0034] According to the traffic monitoring, all pre-stored routing details related to the scheduled VPC under the NFV server gateway cluster branch are deleted.

[0035] A second aspect of the present application provides a traffic scheduling system, which is applicable to a network architecture with a network interconnection channel, wherein the network architecture at least includes a virtual network, an NFV server gateway cluster, a programmable chip switch cluster, a physical network, an edge switch, a dedicated line access switch, and an interconnected data center, and the system includes:

[0036] An acquisition unit, configured to receive a traffic scheduling request, and based on the traffic scheduling request, acquire the VPC traffic of the scheduled VPC from the NFV server gateway cluster and acquire the current cluster traffic usage rate from the programmable chip switch cluster;

[0037] A determination unit, configured to determine whether the programmable chip switch cluster can bear the VPC traffic according to the VPC traffic and the current cluster traffic usage rate;

[0038] A first routing configuration unit, configured to perform routing configuration on the programmable chip switch cluster according to the routing table of the scheduled VPC generated according to the VPC instance information of the scheduled VPC if the programmable chip switch cluster can bear the VPC traffic;

[0039] A second routing configuration unit, configured to perform routing configuration on the edge switch according to the traffic scheduling request of the scheduled VPC;

[0040] A modification unit, used to modify the first traffic scheduling direction corresponding to the scheduled VPC and the next-hop VIP address of the host machine to the VIP address of the programmable chip switch cluster, so as to complete the configuration synchronization of the host machine; the first traffic scheduling direction is the VPC->IDC traffic scheduling direction;

[0041] A first traffic scheduling unit, configured to send the VPC traffic of the scheduled VPC in the host to the IDC through the programmable chip switch cluster and the edge switch when the host completes configuration synchronization, so as to complete the VPC traffic scheduling in the first traffic scheduling direction;

[0042] A signaling sending unit, configured to send a signaling to the programmable chip switch cluster, so that the programmable chip switch cluster publishes the VPC network segment of the scheduled VPC to the edge switch based on the signaling, so that the edge switch completes routing convergence;

[0043] The second traffic scheduling unit is used to send the traffic generated by the IDC to the host machine through the dedicated access switch, the edge switch, the NFV server gateway cluster, and the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction; the second traffic scheduling direction is the IDC->VPC traffic scheduling direction.

[0044] The third aspect of the present application provides an electronic device, comprising: a processor and a memory, wherein the processor and the memory are connected via a communication bus; wherein the processor is used to call and execute a program stored in the memory; and the memory is used to store a program, wherein the program is used to implement the traffic scheduling method provided in the first aspect of the present application.

[0045] The fourth aspect of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored, and the computer-executable instructions are used to execute the traffic scheduling method provided in the first aspect of the present application.

[0046] The embodiments of the present application provide a traffic scheduling method, system, electronic device and storage medium for a network interconnection channel. When the cloud platform receives a traffic scheduling request, it can determine whether the programmable chip switch cluster can bear the VPC traffic of the scheduled VPC according to the VPC traffic of the scheduled VPC and the current cluster traffic usage rate of the programmable chip switch cluster; if it can bear it, the routing table of the scheduled VPC can be further used to perform routing configuration on the programmable chip switch cluster, and the routing configuration information of the border switch can be used to perform routing configuration on the border switch, so that when it is determined that the programmable chip switch cluster and the border switch have completed the corresponding routing configuration, it can be determined that the programmable chip switch cluster and the border switch currently have the ability to correctly forward the VPC traffic of the scheduled VPC, and then the first traffic scheduling direction corresponding to the scheduled VPC and the next-hop VIP address of the host machine are modified to the VIP address of the programmable chip switch cluster, and the configuration synchronization of the host machine is completed. At this time, it can be determined that the cluster and programmable chip switch cluster corresponding to the original next-hop VIP address The switch clusters all have all the routing details of the scheduled VPC and can correctly forward the VPC traffic of the scheduled VPC. Therefore, no packet loss will occur during the traffic forwarding process in the first traffic scheduling direction, ensuring the continuity of cloud services. After completing the VPC traffic scheduling in the first traffic scheduling direction, the VPC network segment of the scheduled VPC can be published to the edge switch, so that the edge switch completes the routing convergence, and the switch that indicates the forwarding of the traffic generated by the core network access router and the WAN dedicated line switch (IDC) in the second traffic scheduling direction is transformed from the dedicated line access switch and the NFV server gateway cluster to the dedicated line access switch + edge switch and the NFV server gateway cluster + programmable chip switch cluster. Finally, the traffic generated by the IDC received by the dedicated line access switch and the edge switch is forwarded to the host corresponding to the scheduled VPC through the NFV server gateway cluster and the programmable chip switch cluster, which can effectively avoid the problem of packet loss in the traffic scheduling process in the second traffic scheduling direction, ensuring the continuity of cloud services. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0048] Figure 1 An example diagram of a network interconnection architecture provided by the prior art;

[0049] Figure 2A diagram showing an example of a network architecture provided in an embodiment of the present application;

[0050] Figure 3 A flow chart of a traffic scheduling method provided in an embodiment of the present application;

[0051] Figure 4 An example diagram of a traffic scheduling method provided in an embodiment of the present application;

[0052] Figure 5 A schematic diagram of the structure of a traffic scheduling system provided in an embodiment of the present application;

[0053] Figure 6 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] In this application, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0056] In order to better understand this application, the technical terms involved in this application are explained below:

[0057] VPC: Virtual Private Cloud, isolated cloud network space.

[0058] DC: Direct Connect, a cloud service that provides fast and reliable connections between user data centers and cloud resources.

[0059] IDC: Internet Data Center, a data center with complete equipment (high-speed Internet access, high-performance local area network, safe and reliable computer room environment), professional management, and complete application service platform.

[0060] Overlay virtual network: A logical network on the cloud, which refers to one or more logical networks built on the physical network through network virtualization technology on the cloud.

[0061] Underlay physical network: refers to the physical network that carries the Overlay virtual network and is actually composed of physical devices such as switches, routers, and physical servers.

[0062] NFV: Network Function Virtualization, which uses network virtualization technology to implement the functions of each network node in the form of software.

[0063] VRF: Virtual Routing Forwarding is a technology that runs multiple routing tables on a routing switching device to achieve service isolation.

[0064] SDN: Software Defined Network, a network management method that supports dynamic programmable network configuration.

[0065] P4: Programming Protocol-Independent Packet Processors, a high-level programming language specifically used to configure the data plane forwarding and processing logic of the switch.

[0066] VIP address, a virtual IP address technology under the virtual network architecture, usually refers to the service address of the cloud gateway cluster. It is introduced into the physical network through a dynamic routing protocol and implements load balancing and fault self-healing of the gateway cluster through equal-cost routing.

[0067] From the above background technology, it can be seen that in the prior art, in the cross-regional and cross-stack network interoperability scenarios in the cloud environment, DC products based on NFV technology and offload products based on programmable switching chip switches should provide network support.

