CN110753072A

CN110753072A - Load balancing system, method, device and equipment

Info

Publication number: CN110753072A
Application number: CN201810816993.8A
Authority: CN
Inventors: 黄毅
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2018-07-24
Filing date: 2018-07-24
Publication date: 2020-02-04
Anticipated expiration: 2038-07-24
Also published as: CN110753072B

Abstract

The application discloses load balancing system includes: the system comprises a gateway node, a first load balancing node cluster and a first back-end server; the gateway node is used for receiving an access data message which is sent by a public network client and has a destination address as an elastic public network address, determining that the access data message needs to be sent to the first load balancing node cluster according to the elastic public network address, packaging the access data message into a first access message by using a protocol of a first tunnel between the gateway node and the first load balancing node cluster, and sending the first access message through the first tunnel; the first load balancing node cluster is used for receiving a first access message through the first tunnel, obtaining an access data message from the first access message, generating a second access message according to the access data message, and sending the second access message to the first back-end server; and the first back-end server is used for receiving the second access message and carrying out service processing. The problem that load balancing examples are difficult to migrate among different load balancing node clusters according to example level granularity is solved.

Description

Load balancing system, method, device and equipment

Technical Field

The application relates to the field of cloud computing, in particular to a load balancing system. The application also relates to a load balancing method and device, electronic equipment and storage equipment. The application also relates to a second load balancing method and device, an electronic device and a storage device.

Background

As cloud computing becomes more and more heavily trafficked, cloud computing platforms providing cloud services typically distribute traffic through a Service Load Balancer (SLB) to one or more real cloud servers that actually process cloud computing traffic requests to extend service capabilities.

The service load balancer is provided by at least one load balancing node in a cluster of load balancing nodes. The service load balancer running at the load balancing node is called a load balancing instance. The load balancing node receives the flow through the load balancing instance and distributes the flow to a real cloud server which actually processes the cloud computing service request, and the real cloud server is also called a Backend server (or backup Servers). The load balancing instance running on the load balancing node generally comprises at least one monitor and at least one back-end server, the cloud computing service request is obtained through the monitor, the cloud computing service request is distributed to the back-end server according to a preset load balancing rule, and the purpose of sharing the load to the plurality of back-end servers is achieved.

The existing load balancing system generally comprises load balancing node clusters and a back-end server which are located in different routing areas, wherein the load balancing node clusters located in the different routing areas provide load balancing examples, the different load balancing node clusters respectively issue routes with different priorities for mutual standby, and migration of the load balancing examples among the different load balancing node clusters depends on route switching of a physical network. When the load balancing node cluster which originally issues the high-priority route fails, the physical network device detects the failure through the dynamic route, and all load balancing instances on the failed load balancing node cluster are migrated to the load balancing node cluster which originally issues the low-priority route. Because the distribution of the route is distributed by taking a network segment as a unit, the existing load balancing system is difficult to migrate load balancing examples among different load balancing node clusters according to example level granularity.

Disclosure of Invention

The application provides a load balancing system to solve the problem that an existing load balancing system is difficult to migrate load balancing examples among different load balancing node clusters according to example level granularity.

The application provides a load balancing system, including:

the system comprises a gateway node, a first load balancing node cluster and a first back-end server;

the gateway node is configured to receive an access data packet sent by a public network client and having a destination address as an elastic public network address, where the elastic public network address includes an address of a first load balancing node cluster, determine, according to the elastic public network address, that the access data packet needs to be sent to the first load balancing node cluster, encapsulate, using a protocol of a first tunnel between the gateway node and the first load balancing node cluster, the access data packet as a first access packet, and send the first access packet to the first load balancing node cluster through the first tunnel;

the first load balancing node cluster is configured to receive a first access packet sent by the gateway node through the first tunnel, obtain the access data packet from the first access packet according to a protocol of the first tunnel, generate a second access packet according to the access data packet, and send the second access packet to the first backend server;

and the first back-end server is used for receiving a second access message sent by the first load balancing node cluster and performing service processing according to the second access message.

Optionally, the load balancing system further includes: the second load balancing node cluster and the second back-end server are positioned in different routing areas from the first load balancing node cluster;

the gateway node is further configured to, if it is determined according to the elastic public network address that the access data packet needs to be sent to the second load balancing node cluster, encapsulate the access data packet into a third access packet using a protocol of a second tunnel between the gateway node and the second load balancing node cluster, and send the third access packet to the second load balancing node cluster through the second tunnel;

the second load balancing node cluster is configured to receive a third access packet sent by the gateway node through the second tunnel, obtain the access data packet from the first access packet according to a protocol of the second tunnel, generate a fourth access packet according to the access data packet, and send the fourth access packet to the second backend server;

and the second back-end server is configured to receive a fourth access packet sent by the second load balancing node cluster, and perform service processing according to the fourth access packet.