[0068] Specifically, in the existing technology, the network interconnection between multiple cloud technology stacks relies on the dedicated line access of the NFV server gateway cluster based on NFV technology to carry the data plane traffic, or the use of cloud network products based on programmable chip switch clusters to carry data plane traffic.

[0069] However, the two have significant differences in the routing design of the underlying network and the protocol level before and after cluster node forwarding. The basic principle of the NFV server gateway cluster based on NFV technology to achieve dedicated line traffic isolation is the tunnel protocol running between the gateway cluster and the dedicated line access switch and the physical port isolation of the dedicated line access switch. Due to the size limit of the table items of traditional switches, only the corresponding VPC network segment routes can be written on the dedicated line access switch, and the detailed routes on the cloud are forwarded on the NFV server gateway cluster at the other end of the tunnel.

[0070] When implementing traffic isolation based on a programmable chip switch cluster, the large routing table of the programmable chip switch cluster can be used to complete network isolation and detailed routing on the cloud on the switches in the programmable chip switch cluster. It is only necessary to deploy a border switch at the cloud edge as a physical network forwarding channel.

[0071] It can be seen that in traditional network cutover technology, due to the characteristic that routes with the same IDC address in the cloud routing table can only specify a unique next-hop VIP address, traffic scheduling in the VPC->IDC traffic scheduling direction can only be achieved by deleting the old next-hop VIP address and adding a new next-hop VIP address. Moreover, at the cloud boundary, the traffic scheduling in the IDC->VPC traffic scheduling direction involves the deletion and delivery of routing configurations in the scheduling process between the dedicated line access switch and the boundary switch. Traffic scheduling in both directions cannot be started and ended at the same time on the cloud platform system and the SDN system. In addition, packet loss will occur in the event that the configuration is not completed or the route has not converged, and business continuity cannot be guaranteed. Among them, the existing VPC->IDC traffic scheduling direction refers to the traffic scheduling direction of Overlay virtual network-IDC, that is, the VPC traffic on the corresponding host machine of the Overlay virtual network is sent to the Internet Data Center (IDC) through the NFV server gateway cluster-Underlay physical network-dedicated line access switch; IDC->VPC traffic scheduling direction refers to the traffic scheduling direction of IDC-Overlay, that is, the traffic generated by IDC is sent to the corresponding host machine in the Overlay virtual network through the dedicated line access switch-Underlay physical network-NFV server gateway cluster.

[0072] Therefore, the present application provides a traffic scheduling method, system, electronic device, and storage medium. When the cloud platform receives a traffic scheduling request, it can determine whether the programmable chip switch cluster can withstand the VPC traffic of the scheduled VPC based on the VPC traffic of the scheduled VPC and the current cluster traffic usage rate of the programmable chip switch cluster; if it can withstand it, the routing table of the scheduled VPC can be further used to perform routing configuration on the programmable chip switch cluster, and the routing configuration information of the border switch can be used to perform routing configuration on the border switch, so that when it is determined that the programmable chip switch cluster and the border switch have completed the corresponding routing configuration, it can be determined that the programmable chip switch cluster and the border switch currently have the ability to correctly forward the VPC traffic of the scheduled VPC, and then the first traffic scheduling direction corresponding to the scheduled VPC and the next-hop VIP address of the host machine are modified to the VIP address of the programmable chip switch cluster, and the configuration synchronization of the host machine is completed. At this time, the cluster corresponding to the original next-hop VIP address can be determined. Both the VPC and the programmable chip switch cluster have all the routing details of the scheduled VPC, and can correctly forward the VPC traffic of the scheduled VPC. Therefore, no data packet loss will occur during the traffic forwarding process in the first traffic scheduling direction. After completing the VPC traffic scheduling in the first traffic scheduling direction, the VPC network segment of the scheduled VPC can be published to the border switch, so that the border switch completes the routing convergence, and the switch that indicates the forwarding of the traffic generated by the core network access router and the WAN dedicated line switch (IDC) in the second traffic scheduling direction is transformed from the dedicated line access switch and the NFV server gateway cluster to the dedicated line access switch + border switch and the NFV server gateway cluster + programmable chip switch cluster. Finally, the traffic generated by the IDC received by the dedicated line access switch and the border switch is forwarded to the host corresponding to the scheduled VPC through the NFV server gateway cluster and the programmable chip switch cluster, which can effectively avoid the problem of data packet loss during the traffic scheduling process in the second traffic scheduling direction.

[0073] See also Figure 2 , showing a schematic diagram of a network architecture with a network interconnection channel provided by an embodiment of the present application, wherein the network architecture includes a Qverlay virtual network, a programmable chip switch cluster, an NFV server gateway cluster, an Underlay physical network, a border switch, a dedicated line access switch, and an interconnected data center; the interconnected data center includes a core network access router and a wide area network dedicated line switch.

[0074] In actual application, the programmable switch chip cluster and the NFV server gateway cluster can publish the VIP addresses of their respective clusters to the Underlay physical network through routing technologies such as BGP or OSPF. When the Overlay virtual network is accessed across the stack, the next-hop VIP address of the VPC indicated in the private network tunnel of the Overlay virtual network can be queried from the general routing table; the next-hop VIP address is pre-specified by the cloud platform controller, and the cloud platform synchronizes the specified next-hop VIP address to the routing table of each host machine. The general routing table includes the routing configuration information of each VPC in the Overlay virtual network; the VPC is deployed on the host machine in the Overlay virtual network.

[0075] The Underlay physical network can distribute traffic (the traffic of the host indicated by the outer source IP address in the outer VIP address of the tunnel) to each gateway server / switch in the cluster (the cluster indicated by the outer destination address (VIP address) in the outer VIP address of the tunnel) according to the outer VIP address of the tunnel; after receiving the traffic, the gateway in the cluster forwards the traffic according to the private network ID and the inner destination IP; among them, the NFV server gateway cluster will re-encapsulate the inner message into a VxLAN message and send the VxLAN message to the dedicated line access switch. The programmable chip switch will extract the inner message and forward it directly, and at the same time route and publish the Overlay virtual network address through BGP; the Underlay physical network will route the Overlay virtual network address to the border switch; the border switch and the dedicated line access switch will publish the Overlay virtual network address through static or BGP, that is, send the traffic to the core network access router or WAN dedicated line switch through static or BGP.

[0076] It should be noted that the private network ID and destination address are obtained by the gateway during the parsing of the tunnel data packet. The tunnel protocol ID is the private network ID, and the destination IP of the tunnel inner layer protocol is the inner layer destination IP.

[0077] Based on the network architecture provided by the above-mentioned embodiment of the present application, accordingly, the embodiment of the present application provides a traffic scheduling method, which can realize the traffic scheduling of the network architecture across clouds, instances and stacks, that is, realize the traffic scheduling of the network architecture with interconnection channels, such as Figure 3 As shown, the traffic scheduling method is applied to a cloud platform, and the method specifically includes the following steps:

[0078] S301: Receive a traffic scheduling request, and based on the traffic scheduling request, obtain the VPC traffic of the scheduled VPC from the NFV server gateway cluster and obtain the current cluster traffic usage rate from the programmable chip switch cluster.