Optionally, the gateway node is specifically configured to:

acquiring indication information corresponding to the elastic public network address and used for indicating a tunnel between the elastic public network address and the load balancing node cluster;

if the indication information is the tunnel IP of the first tunnel, determining to send the access data message to the first load balancing node cluster; the tunnel IP of the first tunnel comprises at least one of a gateway node end IP of the first tunnel and a load balancing node cluster end tunnel IP of the first tunnel;

and if the indication information and the tunnel IP of the second tunnel are the same, determining to send the access data message to the second load balancing node cluster, wherein the tunnel IP of the second tunnel comprises at least one of a gateway node end IP of the second tunnel and a tunnel IP of the load balancing node cluster end of the second tunnel.

Optionally, the gateway node is specifically configured to:

and using the tunnel IP of the load balancing node cluster end of the first tunnel as a destination IP, using the tunnel IP of the gateway node end of the first tunnel as a source IP, and encapsulating the access data message into a first access message.

Optionally, the first load balancing node cluster is specifically configured to:

determining the first back-end server according to the destination address of the access data message and a preset rule;

modifying the source address of the access data message into a load balancing node IP of the first load balancing node cluster, and modifying the destination address of the access data message into a back-end server IP of the first back-end server to obtain a second access message; the load balancing node IP is an IP which is provided by the first load balancing node cluster and used for transmitting messages with the first back-end server in a load balancing mode; the back-end server IP is provided for the first back-end server and used for transmitting messages with the first load balancing node cluster in a load balancing mode.

Optionally, the first backend server is further configured to: generating a second response message, and sending the second response message to the first load balancing node cluster; the destination address of the second response message is the load balancing node IP of the first load balancing node cluster, and the source address of the second response message is the first back-end server IP;

the first load balancing node cluster is further configured to:

receiving the second response message sent by the first back-end server, generating a response message aiming at the access data message according to the second response message, packaging the response message aiming at the access data message into a first response message by using a protocol of the first tunnel, and sending the first response message to the gateway node through the first tunnel;

the gateway node is further configured to:

receiving a first response message sent by the first load balancing node cluster, obtaining a response message aiming at the access data message from the first response message, and sending the response message aiming at the access data message to the public network client.

obtaining session information of a public network client; the session information of the public network client at least comprises an address of the public network client, the elastic public network address, a load balancing node cluster end tunnel IP of the first tunnel and a gateway node end tunnel IP of the first tunnel;

and modifying the source address of the second response message into an elastic public network address according to the session information of the public network client, and modifying the destination address of the second response message into the address of the public network client to obtain a response message aiming at the access data message.

and using the tunnel IP of the load balancing node cluster end of the first tunnel as a source IP, using the tunnel IP of the gateway node end of the first tunnel as a destination IP, and encapsulating the response message aiming at the access data into a first response message according to a protocol of the first tunnel.

Optionally, the first load balancing node cluster is further configured to:

sending the source address of the access data message to the first back-end server;

correspondingly, the first backend server is further configured to:

performing service processing according to the source address of the access data message and the second access message;

and the source address of the access data message is the IP of the public network client.

The present application further provides a load balancing method, including:

receiving an access data message which is sent by a public network client and has a destination address which is an elastic public network address, wherein the elastic public network address comprises an address of a first load balancing node cluster;

determining that the access data message needs to be sent to the first load balancing node cluster according to the elastic public network address;

packaging the access data message into a first access message by using a protocol of a first tunnel between a gateway node and the first load balancing node cluster;

and sending the first access message to the first load balancing node cluster through the first tunnel.

Optionally, the load balancing method further includes:

if the access data message is determined to be required to be sent to the second load balancing node cluster according to the elastic public network address, using a protocol of a second tunnel between the gateway node and the second load balancing node cluster to package the access data message into a third access message;

and sending the third access packet to the second load balancing node cluster through the second tunnel.

Optionally, the load balancing method further includes:

if the indication information is the tunnel IP of the first tunnel, determining to send the access data message to the first load balancing node cluster, wherein the tunnel IP of the first tunnel comprises at least one of a gateway node end IP of the first tunnel and a load balancing node cluster end tunnel IP of the first tunnel;

and if the indication information is the tunnel IP of the second tunnel, determining to send the access data message to the second load balancing node cluster, wherein the tunnel IP of the second tunnel comprises at least one of a gateway node end IP of the second tunnel and a load balancing node cluster end tunnel IP of the second tunnel.

Optionally, the load balancing method further includes:

if the traffic of the first load balancing node cluster reaches a traffic threshold, and/or,

all or a portion of the load balancing nodes of the first cluster of load balancing nodes fail,

modifying the indication information into the tunnel IP of the second tunnel.

Optionally, the encapsulating, by using a protocol of a first tunnel between the gateway node and the first load balancing node cluster, the access data packet into a first access packet includes:

and using the tunnel IP of the load balancing node cluster end of the first tunnel as a destination IP, using the tunnel IP of the gateway node end of the first tunnel as a source IP, and encapsulating the received access data message into a first access message according to the protocol of the first tunnel.

Optionally, the load balancing method further includes:

receiving a first response message sent by the first load balancing node cluster through the first tunnel;

obtaining a response message aiming at the access data message from the first response message according to the protocol of the first tunnel;

and sending the response message aiming at the access data message to the public network client.

Optionally, the load balancing method further includes:

obtaining configuration information of the first tunnel;

saving the configuration information of the first tunnel;

the configuration information includes a protocol for establishing a first tunnel between the gateway node and the first load balancing node cluster, a gateway node end IP of the first tunnel, and a load balancing node cluster end tunnel IP of the first tunnel.

The present application further provides a load balancing method, including:

receiving a first access message which is sent by a gateway node and encapsulated according to a protocol of a tunnel through the tunnel between the gateway node and the gateway node;

obtaining an access data message from the first access message according to the protocol of the tunnel; the access data message is an access data message which is sent by a public network client and has a destination address of an elastic public network address, and the elastic public network address is an address of a load balancing node cluster;

generating a second access message according to the access data message;

and sending the second access message to a back-end server.

Optionally, the generating a second access packet according to the access data packet includes:

determining a back-end server according to the destination address of the access data message and a preset rule;

modifying the source address of the access data message into the load balancing node IP of the load balancing node cluster, and modifying the destination address of the access data message into the IP of the back-end server to obtain a second access message; the load balancing node IP is an IP which is provided by the load balancing node cluster and used for transmitting messages with the back-end server in a load balancing mode; the IP of the back-end server is the back-end server IP which is provided by the back-end server and used for transmitting the message with the load balancing node cluster in a load balancing mode.

Optionally, the load balancing method further includes:

receiving a second response message aiming at the second access message and sent by the back-end server, wherein the destination address of the second response message is the IP of the load balancing node, and the source address of the second response message is the IP of the back-end server;

generating a response message aiming at the access data message according to the second response message;

packaging the response message aiming at the access data message into a first response message by using the protocol of the tunnel;

and sending the first response message to the gateway node through the tunnel.

Optionally, the generating a response packet for the access data packet according to the second response packet includes:

obtaining session information of a public network client; the session information of the public network client at least comprises the address of the public network client, the elastic public network address, the tunnel IP of the load balancing node cluster end of the tunnel and the tunnel IP of the gateway node end of the tunnel;

and modifying the source address of the second response message into the elastic public network address according to the session information of the public network client, and modifying the destination address of the second response message into the address of the public network client to obtain the response message aiming at the access data message.

Optionally, the encapsulating, by using the protocol of the tunnel, the response packet for the access data packet into a first response packet includes:

and using the tunnel IP of the load balancing node cluster end of the tunnel as a source IP, using the tunnel IP of the gateway node end of the tunnel as a destination IP, and encapsulating the response message aiming at the access data into a first response message according to the protocol of the tunnel.

Optionally, the load balancing method further includes:

obtaining configuration information of the tunnel;

storing the configuration information of the tunnel;

the configuration information at least comprises a protocol for establishing the tunnel, a gateway node end tunnel IP and a load balancing node cluster end tunnel IP.

The present application further provides a load balancing apparatus, including:

the access message receiving unit is used for receiving an access data message which is sent by the public network client and has a destination address of an elastic public network address; the elastic public network address comprises an address of a first load balancing node cluster;

a first tunnel determining unit, configured to determine, according to the elastic public network address, that the access data packet needs to be sent to the first load balancing node cluster;

a first tunnel encapsulation unit, configured to encapsulate the access data packet into a first access packet using a protocol of a first tunnel with the first load balancing node cluster;

and the sending unit is used for sending the first access message to the first load balancing node cluster through the first tunnel.

The present application further provides a load balancing apparatus, including:

the access message receiving unit is used for receiving a first access message which is sent by the gateway node and is encapsulated according to a protocol of a tunnel through the tunnel between the gateway node and the access message receiving unit;

a decapsulation unit, configured to obtain an access data packet from the first access packet according to the protocol of the tunnel; the access data message is an access data message which is sent by a public network client and has a destination address of an elastic public network address, and the elastic public network address is an address of a load balancing node cluster;

the load balancing message generating unit is used for generating a second access message according to the access data message;

and the load balancing message sending unit is used for sending the second access message to the back-end server.

The present application further provides an electronic device, comprising:

a memory, and a processor;

the memory is to store computer-executable instructions, and the processor is to execute the computer-executable instructions to:

The present application further provides an electronic device, comprising:

a memory, and a processor;

generating a second access message according to the access data message;

and sending the second access message to a back-end server.

The present application further provides a storage device storing instructions that can be loaded by a processor and perform the steps of:

generating a second access message according to the access data message;

and sending the second access message to a back-end server.

Compared with the prior art, the method has the following advantages:

the application provides a load balancing system, includes: the system comprises a gateway node, a first load balancing node cluster and a first back-end server; the gateway node receives an access data message which is sent by a public network client and has a destination address as an elastic public network address, and determines a first load balancing node cluster according to the elastic public network address, so that when the access data message is determined to be sent to a load balancing node cluster different from the first load balancing node cluster according to the elastic public network address, the access data message can be sent to the corresponding load balancing node cluster, the load balancing example can be migrated from one load balancing node cluster to another load balancing node cluster, and the problem that the load balancing example is difficult to migrate among different load balancing node clusters according to example level granularity is solved.