[0079] In the process of specifically executing step S301, the cloud platform can detect in real time whether there is a traffic scheduling operation by the operation and maintenance personnel. When the traffic scheduling operation by the operation and maintenance personnel is detected, the cloud platform can receive a traffic scheduling request corresponding to the traffic scheduling operation, wherein the traffic scheduling request at least indicates the scheduled VPC, the model of the border switch, the routing type, etc.; and send a VPC traffic acquisition request to the NFV server gateway cluster, so that when the NFV server gateway cluster receives the VPC traffic acquisition request, it will feedback to the cloud platform the VPC traffic of the scheduled VPC indicated by the VPC traffic acquisition request.

[0080] After receiving the VPC traffic of the scheduled VPC fed back by the NFV server gateway cluster, the cloud platform can send a traffic usage request to the programmable chip switch cluster, so that when the programmable chip switch cluster receives the traffic usage request, it feeds back the current cluster traffic usage to the cloud platform, so that the cloud platform can determine whether the programmable chip switch cluster can bear the VPC traffic of the scheduled VPC based on the VPC traffic fed back by the NFV server gateway cluster and the current cluster traffic usage fed back by the programmable chip switch cluster.

[0081] It should be noted that the VPC traffic of the scheduled VPC includes the VPC traffic statistics of each gateway in the NFV server gateway cluster; the current cluster traffic usage of the programmable chip switch cluster includes the current traffic usage of each switch in the programmable chip switch cluster.

[0082] S302: Determine whether the programmable chip switch cluster can bear the VPC traffic according to the VPC traffic and the current cluster traffic usage rate. If the programmable chip switch cluster can bear the VPC traffic of the scheduled VPC, execute step S303.

[0083] During the specific execution of step S302, after receiving the VPC traffic of the scheduled VPC and the current cluster traffic usage of the programmable chip switch cluster, the cloud platform can determine whether the programmable chip switch cluster can withstand the VPC traffic based on the VPC traffic statistics of each gateway in the VPC traffic of the scheduled VPC and the current traffic usage of each switch in the current cluster traffic usage.

[0084] Optionally, the process of determining whether the programmable chip switch cluster can withstand VPC traffic based on VPC traffic and current cluster traffic usage can be specifically as follows: for each switch, determine the remaining capacity statistics of the switch based on the current traffic usage of the switch; and judge whether the programmable chip switch cluster can withstand VPC traffic based on the VPC traffic statistics of each gateway and the remaining capacity statistics of each switch.

[0085] In the actual application process, for each gateway in the NFV server gateway cluster, the maximum VPC traffic statistics value of the gateway within a preset statistical period can be determined from the VPC traffic statistics value of the gateway, and the maximum VPC traffic statistics value of each gateway in the NFV server gateway cluster is accumulated to obtain a first accumulated value; for each switch in the programmable chip switch cluster, the total traffic usage rate of the switch can be subtracted from the current traffic usage rate of the switch to determine the remaining traffic usage rate of the switch, that is, to obtain the remaining capacity statistics value of the switch, and the maximum VPC traffic usage rate of the switch within the preset statistical period can be determined from the remaining capacity statistics value of the switch. The maximum remaining capacity statistics of the switch, accumulate the maximum remaining capacity statistics of each switch, and obtain a second accumulated value; determine whether the first accumulated value is less than the second accumulated value; if the first accumulated value is less than the second accumulated value, it can be determined that the programmable chip switch cluster can withstand the VPC traffic of the scheduled VPC; if the first accumulated value is not less than the second accumulated value, it can be determined that the programmable chip switch cluster cannot withstand the VPC traffic of the scheduled VPC, and the corresponding prompt information can be output at this time to prompt the operation and maintenance personnel that the programmable chip switch cluster cannot withstand the VPC traffic of the scheduled VPC, so that the operation and maintenance personnel can make corresponding operation and maintenance measures in time according to the prompt information. Among them, the specific process of judging whether the programmable chip switch cluster can withstand the VPC traffic based on the VPC traffic statistics of each gateway and the remaining capacity statistics of each switch can be as shown in the public

[0086] As shown in formula 1.

[0087]

[0088] Where M is the number of gateways in the NFV server gateway cluster, F i,t-T,t is the VPC traffic statistics value of the ith gateway between the current time t and the monitoring statistical period tT, that is, the VPC traffic statistics value of the ith gateway within the preset statistical period T; N is the number of switches in the programmable chip switch cluster, C j,t-T,t It is the remaining capacity statistical value of the j-th switch between the current time t and the monitoring statistical period tT, that is, the remaining flux statistical value of the j-th switch within the preset statistical period T.

[0089] S303: Generate a routing table of the scheduled VPC according to the VPC instance information of the scheduled VPC, and perform routing configuration on the programmable chip switch cluster according to the routing table of the scheduled VPC.

[0090] In the specific process of executing step S303, when the cloud platform determines that the programmable chip switch cluster can withstand the VPC traffic of the scheduled VPC, it can further obtain the VPC instance information of the scheduled VPC, and generate a routing table of the scheduled VPC based on the VPC instance information of the scheduled VPC, and finally use the generated routing table of the scheduled VPC to perform routing configuration for the programmable chip switch cluster.

[0091] Optionally, the process of performing routing configuration on the programmable chip switch cluster based on the routing table of the scheduled VPC generated according to the VPC instance information of the scheduled VPC can be specifically as follows: obtaining the VPC instance information of the scheduled VPC; obtaining all routing details related to the scheduled VPC according to the VPC instance information of the scheduled VPC, and generating a routing table of the scheduled VPC according to all routing details related to the scheduled VPC; synchronizing the routing table of the scheduled VPC to the programmable chip switch cluster, so that when the programmable chip switch cluster receives the routing table of the scheduled VPC, it performs an integrity check on the routing table of the scheduled VPC through each switch, and uses the routing table of the scheduled VPC to update its own stored routing table when the integrity check passes, so as to complete the routing configuration of the programmable chip switch.

[0092] In actual application, the cloud platform can obtain the VPC instance information of the scheduled VPC, and query all routing details related to the scheduled VPC from the total routing table pre-stored in the cloud platform routing management platform based on the VPC instance information, so as to generate the routing table of the scheduled VPC based on all the queried routing details related to the scheduled VPC; the routing table of the scheduled VPC is synchronized to the programmable chip switch cluster to trigger all switches in the programmable chip switch cluster to perform routing integrity check and update based on the received routing table of the scheduled VPC.

[0093] Specifically, for each switch in the programmable chip switch cluster, each switch has a corresponding routing table stored in advance; when the switch receives the routing table of the scheduled VPC synchronized by the cloud platform, it can determine whether the received routing table of the scheduled VPC is consistent with the routing table of the scheduled VPC synchronized by the cloud platform. If they are consistent, it can be considered that the integrity check of the received route of the scheduled VPC has passed, and then the routing table of the scheduled VPC can be used to update the routing table stored in itself to complete the routing configuration of the programmable chip switch cluster. At this time, the programmable chip switch cluster has written all the routing details of the scheduled VPC. In the subsequent process of traffic forwarding, the traffic passing through any switch in the programmable chip switch cluster can be correctly forwarded according to the configured routing details.

[0094] If the received routing table of the scheduled VPC is inconsistent with the routing table of the scheduled VPC synchronized with the cloud platform, it can be considered that the integrity check of the received routing of the scheduled VPC has failed. At this time, a corresponding prompt message can be output to prompt the operation and maintenance personnel that the routing table of the scheduled VPC received by the programmable chip switch cluster is incomplete at this time, so that the operation and maintenance personnel can take corresponding operation and maintenance measures in time.

[0095] It should be noted that after completing the corresponding routing configuration, the programmable chip switch cluster can feedback to the cloud platform the prompt information that the corresponding routing configuration is completed and the routing table of the scheduled VPC has passed the complete check, so that the cloud platform can promptly understand the routing configuration completion of the programmable chip switch cluster.

[0096] S304: Perform routing configuration on the border switch according to the scheduled VPC traffic scheduling request.

[0097] In the specific execution of step S304, when the cloud platform receives the prompt information of the completion of the routing configuration fed back by the programmable chip switch cluster, it can determine that the programmable chip switch has completed the routing configuration, and can further perform routing configuration on the border switch, so that in the subsequent process of traffic forwarding, the traffic passing through the border switch can be correctly forwarded. Specifically, the routing configuration information of the border switch can be generated according to the received scheduled VPC and traffic scheduling request, and the routing configuration information of the border switch can be written into the border switch, so that the border switch writes the relevant information of the scheduled VPC and the VPC traffic scheduling direction, optimizes the transmission path of the VPC traffic, and enables the VPC traffic passing through the border switch to be correctly forwarded.

[0098] Optionally, when the programmable chip switch cluster completes the routing configuration, the process of performing routing configuration on the border switch according to the scheduled VPC and the traffic scheduling request can be specifically as follows: when the programmable chip switch cluster completes the routing configuration, obtain the VPC network segment of the scheduled VPC; generate the routing configuration information of the border switch according to the VPC network segment and the border switch model and routing type indicated in the traffic scheduling request; write the routing configuration information of the border switch into the border switch to complete the routing configuration of the border switch.

[0099] In actual application, when the cloud platform determines that the programmable chip switch cluster has completed the routing configuration, it can further obtain the VPC network segment corresponding to the scheduled VPC, and generate the routing configuration information of the border switch according to the VPC network segment of the scheduled VPC and the border switch model and routing type indicated in the traffic scheduling request, so as to write the routing configuration information of the border switch into the border switch through the cloud platform's SDN controller using methods such as netconf to complete the routing configuration of the border switch. At this time, the border switch writes the relevant information about the scheduled VPC and the VPC traffic scheduling direction, so that the VPC traffic of the scheduled VPC passing through the border switch can be forwarded normally.

[0100] It should be noted that after the routing configuration information of the border switch is successfully written, the border switch can feedback a prompt message of the completion of the routing configuration to the cloud platform, so that when the cloud platform receives the prompt message feedback from the border switch, it can promptly understand that the border switch has completed the corresponding routing configuration.

[0101] S305: Modify the first traffic scheduling direction corresponding to the scheduled VPC and the next-hop VIP address of the host machine into the VIP address of the programmable chip switch cluster to complete the configuration synchronization of the host machine; the first traffic scheduling direction is the VPC->IDC traffic scheduling direction.

[0102] In the specific process of executing step S305, when the cloud platform determines that both the programmable chip switch cluster and the edge switch have completed the corresponding routing configuration, it can be determined that both the programmable chip switch cluster and the edge switch currently have the ability to correctly forward the VPC traffic of the scheduled VPC, and then the next-hop VIP address corresponding to the first traffic scheduling direction corresponding to the scheduled VPC stored on the routing management platform of the cloud platform is modified to the VIP address of the programmable chip switch cluster, and at the same time, the next-hop VIP address stored on the host corresponding to the scheduled VPC is also modified to the VIP address of the programmable chip switch cluster to complete the configuration synchronization of the host. At this time, the traffic in the first traffic scheduling direction of the scheduled VPC will pass through the programmable chip switch cluster, and the traffic in the second traffic scheduling direction of the scheduled VPC will pass through the cluster corresponding to the original next-hop VIP address. It can be seen that the cluster corresponding to the original next-hop VIP address and the programmable chip switch cluster both have all the routing details of the scheduled VPC, and can correctly forward the VPC traffic of the scheduled VPC based on the routing details of the scheduled VPC. Therefore, no data packet loss will occur during the traffic forwarding process in the first traffic scheduling direction.

[0103] It should be noted that the next-hop VIP address can be the VIP address of the NFV server gateway cluster; the first traffic scheduling direction is VPC->IDC traffic scheduling direction, that is, the first traffic scheduling direction indicates that the VPC traffic on the host corresponding to the scheduled VPC is sent to the Internet Data Center (IDC) through the NFV server gateway cluster-Underlay physical network-dedicated line access switch; the second traffic scheduling direction is IDC->VPC traffic scheduling direction, that is, the second traffic scheduling direction indicates that the traffic generated by the IDC is sent to the host corresponding to the scheduled VPC through the dedicated line access switch-Underlay physical network-NFV server gateway cluster; that is, after the configuration synchronization of the host is completed, the first traffic scheduling direction indicates that the traffic generated by the IDC is sent to the host corresponding to the scheduled VPC through the programmable chip switch-Underlay physical network-border switch The switch-IDC sends the VPC traffic on the host corresponding to the scheduled VPC to the Internet data center (IDC), and the second traffic scheduling direction still indicates that the traffic generated by the IDC is sent to the host corresponding to the scheduled VPC through the dedicated line access switch-Underlay physical network-NFV server gateway cluster. It can be seen that the traffic in the first traffic scheduling direction of the scheduled VPC will pass through the programmable chip switch cluster, and the traffic in the second traffic scheduling direction of the scheduled VPC will pass through the cluster corresponding to the original next-hop VIP address, that is, the NFV server gateway cluster and the programmable chip switch cluster both have all the routing details of the scheduled VPC, and can correctly forward the VPC traffic of the scheduled VPC. Therefore, no data packet loss will occur during the traffic forwarding process in the first traffic scheduling direction.

[0104] S306: Send the VPC traffic of the scheduled VPC in the host to the IDC through the programmable chip switch cluster and the edge switch that have completed the routing configuration.

[0105] In the specific execution process of step S306, after completing the configuration synchronization of the host machine, the cloud platform can send the VPC traffic of the scheduled VPC in the host machine to the IDC through the programmable chip switch cluster and the edge switch that have completed the routing configuration to complete the traffic scheduling in the first traffic scheduling direction.

[0106] S307: Send signaling to the programmable chip switch cluster, so that the programmable chip switch cluster publishes the VPC network segment of the scheduled VPC to the edge switch based on the signaling, so that the edge switch completes routing convergence.