Drawings

Fig. 1 is a schematic diagram of a load balancing system according to a first embodiment of the present application;

fig. 2 is a processing flow chart of a load balancing method according to a second embodiment of the present application;

fig. 3 is a schematic diagram illustrating message modification and flow provided in a second embodiment of the present application;

fig. 4 is a processing flow chart of a load balancing method according to a third embodiment of the present application;

fig. 5 is a schematic diagram illustrating message modification and flow provided in a third embodiment of the present application;

fig. 6 is a schematic diagram of a load balancing apparatus according to a fourth embodiment of the present application;

fig. 7 is a schematic diagram of a load balancing apparatus according to a fifth embodiment of the present application;

fig. 8 is a schematic diagram of an electronic device of the present application.

Detailed Description

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. This application is capable of implementation in many different ways than those herein set forth and of similar import by those skilled in the art without departing from the spirit of this application and is therefore not limited to the specific implementations disclosed below.

The application provides a load balancing system. The application also provides a load balancing method and device, the electronic equipment and the storage equipment. The application also provides a second load balancing method and device, an electronic device and a storage device. Details are described in the following examples one by one.

A first embodiment of the present application provides a load balancing system.

Referring to fig. 1, a schematic diagram of a load balancing system according to a first embodiment of the present application is shown.

The load balancing system shown in fig. 1 includes: a gateway node 101, a first load balancing node cluster 102, and a first back-end server 103;

the gateway node 101 is configured to receive an access data packet with a destination address as an elastic public network address sent by a public network client, where the elastic public network address includes an address of a first load balancing node cluster, determine, according to the elastic public network address, that the access data packet needs to be sent to the first load balancing node cluster 102, encapsulate, using a protocol of a first tunnel between the gateway node and the first load balancing node cluster, the access data packet as a first access packet, and send the first access packet to the first load balancing node cluster 102 through the first tunnel.

The elastic public network address comprises an address of a first load balancing node cluster. In practical applications, the elastic public network address may specifically be a service address of a load balancing node in a load balancing node cluster, for example, an elastic public network IP, or referred to as eip (elastic IP), a public network client accesses the elastic public network address through a public network (Internet), the elastic public network address may be provided by the gateway node 101, or a packet with a destination address of the elastic public network address may be routed to the gateway node 101, so that an access data packet with a destination address of the elastic public network address sent by the public network client may reach the gateway node 101.

For ease of understanding, the concept of a tunnel is first introduced. The tunnel is an end-to-end connection established between two network devices according to a tunnel protocol and used for receiving and/or sending data messages encapsulated according to the tunnel protocol, two sides of the established tunnel respectively provide tunnel IPs at two ends of the tunnel, and the same tunnel protocol is used. The tunneling protocol may employ GRE, L2TP, VXLAN, or the like.

The following is an example, taking the example information in fig. 1 as an example. The public network client IP is 4.32.100.20, and the elastic public network address is EIP 220.181.6.19. A first tunnel is established between the gateway node 101 and the first load balancing node cluster 102 for transmitting the first access packet. The tunnel IP of both ends of the tunnel of the first tunnel includes: a gateway node end tunnel IP at the gateway node 101 side and a load balancing node cluster end tunnel IP at the first load balancing node cluster 102 side. In fig. 1, IGW Tunnel-IP represents a gateway node end Tunnel IP, SLB Tunnel-IP represents a load balancing node cluster end Tunnel IP, igwtunel-IP is 100.68.2.2, and SLB Tunnel-IP is 100.68.3.2.

In this embodiment of the application, the gateway node 101 is specifically configured to: and using the tunnel IP of the load balancing node cluster end of the first tunnel as a destination IP, using the tunnel IP of the gateway node end of the first tunnel as a source IP, and encapsulating the access data message into a first access message. The first tunnel sends the first access packet to the first load balancing node cluster 102.

The first load balancing node cluster 102 is configured to receive a first access packet sent by the gateway node 101 through the first tunnel, obtain the access data packet from the first access packet according to a protocol of the first tunnel, generate a second access packet according to the access data packet, and send the second access packet to the first backend server 103.

The slave load balancing node cluster can forward the access data message to a back-end server by adopting various load balancing modes, such as a DR (direct route) mode and a fullnat mode. The DR mode is to modify the MAC address of the message, and the load balancing node and the back-end server are required to be in the same virtual local area network. In the fullnat mode, the source IP and the destination IP of the access data message need to be changed, the modified message is sent to a back-end server, and the back-end server performs service processing.

In the embodiment of the application, a fullnat conversion mode is adopted. Specifically, the first load balancing node cluster 102 is specifically configured to:

determining the first back-end server 103 according to the destination address of the access data message and a preset rule;

modifying the source address of the access data message into the load balancing node IP of the first load balancing node cluster 102, and modifying the destination address of the access data message into the back-end server IP of the first back-end server 103 to obtain a second access message; the load balancing node IP is an IP provided by the first load balancing node cluster 102 and used for transmitting a packet with the first backend server 103 in a load balancing manner; the back-end server IP is a back-end server IP provided by the first back-end server 103 and used for transmitting a packet with the first load balancing node cluster 102 in a load balancing manner.