[0107] In the specific process of executing step S307, when the cloud platform completes the VPC traffic scheduling in the first traffic scheduling direction, it can send corresponding signaling to the programmable chip switch according to the VPC network segment of the scheduled VPC, wherein the signaling carries the VPC network segment of the scheduled VPC; after receiving the signaling, the programmable chip switch cluster can publish the VPC network segment carried in the signaling to the edge switch through the Underlay physical network and EBGP, so that the edge switch completes the corresponding routing configuration based on the received VPC network segment to complete the corresponding routing convergence, and when the routing convergence is completed, feedback the prompt information of the completion of routing convergence to the cloud platform.

[0108] S308: Send the traffic generated by the IDC to the host machine through the dedicated line access switch, the edge switch, the NFV server gateway cluster, and the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction.

[0109] In the specific execution of step S308, when the cloud platform detects that the border switch receives the VPC network segment, that is, when it is determined that the border switch has completed routing convergence, it can be determined that the border switch that has completed routing convergence and the programmable chip switch that has completed routing configuration have the ability to accurately forward the traffic generated by the IDC. At this time, it is determined that the switch that indicates forwarding the traffic generated by the core network access router and the wide area network dedicated line switch (IDC) in the second traffic scheduling direction is changed from the dedicated line access switch and the NFV server gateway cluster to the dedicated line access switch + border switch and the NFV server gateway cluster + programmable chip switch cluster, and then the traffic generated by the IDC can be balanced on the dedicated line access switch and the border switch, and then the traffic received by the dedicated line access switch and the border switch is forwarded to the host corresponding to the scheduled VPC through the NFV server gateway cluster and the programmable chip switch cluster to complete the traffic scheduling in the second traffic direction.

[0110] Optionally, when it is detected that the border switch receives the VPC network segment, the traffic generated by the IDC is sent to the host machine through the dedicated access switch, the border switch, the NFV server gateway cluster, and the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction. The specific process can be: when it is detected that the border switch receives the VPC network segment, the traffic generated by the IDC is sent to the dedicated access switch and the border switch in a balanced manner through the hash rule; the traffic received by the dedicated access switch is forwarded to the host machine through the NFV server gateway cluster, and the traffic received by the border switch is forwarded to the host machine through the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction.

[0111] It should be noted that by changing the switch that indicates forwarding the traffic generated by the core network access router and the wide area network dedicated switch (IDC) in the second traffic scheduling direction from a dedicated access switch and an NFV server gateway cluster to a dedicated access switch + edge switch and an NFV server gateway cluster + a programmable chip switch cluster, so that the traffic received by the dedicated access switch is forwarded to the host machine through the NFV server gateway cluster and the traffic received by the edge switch is forwarded to the host machine through the programmable chip switch cluster, the problem of data packet loss in the traffic scheduling process in the second traffic scheduling direction can be effectively avoided.

[0112] It should also be noted that the above is merely a preferred method for completing traffic scheduling in the second traffic scheduling direction provided by an embodiment of the present application.

[0113] Furthermore, the embodiments of the present application can also provide another traffic scheduling method for completing the second traffic scheduling direction; specifically, it can be: when it is detected that the border switch receives the VPC network segment, that is, when it is determined that the border switch completes routing convergence based on the received VPC network segment, the session between the dedicated line access switch and the IDC is stopped, or the session between the NFV server gateway cluster and the IDC is stopped, the traffic generated by the IDC is sent to the border switch, and the traffic received by the border switch is forwarded to the host machine through the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction.

[0114] In actual application, when the cloud platform determines that the border switch completes routing convergence based on the received VPC network segment, the SDN controller can be used to stop the BFD session between the dedicated access switch and the cloud border core network router / WAN dedicated switch (IDC) (for dynamic routing) or stop the BFD session between the NFV server gateway cluster and the cloud border core network router / WAN dedicated switch (IDC) (for static routing). At this time, the switch indicating the forwarding of traffic generated by the core network access router and the WAN dedicated switch (IDC) in the second traffic scheduling direction can be changed from the dedicated access switch and the NFV server gateway cluster to the dedicated access switch + border switch and the NFV server gateway cluster + programmable chip switch cluster, and then to the border switch and the programmable chip switch cluster, so that all the traffic generated by the IDC can be sent to the border switch, and the traffic received by the border switch can be forwarded to the host through the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction.

[0115] It should be noted that since the programmable chip switch cluster and the edge switch are configured with the routing details of the scheduled VPC, when the BFD session between the dedicated access switch and the cloud edge core network router / WAN dedicated switch (IDC) is stopped through the SDN controller, or the BFD session between the NFV server gateway cluster and the cloud edge core network router / WAN dedicated switch (IDC) is stopped, the programmable chip switch cluster and the edge switch can correctly forward the traffic generated by the IDC to the host corresponding to the scheduled VPC based on the routing details of the scheduled VPC, and can also effectively avoid the problem of data packet loss during the traffic scheduling process in the second traffic scheduling direction.

[0116] Further, in the embodiment of the present application, after completing the VPC traffic scheduling in the first traffic scheduling direction and the traffic scheduling in the second traffic scheduling direction, the traffic monitoring of the scheduled VPC is collected from the dedicated line access switch; and all the route details related to the scheduled VPC under the NFV server gateway cluster branch stored in advance are deleted according to the traffic monitoring. Among them, the traffic monitoring includes the channel traffic of the scheduled VPC.

[0117] In actual application, after completing the VPC traffic scheduling in the first traffic scheduling direction and the traffic scheduling in the second traffic scheduling direction, the cloud platform can apply to the dedicated line access switch for the channel traffic of the scheduled VPC, so that the dedicated line access switch can feedback the channel traffic of the scheduled VPC to the cloud platform according to the application submitted by the cloud platform, so that the cloud platform can judge whether the VPC traffic of the scheduled VPC in the host has been completely migrated to the IDC according to the channel traffic of the scheduled VPC and the VPC traffic of the scheduled VPC fed back by the dedicated line access switch; if it is determined that the VPC traffic of the scheduled VPC has been completely migrated, all routing details related to the scheduled VPC under the NFV server gateway cluster branch pre-stored in the routing management platform can be deleted.

[0118] The embodiment of the present application provides a traffic scheduling method. When the cloud platform receives a traffic scheduling request, it can determine whether the programmable chip switch cluster can bear the VPC traffic of the scheduled VPC according to the VPC traffic of the scheduled VPC and the current cluster traffic usage rate of the programmable chip switch cluster; if it can bear it, the routing table of the scheduled VPC can be further used to configure the routing of the programmable chip switch cluster, and the routing configuration information of the border switch can be used to configure the routing of the border switch, so that when it is determined that the programmable chip switch cluster and the border switch have completed the corresponding routing configuration, it can be determined that the programmable chip switch cluster and the border switch currently have the ability to correctly forward the VPC traffic of the scheduled VPC, and then the first traffic scheduling direction corresponding to the scheduled VPC and the next-hop VIP address of the host machine are modified to the VIP address of the programmable chip switch cluster, and the configuration synchronization of the host machine is completed. At this time, the cluster and programmable chip corresponding to the original next-hop VIP address can be determined. The switch clusters all have all the routing details of the scheduled VPC and can correctly forward the VPC traffic of the scheduled VPC. Therefore, no packet loss will occur during the traffic forwarding process in the first traffic scheduling direction. After completing the VPC traffic scheduling in the first traffic scheduling direction, the VPC network segment of the scheduled VPC can be published to the edge switch, so that the edge switch completes the routing convergence, and the switch that indicates the forwarding of the traffic generated by the core network access router and the WAN dedicated line switch (IDC) in the second traffic scheduling direction is transformed from the dedicated line access switch and the NFV server gateway cluster to the dedicated line access switch + edge switch and the NFV server gateway cluster + programmable chip switch cluster. Finally, the traffic generated by the IDC received by the dedicated line access switch and the edge switch is forwarded to the host corresponding to the scheduled VPC through the NFV server gateway cluster and the programmable chip switch cluster, which can effectively avoid the problem of packet loss during the traffic scheduling process in the second traffic scheduling direction.

[0119] In order to better understand the contents provided in the embodiments of the present application, the following is explained by way of example. Figure 4 shown.

[0120] When the cloud platform detects that the operation and maintenance personnel start traffic scheduling for the scheduled VPC, it sends a VPC traffic acquisition request to the NFV server gateway cluster.

[0121] The NFV server gateway cluster feeds back the VPC traffic of the scheduled VPC based on the VPC traffic acquisition request.

[0122] After receiving the VPC traffic of the scheduled VPC fed back by the NFV server gateway cluster, the cloud platform can send a traffic usage rate request to the programmable chip switch cluster.

[0123] When the programmable chip switch cluster receives a traffic usage request, it feeds back the current cluster traffic usage to the cloud platform.

[0124] The cloud platform determines whether the programmable chip switch cluster can withstand the VPC traffic of the scheduled VPC based on the VPC traffic and the current cluster traffic usage.

[0125] If it is determined that the programmable chip switch cluster can bear the VPC traffic of the scheduled VPC, the cloud platform generates a routing table of the scheduled VPC based on the VPC instance information of the scheduled VPC.

[0126] The cloud platform synchronizes the routing table of the scheduled VPC to the programmable chip switch cluster.

[0127] The programmable chip switch cluster performs an integrity check on the routing table of the scheduled VPC, and updates it with the routing table of the scheduled VPC when the integrity check passes, to complete the corresponding routing configuration.

[0128] The programmable chip switch feeds back to the cloud platform prompt information that the routing configuration update is complete and the integrity check of the routing table of the scheduled VPC has passed.

[0129] The cloud platform generates routing configuration information for the border switch based on the VPC network segment of the scheduled VPC and the border switch model and routing type indicated in the traffic scheduling request.

[0130] The SDN controller of the cloud platform uses methods such as netconf to write the routing configuration information of the border switch into the border switch to complete the routing configuration of the border switch.

[0131] After completing the routing configuration, the edge switch feeds back a prompt message indicating that the routing configuration is completed to the cloud platform.

[0132] The cloud platform modifies the first traffic scheduling direction corresponding to the scheduled VPC and the next-hop VIP address of the host machine into the VIP address of the programmable chip switch cluster, completes the configuration synchronization of the host machine, and sends the VPC traffic of the scheduled VPC in the host machine to the IDC through the programmable chip switch cluster and the edge switch that have completed the routing configuration.

[0133] The cloud platform sends signaling to the programmable chip switch cluster, so that the programmable chip switch cluster publishes the VPC network segment of the scheduled VPC to the edge switch based on the signaling, so that the edge switch completes the corresponding routing configuration based on the received VPC network segment to complete the corresponding routing convergence.

[0134] The cloud platform closes the BFD session (for dynamic routing) between the dedicated access switch and the cloud edge core network router / WAN dedicated switch (IDC) to send all the traffic generated by the IDC to the edge switch, and forwards the traffic received by the edge switch to the host machine through the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction.

[0135] The cloud platform sends a traffic monitoring request for the scheduled VPC to the dedicated access switch.

[0136] The dedicated access switch feeds back the channel traffic of the scheduled VPC to the cloud platform based on the traffic monitoring request of the scheduled VPC.

[0137] When the cloud platform determines that the VPC traffic of the scheduled VPC has been completely migrated based on the traffic passing through the scheduled VPC, all route details related to the scheduled VPC under the NFV server gateway cluster branch pre-stored in the route management platform are deleted.

[0138] Based on the traffic scheduling method provided by the above embodiment of the present invention, the embodiment of the present invention also provides a traffic scheduling system, such as Figure 5 As shown, it is applicable to a network architecture with a network interconnection channel. The network architecture at least includes a virtual network, an NFV server gateway cluster, a programmable chip switch cluster, a physical network, a border switch, a dedicated line access switch and an interconnected data center. The system includes:

[0139] The acquisition unit 51 is used to receive a traffic scheduling request, and based on the traffic scheduling request, acquire the VPC traffic of the scheduled VPC from the NFV server gateway cluster and acquire the current cluster traffic usage rate from the programmable chip switch cluster;

[0140] A determination unit 52, configured to determine whether the programmable chip switch cluster can bear the VPC traffic according to the VPC traffic and the current cluster traffic usage rate;

[0141] A first routing configuration unit 53 is configured to perform routing configuration on the programmable chip switch cluster according to a routing table of the scheduled VPC generated according to the VPC instance information of the scheduled VPC if the programmable chip switch cluster can bear the VPC traffic;

[0142] A second routing configuration unit 54 is used to perform routing configuration on the edge switch according to the scheduled VPC traffic scheduling request;

[0143] The modification unit 55 is used to modify the first traffic scheduling direction corresponding to the scheduled VPC and the next-hop VIP address of the host machine into the VIP address of the programmable chip switch cluster to complete the configuration synchronization of the host machine; the first traffic scheduling direction is the VPC->IDC traffic scheduling direction;

[0144] The first traffic scheduling unit 56 is used to send the VPC traffic of the scheduled VPC in the host to the IDC through the programmable chip switch cluster and the edge switch to complete the VPC traffic scheduling in the first traffic scheduling direction;

[0145] The signaling sending unit 57 is used to send a signaling to the programmable chip switch cluster, so that the programmable chip switch cluster publishes the VPC network segment of the scheduled VPC to the edge switch based on the signaling, so that the edge switch completes routing convergence;

[0146] The second traffic scheduling unit 58 is used to send the traffic generated by the IDC to the host machine through the dedicated line access switch, the edge switch, the NFV server gateway cluster, and the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction.

[0147] The specific principles and execution processes of each unit in the power generation adjustment device of the extended-range system disclosed in the above-mentioned embodiment of the present invention are the same as the power generation adjustment method of the extended-range system disclosed in the above-mentioned embodiment of the present invention. Please refer to the corresponding parts of the power generation adjustment method of the extended-range system disclosed in the above-mentioned embodiment of the present invention, and will not be repeated here.