Following the example with the example information in fig. 1. A source IP of the first access packet received by the first load balancing node cluster 102 is a gateway node end tunnel IP of the first tunnel, that is, 100.68.2.2, and a target IP is a load balancing node cluster end tunnel IP of the first tunnel, that is, 100.68.3.2; the first load balancing node cluster 102 decapsulates the first access packet according to a protocol of a first tunnel to obtain the access data packet, where a source IP of the access data packet is a public network client IP, that is, 4.32.100.20, and a destination IP is the elastic public network address, that is, 220.181.6.19. If the load balancing node IP determined by the first load balancing node cluster 102 according to 220.181.6.19 and a preset rule is 100.64.10.2 and the back-end server IP is 10.100.2.2, the source IP of the access data packet is modified to 100.64.10.2 and the destination IP of the access data packet is modified to 10.100.2.2, a second access packet is obtained, and the second access packet is sent to 10.100.2.2.

Preferably, the first load balancing node cluster 102 is further configured to: and sending the source address of the access data message to the first backend server 103. So as to be suitable for the scenario that the first backend server 103 needs to use the public network client address for service processing. For example, the real address of the public network client is passed to the first backend server 103 by adding an option in the TCP packet. The first back-end server 103 is configured to receive a second access packet sent by the first load balancing node cluster, and perform service processing according to the second access packet.

In this embodiment of the present application, the first backend server 103 is further configured to:

receiving a source address of the access data message;

In this embodiment of the application, the first backend server 103 is further configured to:

generating a second response message, and sending the second response message to the first load balancing node cluster; the second response message is generated by performing service processing according to a second access message, a destination address of the second response message is a load balancing node IP of the first load balancing node cluster, and a source address of the second response message is the first backend server IP.

The first load balancing node cluster 102 is further configured to:

receiving the second response packet sent by the first backend server 103, generating a response packet for the access data packet according to the second response packet, encapsulating the response packet for the access data packet into a first response packet by using the protocol of the first tunnel, and sending the first response packet to the gateway node 101 through the first tunnel. The first load balancing node cluster 103 is specifically configured to:

according to the session information of the public network client, modifying the source address of the second response message into an elastic public network address, and modifying the destination address of the second response message into the address of the public network client to obtain a response message aiming at the access data message;

The gateway node 101 is further configured to:

receiving a first response message sent by the first load balancing node cluster 102, obtaining a response message for the access data message from the first response message, and sending the response message for the access data message to the public network client.

In this embodiment of the present application, a second tunnel is further established between the gateway node 101 and a second load balancing node cluster 104, and the second load balancing node cluster 104 and the first load balancing node cluster are located in different routing areas. As shown in fig. 1, the first load balancing node cluster 103 and the gateway node 101 are located in a first routing area, and the second load balancing node cluster 104 is located in a second routing area. The second load balancing node cluster and the second back-end server 105 transfer the message in a load balancing manner.

The gateway node 101 is further configured to, if it is determined according to the elastic public network address that the access data packet needs to be sent to the second load balancing node cluster 104, use a protocol of a second tunnel between the gateway node 101 and the second load balancing node cluster 104 to encapsulate the access data packet into a third access packet, and send the third access packet to the second load balancing node cluster 104 through the second tunnel;

the second load balancing node cluster 104 is configured to receive a third access packet sent by the gateway node 101 through the second tunnel, obtain the access data packet from the first access packet according to a protocol of the second tunnel, generate a fourth access packet according to the access data packet, and send the fourth access packet to the second backend server 105;

the second back-end server 105 is configured to receive a fourth access packet sent by the second load balancing node cluster 104, and perform service processing according to the fourth access packet.

In this embodiment of the application, the gateway node 101 is specifically configured to:

Following the example of information illustrated in fig. 1. The corresponding relationship between the elastic public network address and the tunnel IP is stored on the gateway node 101, for example, 220.181.6.19 preferentially corresponds to the tunnel IP 100.68.2.2 at the load balancing node cluster end of the first tunnel, when the traffic carried by the first load balancing node cluster 102 reaches the traffic threshold (or capacity threshold), 220.181.6.19 is modified to the tunnel IP at the load balancing node cluster end corresponding to the second tunnel, for example, 100.68.30.2, so that the traffic with the destination address of 220.181.6.19 is migrated to the second load balancing node cluster through the second tunnel. It should be noted that the gateway node end tunnel IP of the second tunnel may be the same as or different from the gateway node end tunnel IP of the first tunnel.

Based on the foregoing embodiments, a second embodiment of the present application provides a load balancing method, and please refer to the corresponding description of the foregoing embodiments for related parts.

The load balancing method provided in the second embodiment of the present application may be deployed in the gateway node 101 shown in fig. 1, and by receiving an access data packet whose destination address is an elastic public network address and sent by a public network client, if it is determined that the access data packet needs to be sent to the first load balancing node cluster according to the elastic public network address, the access data packet is encapsulated as a first access packet by using a protocol of a first tunnel between a gateway node and the first load balancing node cluster; and sending the first access message to the first load balancing node cluster through the first tunnel. And when the access data message is determined to be required to be sent to a load balancing node cluster different from the first load balancing node cluster according to the elastic public network address, sending the access data message to the load balancing node cluster indicated by the corresponding information, and realizing the migration of the load balancing example from one load balancing node cluster to another load balancing node cluster.