[0148] The embodiment of the present application provides a traffic scheduling system, which is applied to a cloud platform. When the cloud platform receives a traffic scheduling request, it can determine whether the programmable chip switch cluster can bear the VPC traffic of the scheduled VPC according to the VPC traffic of the scheduled VPC and the current cluster traffic usage rate of the programmable chip switch cluster; if it can bear it, the routing table of the scheduled VPC can be further used to configure the routing of the programmable chip switch cluster, and the routing configuration information of the border switch can be used to configure the routing of the border switch, so that when it is determined that the programmable chip switch cluster and the border switch have completed the corresponding routing configuration, it can be determined that the programmable chip switch cluster and the border switch currently have the ability to correctly forward the VPC traffic of the scheduled VPC, and then the first traffic scheduling direction corresponding to the scheduled VPC and the next-hop VIP address of the host machine are modified to the VIP address of the programmable chip switch cluster, and the configuration synchronization of the host machine is completed. At this time, the cluster and programmable chip switch cluster corresponding to the original next-hop VIP address can be determined. The programmable chip switch clusters all have all the routing details of the scheduled VPC and can correctly forward the VPC traffic of the scheduled VPC. Therefore, no packet loss will occur during the traffic forwarding process in the first traffic scheduling direction. After completing the VPC traffic scheduling in the first traffic scheduling direction, the VPC network segment of the scheduled VPC can be published to the border switch, so that the border switch completes the routing convergence, and the switch that indicates the forwarding of the traffic generated by the core network access router and the WAN dedicated line switch (IDC) in the second traffic scheduling direction is transformed from the dedicated line access switch and the NFV server gateway cluster to the dedicated line access switch + border switch and the NFV server gateway cluster + programmable chip switch cluster. Finally, the traffic generated by the IDC received by the dedicated line access switch and the border switch is forwarded to the host corresponding to the scheduled VPC through the NFV server gateway cluster and the programmable chip switch cluster, which can effectively avoid the problem of packet loss during the traffic scheduling process in the second traffic scheduling direction.

[0149] Optionally, the VPC traffic includes a VPC traffic statistic value of each gateway in the NFV server gateway cluster, and the current cluster traffic usage includes a current traffic usage of each switch in the programmable chip switch cluster; the determining unit includes:

[0150] A remaining capacity statistical value determining unit, configured to determine, for each switch, a remaining capacity statistical value of the switch according to the current traffic usage rate of the switch;

[0151] The judging unit is used to judge whether the programmable chip switch cluster can bear the VPC traffic according to the VPC traffic statistics of each gateway and the remaining capacity statistics of each switch.

[0152] Optionally, the first routing configuration unit includes:

[0153] A VPC instance information acquisition unit, used to acquire VPC instance information of the scheduled VPC;

[0154] A routing table generating unit, configured to obtain all routing details related to the scheduled VPC according to the VPC instance information of the scheduled VPC, and generate a routing table of the scheduled VPC according to all routing details related to the scheduled VPC;

[0155] The first routing configuration subunit is used to synchronize the routing table of the scheduled VPC to the programmable chip switch cluster, so that when the programmable chip switch cluster receives the routing table of the scheduled VPC, it performs an integrity check on the routing table of the scheduled VPC through each switch, and uses the routing table of the scheduled VPC to update its own stored routing table when the integrity check passes, so as to complete the routing configuration of the programmable chip switch.

[0156] Optionally, the second routing configuration unit includes:

[0157] A VPC network segment acquisition unit, used to acquire the VPC network segment of the scheduled VPC;

[0158] A routing configuration information generating unit, configured to generate routing configuration information of a border switch according to the VPC network segment and the border switch model and routing type indicated in the traffic scheduling request;

[0159] The second routing configuration subunit is used to write the routing configuration information of the border switch into the border switch to complete the routing configuration of the border switch.

[0160] Optionally, the second traffic scheduling unit includes:

[0161] Traffic balancing unit, used to send traffic generated by IDC to dedicated line access switch and edge switch in a balanced manner through hash rules;

[0162] The traffic forwarding unit is used to forward the traffic received by the dedicated line access switch to the host machine through the NFV server gateway cluster, and to forward the traffic received by the edge switch to the host machine through the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction.

[0163] Optionally, the traffic scheduling system provided in the embodiment of the present application further includes:

[0164] The third traffic scheduling unit is used to stop the session between the dedicated line access switch and the IDC, or stop the session between the NFV server gateway cluster and the IDC, send the traffic generated by the IDC to the edge switch, and forward the traffic received by the edge switch to the host machine through the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction.

[0165] Optionally, the traffic scheduling system provided in the embodiment of the present application further includes:

[0166] A collection unit, configured to collect the channel traffic of the scheduled VPC in the dedicated line access switch after completing the VPC traffic scheduling in the first traffic scheduling direction and the traffic scheduling in the second traffic scheduling direction;

[0167] The deletion unit is used to delete all pre-stored route details related to the scheduled VPC under the NFV server gateway cluster branch according to traffic monitoring.

[0168] The present application embodiment provides an electronic device, such as Figure 6 As shown, the electronic device includes a processor 601 and a memory 602, the memory 602 is used to store program codes and data for traffic scheduling, and the processor 601 is used to call program instructions in the memory to execute the steps shown in the traffic scheduling method in the above embodiment.

[0169] An embodiment of the present application provides a storage medium, which includes a storage program, wherein when the program is running, the device where the storage medium is located is controlled to execute the traffic scheduling method shown in the above embodiment.

[0170] The processor in this article can be a CPU of a terminal, or an MCU integrated in the terminal, or a combination of a CPU and an MCU. Moreover, the processor includes a kernel, which calls a corresponding program from a memory, and one or more kernels can be set.

[0171] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0172] For the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the order of the actions described, because according to the present invention, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0173] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0174] The steps in the methods of the embodiments of the present invention may be adjusted in order, combined, or deleted according to actual needs.

[0175] The modules and sub-modules in the apparatus and terminal in each embodiment of the present invention may be combined, divided or deleted according to actual needs.

[0176] In the several embodiments provided by the present invention, it should be understood that the disclosed terminals, devices and methods can be implemented in other ways. For example, the terminal embodiments described above are only schematic, for example, the division of modules or submodules is only a logical function division, and there may be other division methods in actual implementation, such as multiple submodules or modules can be combined or integrated into another module, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0177] The modules or submodules described as separate components may or may not be physically separated, and the components of the modules or submodules may or may not be physical modules or submodules, that is, they may be located in one place, or they may be distributed on multiple network modules or submodules. Some or all of the modules or submodules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0178] In addition, each functional module or submodule in each embodiment of the present invention may be integrated into one processing module, or each module or submodule may exist physically separately, or two or more modules or submodules may be integrated into one module. The above-mentioned integrated modules or submodules may be implemented in the form of hardware or in the form of software functional modules or submodules.