Fig. 2 shows a processing flow of the load balancing method, which includes: step S201 to step S204.

Step S201, receiving an access data message with a destination address as an elastic public network address sent by a public network client.

The elastic public network address comprises an address of a first load balancing node cluster. In practical application, the elastic public network address may be a service address of a load balancing node in a load balancing node cluster.

The step is to receive the access data message sent by the public network client.

In this embodiment, the public network client accesses the Elastic public network address through the public network (Internet), and the destination address of the sent access data message is an Elastic public network address, such as an Elastic IP (Elastic IP or EIP).

Examples are as follows: fig. 3 is a modification and flow direction of a message in an exemplary implementation manner provided in a second embodiment of the present application. In the example shown in fig. 3, the IP of the public network client is 4.32.100.20, the EIP of the elastic public network address is 220.181.6.19, and 301 shows the structure of the access data packet with the destination address of 220.181.6.19 sent by the public network client. And the access data message is sent to the gateway node through the Internet.

Step S202, according to the elastic public network address, determining that the access data message needs to be sent to the first load balancing node cluster.

This step is to determine that the access data packet needs to be sent to the first load balancing node cluster.

In the embodiment of the present application, the load balancing node cluster to which the access data packet is sent is specifically determined according to the elastic public network address. For example, determining to send a message, which is destined to the elastic public network address, to the first load balancing node cluster through the first tunnel according to routing information; or determining the load balancing node cluster to which the access data message is sent according to indication information corresponding to the elastic public network address and used for indicating a tunnel between the load balancing node cluster and the elastic public network address. The method specifically comprises the following steps:

and if the indication information is the tunnel IP of the second tunnel, determining to send the access data message to the second load balancing node cluster, wherein the tunnel IP of the second tunnel comprises at least one of a gateway node end IP of the second tunnel and a load balancing node cluster end tunnel IP of the second tunnel. Further, the access data packet is encapsulated into a third access packet by using a protocol of a second tunnel between the gateway node and the second load balancing node cluster; and sending the third access packet to the second load balancing node cluster through the second tunnel.

Step S203, using a protocol of a first tunnel between the gateway node and the first load balancing node cluster, to encapsulate the access data packet into a first access packet.

In this step, the received access data message is encapsulated into a first access message by using a protocol of a first tunnel.

In this embodiment of the present application, the first tunnel is a tunnel established according to a tunnel protocol between the gateway node and the first load balancing node cluster, and configured to receive and/or send a packet encapsulated according to the tunnel protocol. The protocol of the first tunnel may adopt a GRE, L2TP, VXLAN, and other tunneling protocols. Tunnel IP at both ends of the first tunnel: a gateway node end tunnel IP at the gateway node side and a load balancing node cluster end tunnel IP at the first load balancing node cluster side. For example, in fig. 3, IGW Tunnel-IP represents a gateway node end Tunnel IP, SLB Tunnel-IP represents a load balancing node cluster end Tunnel IP, IGW Tunnel-IP is 100.68.2.2, and SLB Tunnel-IP is 100.68.3.2.

In this embodiment of the present application, the following processing is specifically adopted to encapsulate the access data packet into a first access packet, including:

Following the example shown in fig. 3. The gateway node end tunnel IP is 100.68.2.2 as a source IP, the access data message shown in 301 is encapsulated on the outer layer of the tunnel by using the load balancing node cluster end tunnel IP100.68.3.2 as a destination IP according to a tunnel protocol to obtain a first access message shown in 302, the destination address of the outer layer of the first access message is 100.68.3.2, the source address is 100.68.2.2, the tunnel encapsulation inner layer IP message is the access data message, and the load (payload) is kept unchanged in the encapsulating process.

In the embodiment of the present application, the following processing is further included to establish the tunnel:

obtaining configuration information of the first tunnel;

saving the configuration information of the first tunnel;

Step S204, sending the first access packet to the first load balancing node cluster through the first tunnel.

The first access message is sent to the first load balancing node cluster through the first tunnel. The first access message is a message encapsulated according to a protocol of the first tunnel.

In the embodiment of the application, the method further includes obtaining indication information corresponding to the elastic public network address and used for indicating a tunnel between the elastic public network address and the load balancing node cluster; and determining the load balancing node cluster to which the access data message is sent according to the indication information. The method specifically comprises the following steps:

In the embodiment of the application, if the load balancing instance corresponding to the elastic public network address originally operates in the first load balancing node cluster, a first tunnel is established between the gateway node and the first load balancing node cluster and used for transmitting the first access message, and the elastic public network address corresponds to a tunnel IP at the load balancing end of the first tunnel; when the load balancing example to be migrated corresponding to the elastic public network address needs to be migrated from the first load balancing node cluster to the second load balancing node cluster, a load balancing example providing the same service of the load balancing example to be migrated is created in the second load balancing node cluster, a second tunnel is established between the gateway node and the second load balancing node cluster, and a tunnel IP corresponding to the elastic public network address is modified into a tunnel IP at the load balancing node cluster end of the second tunnel, so that an access data message aiming at the elastic public network address can be sent to the second load balancing node cluster by using the second tunnel, and the purpose of migrating the load balancing example is achieved.

Preferably, if the traffic of the first load balancing node cluster reaches a traffic threshold, and/or,

modifying the indication information into the tunnel IP of the second tunnel.

In this embodiment of the present application, the gateway node further needs to obtain a response packet for the access data packet, and send the response packet for the access data packet to the public network client. The specific treatment comprises the following steps:

Specifically, a response packet for the access data packet is obtained from the first response packet according to the tunneling protocol of the first tunnel.

Following the example shown in fig. 3. Receiving, through the first tunnel, a packet which is sent by a first load balancing node cluster and has a tunnel encapsulation added to an outer layer, such as a first response packet shown in 303 in the drawing, where an inner layer of the tunnel encapsulation of the first response packet is a response packet for the access data packet, that is, a packet source IP of the inner layer is an elastic public network IP: 220.181.6.19, the message destination IP of the inner layer is public network client IP: 4.32.100.20, respectively; the source IP encapsulated by the outer layer tunnel is the tunnel IP of the load balancing node cluster end: 100.68.3.2, the destination IP of the outer tunnel encapsulation is the gateway node end tunnel IP: 100.68.2.2. the load payload is data content of a response message generated by the first back-end server for service processing, and remains unchanged in the source IP and destination IP modification or encapsulation process of the message.

Based on the foregoing embodiments of the present application, a third embodiment of the present application provides a second load balancing method. The load balancing method provided in the third embodiment may be deployed in a load balancing node cluster including at least one load balancing node.

The load balancing method provided by the third embodiment is described below with reference to fig. 4 to 5. Since the third embodiment is based on the above embodiments, the description is simple, and the relevant portions should be referred to the corresponding description of the above embodiments.

The load balancing method shown in fig. 4 includes: step S401 to step S404.

Step S401, a first access message which is sent by a gateway node and encapsulated according to a protocol of a tunnel is received through the tunnel between the gateway node and the gateway node.

The method comprises the steps of receiving a first access message with tunnel encapsulation, wherein the first access message encapsulates an access data message which is sent by a public network client and has a destination address as an elastic public network address by using a tunnel protocol.

In this embodiment of the present application, the tunnel is an end-to-end connection established between the gateway node and the load balancing node cluster according to a tunnel protocol by using a tunnel IP for receiving and/or sending a packet encapsulated according to the tunnel protocol. The tunneling protocol can adopt GRE, L2TP, VXLAN and other tunneling protocols. The tunnel IP provided by the two sides for establishing the tunnel comprises the following steps: and the gateway node end tunnel IP at the gateway node side and the load balancing node cluster end tunnel IP at the load balancing node cluster side.

For example, in fig. 5, IGW Tunnel-IP represents a gateway node end Tunnel IP, SLB Tunnel-IP represents a load balancing node cluster end Tunnel IP, IGW Tunnel-IP is 100.68.2.2, and SLB Tunnel-IP is 100.68.3.2. . An example of a received first access packet is shown at 501 in fig. 5: the source IP of the inner layer IP message is a public network client IP, namely 4.32.100.20; the destination IP of the inner layer IP message is an elastic public network address EIP (an EIP 220.181.6.19); the source IP encapsulated by the outer layer tunnel is a gateway node end tunnel IP, namely an IGWTUNNel-IP in the figure, which is 100.68.2.2; the destination IP of the outer layer Tunnel encapsulation is a load balancing node cluster end Tunnel IP, i.e. the SLB Tunnel-IP in the figure is 100.68.3.2.

In addition, in the embodiment of the present application, the following processing is further included:

obtaining configuration information of the tunnel;

storing the configuration information of the tunnel;

Step S402, obtaining access data message from the first access message according to the protocol of the tunnel.

The access data message is an access data message which is sent by a public network client and has a destination address of an elastic public network address, and the elastic public network address is an address of a load balancing node cluster. Specifically, the elastic public network address may specifically be a service address of a load balancing node in a load balancing node cluster

In the step, the access data message is obtained from the first access message, so that the access data message is transmitted in a load balancing mode in the subsequent steps.

In this embodiment, the method specifically includes taking out an inner-layer encapsulated packet from the first access packet according to the tunneling protocol of the tunnel. Along the example shown in fig. 5, the source IP of the extracted inner layer message is 4.32.100.20; the destination IP is 220.181.6.19, the load (payload) of the inner layer message remains unchanged, and the fetched memory message is an access data message with the destination address being an elastic public network address sent by the public network client.

Step S403, generating a second access packet according to the access data packet.

In this step, a second access message is generated for further processing by sending to the back-end server in the subsequent steps.

The load balancing node cluster may transmit the access data packet obtained in step S402 to the backend server in a plurality of load balancing manners. Such as DR (direct route) mode and fullnat mode. The DR mode is to modify the MAC address of the message, and the load balancing node and the back-end server are required to be in the same virtual local area network. In the fullnat mode, the source IP and the destination IP of the access data message need to be changed, the modified message is sent to a back-end server, and the back-end server performs service processing.