[0179] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0180] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0181] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A traffic scheduling method, characterized in that: Applicable to a network architecture with a network interconnection channel, the network architecture at least includes a virtual network, an NFV server gateway cluster, a programmable chip switch cluster, a physical network, an edge switch, a dedicated line access switch and an interconnected data center, the method includes: Receive a traffic scheduling request, and based on the traffic scheduling request, obtain the VPC traffic of the scheduled VPC from the NFV server gateway cluster and obtain the current cluster traffic usage rate from the programmable chip switch cluster; Determine whether the programmable chip switch cluster can bear the VPC traffic according to the VPC traffic and the current cluster traffic usage rate; If the programmable chip switch cluster can bear the VPC traffic, performing routing configuration on the programmable chip switch cluster according to the routing table of the scheduled VPC generated according to the VPC instance information of the scheduled VPC; Performing routing configuration on the border switch according to the traffic scheduling request of the scheduled VPC; Modify the first traffic scheduling direction corresponding to the scheduled VPC and the next-hop VIP address of the host machine to the VIP address of the programmable chip switch cluster to complete the configuration synchronization of the host machine; the first traffic scheduling direction is the VPC->IDC traffic scheduling direction; Sending the VPC traffic of the scheduled VPC in the host to the IDC through the programmable chip switch cluster and the edge switch to complete the VPC traffic scheduling in the first traffic scheduling direction; Sending a signaling to the programmable chip switch cluster, so that the programmable chip switch cluster publishes the VPC network segment of the scheduled VPC to the edge switch based on the signaling, so that the edge switch completes routing convergence; The traffic generated by the IDC is sent to the host machine through the dedicated access switch, the edge switch, the NFV server gateway cluster, and the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction; the second traffic scheduling direction is the IDC->VPC traffic scheduling direction.

2. The method according to claim 1, characterized in that The VPC traffic includes the VPC traffic statistics of each gateway in the NFV server gateway cluster, and the current cluster traffic usage includes the current traffic usage of each switch in the programmable chip switch cluster; The determining, according to the VPC traffic and the current cluster traffic usage rate, whether the programmable chip switch cluster can withstand the VPC traffic includes: For each of the switches, determining a statistical value of the remaining capacity of the switch according to the current traffic usage rate of the switch; According to the VPC traffic statistics of each gateway and the remaining capacity statistics of each switch, it is determined whether the programmable chip switch cluster can withstand the VPC traffic.

3. The method according to claim 1, characterized in that The routing table of the scheduled VPC generated according to the VPC instance information of the scheduled VPC performs routing configuration on the programmable chip switch cluster, including: Obtain VPC instance information of the scheduled VPC; Acquire all routing details related to the scheduled VPC according to the VPC instance information of the scheduled VPC, and generate a routing table of the scheduled VPC according to all routing details related to the scheduled VPC; The routing table of the scheduled VPC is synchronized to the programmable chip switch cluster, so that when the programmable chip switch cluster receives the routing table of the scheduled VPC, it performs an integrity check on the routing table of the scheduled VPC through each switch, and uses the routing table of the scheduled VPC to update its own stored routing table when the integrity check passes, so as to complete the routing configuration of the programmable chip switch.

4. The method according to claim 1, characterized in that: The performing routing configuration on the edge switch according to the scheduled VPC and the traffic scheduling request includes: Obtain the VPC network segment of the scheduled VPC; Generate routing configuration information of the border switch according to the VPC network segment and the border switch model and routing type indicated in the traffic scheduling request; The routing configuration information of the border switch is written into the border switch to complete the routing configuration of the border switch.

5. The method according to claim 1, characterized in that The sending the traffic generated by the IDC to the host machine through the dedicated line access switch, the edge switch, the NFV server gateway cluster, and the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction includes: The traffic generated by the IDC is sent to the dedicated line access switch and the edge switch in a balanced manner through a hash rule; The traffic received by the dedicated access switch is forwarded to the host machine through the NFV server gateway cluster, and the traffic received by the edge switch is forwarded to the host machine through the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction.

6. The method according to claim 1, characterized in that After the border switch completes routing convergence, the method further includes: Stop the session between the dedicated access switch and the IDC, or stop the session between the NFV server gateway cluster and the IDC, send the traffic generated by the IDC to the border switch, and forward the traffic received by the border switch to the host machine through the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction.

7. The method according to claim 1, characterized in that The method further comprises: After completing the VPC traffic scheduling in the first traffic scheduling direction and the traffic scheduling in the second traffic scheduling direction, collecting the channel traffic of the scheduled VPC from the dedicated line access switch; According to the traffic monitoring, all pre-stored routing details related to the scheduled VPC under the NFV server gateway cluster branch are deleted.

8. A traffic scheduling system, characterized in that: Applicable to a network architecture with a network interconnection channel, the network architecture at least includes a virtual network, an NFV server gateway cluster, a programmable chip switch cluster, a physical network, an edge switch, a dedicated line access switch and an interconnected data center, the system includes: An acquisition unit, configured to receive a traffic scheduling request, and based on the traffic scheduling request, acquire the VPC traffic of the scheduled VPC from the NFV server gateway cluster and acquire the current cluster traffic usage rate from the programmable chip switch cluster; A determination unit, configured to determine whether the programmable chip switch cluster can bear the VPC traffic according to the VPC traffic and the current cluster traffic usage rate; A first routing configuration unit, configured to perform routing configuration on the programmable chip switch cluster according to the routing table of the scheduled VPC generated according to the VPC instance information of the scheduled VPC if the programmable chip switch cluster can bear the VPC traffic; A second routing configuration unit, configured to perform routing configuration on the edge switch according to the traffic scheduling request of the scheduled VPC; A modification unit, used to modify the first traffic scheduling direction corresponding to the scheduled VPC and the next-hop VIP address of the host machine to the VIP address of the programmable chip switch cluster, so as to complete the configuration synchronization of the host machine; the first traffic scheduling direction is the VPC->IDC traffic scheduling direction; A first traffic scheduling unit, configured to send the VPC traffic of the scheduled VPC in the host to the IDC through the programmable chip switch cluster and the edge switch when the host completes configuration synchronization, so as to complete the VPC traffic scheduling in the first traffic scheduling direction; A signaling sending unit, configured to send a signaling to the programmable chip switch cluster when completing the VPC traffic scheduling in the first traffic scheduling direction, so that the programmable chip switch cluster publishes the VPC network segment of the scheduled VPC to the edge switch based on the signaling, so that the edge switch completes routing convergence; The second traffic scheduling unit is used to send the traffic generated by the IDC to the host machine through the dedicated access switch, the edge switch, the NFV server gateway cluster, and the programmable chip switch cluster to complete the traffic scheduling in the second traffic scheduling direction; the second traffic scheduling direction is the IDC->VPC traffic scheduling direction.

9. An electronic device, characterized in that: include: A processor and a memory, wherein the processor and the memory are connected via a communication bus; wherein the processor is used to call and execute a program stored in the memory; The memory is used to store a program, and the program is used to implement the traffic scheduling method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to execute the traffic scheduling method according to any one of claims 1 to 7.

Images