In the embodiment of the present application, the second access packet is generated specifically by the following processing:

Following the example shown in fig. 5, the explanation will be given by taking a fullnat method as an example for forwarding a message. In fig. 5, the load balancing node IP is 100.64.10.2, and the back-end server provides 10.100.2.2 IP; modifying the source IP of the data access message from 4.32.100.20 to 100.64.10.2; the destination IP is modified from 220.181.6.19 to 10.100.2.2; the payload (payload) of the message remains unchanged. The generated second access packet is shown as 502: the source IP is 100.64.10.2, the destination IP is 10.100.2.2, and the payload (payload) is the payload of the access data packet.

And step S404, sending the second access message to a back-end server.

The second access message is sent to the back-end server.

In this embodiment of the present application, the source address of the access data packet is further sent to the back-end server. The method is suitable for the scene that the address of the public network client is needed for the back-end server to process the service. For example, the real address of the public network client is transferred to the back-end server by adding an option in the TCP packet.

In the embodiment of the present application, the method further includes:

and sending the first response message to the gateway node through the tunnel.

Following the example shown in fig. 5. The load of the second response packet 503 in fig. 5 is the service processing result of the backend server, the source IP is the IP provided by the backend server, i.e., 10.100.2.2, and the source IP is the IP of the load balancing node, i.e., 100.64.10.2.

In this embodiment, specifically, generating a response packet for the access data packet according to the second response packet through the following processing includes:

Following the example shown in fig. 5. Modifying the source IP in the second response message shown in 503 from 10.100.2.2 to an elastic public network address 220.181.6.19, modifying the destination IP from 100.64.10.2 to a public network client IP4.32.100.20, and keeping the load (payload) of the second response message unchanged, thereby obtaining the response message for the access data message.

In this embodiment of the application, the encapsulating the response packet to the access data packet into the first response packet by using the protocol of the tunnel through the following processing specifically includes:

Following the example shown in fig. 5. The first response packet encapsulated by the load balancing node cluster by using the tunnel IP is an example shown as 504 in fig. 5: the source IP of the inner layer IP message is an elastic public network address, namely EIP220.181.6.19, and the destination IP is a public network client IP, namely 4.32.100.20; after tunnel IP encapsulation is used, the outer layer source IP is the load balancing node cluster end tunnel IP100.68.3.2, and the destination IP is the gateway node end tunnel IP 100.68.2.2.

Corresponding to the load balancing method provided in the second embodiment of the present application, a fourth embodiment of the present application provides a load balancing apparatus.

Referring to fig. 6, a schematic diagram of a load balancing apparatus according to a fourth embodiment of the present application is shown. Since the device embodiment is basically similar to the method embodiment, the description is relatively simple, and the relevant portions only need to refer to the corresponding description of the method embodiment.

The load balancing apparatus shown in fig. 6 includes:

an access packet receiving unit 601, configured to receive an access data packet sent by a public network client and having a destination address as an elastic public network address;

a first tunnel determining unit 602, configured to determine, according to the elastic public network address, that the access data packet needs to be sent to the first load balancing node cluster;

a first tunnel encapsulation unit 603, configured to encapsulate the access data packet into a first access packet using a protocol of a first tunnel between a gateway node and the first load balancing node cluster;

a sending unit 603, configured to send the first access packet to the first load balancing node cluster through the first tunnel.

Optionally, the load balancing apparatus further includes a second tunnel encapsulation unit, where the second tunnel encapsulation unit is configured to:

Optionally, the load balancing apparatus further includes a determining unit, where the determining unit is configured to:

Optionally, the load balancing apparatus further includes a migration unit, where the migration unit is configured to:

modifying the indication information into the tunnel IP of the second tunnel.

Optionally, the first tunnel encapsulation unit 603 is specifically configured to:

Optionally, the load balancing apparatus further includes a receiving unit, where the receiving unit is configured to:

Optionally, the load balancing apparatus further includes a configuration unit, where the configuration unit is configured to:

obtaining configuration information of the first tunnel;

saving the configuration information of the first tunnel;

Corresponding to the load balancing method provided in the second embodiment of the present application, a fourth embodiment of the present application provides a second load balancing apparatus.

Referring to fig. 7, a schematic diagram of a load balancing apparatus according to a fourth embodiment of the present application is shown. Since the device embodiment is basically similar to the method embodiment, the description is relatively simple, and the relevant portions only need to refer to the corresponding description of the method embodiment.

The load balancing apparatus shown in fig. 7 includes:

an access packet receiving unit 701, configured to receive, through a tunnel with a gateway node, a first access packet that is sent by the gateway node and encapsulated according to a protocol of the tunnel;

a decapsulation unit 702, configured to obtain an access data packet from the first access packet according to the protocol of the tunnel; the access data message is an access data message which is sent by a public network client and has a destination address of an elastic public network address, and the elastic public network address is an address of a load balancing node cluster;

a load balancing packet generating unit 703, configured to generate a second access packet according to the access data packet;

and a load balancing message sending unit 704, configured to send the second access message to the backend server.

Optionally, the load balancing packet generating unit 703 is specifically configured to:

Optionally, the load balancing apparatus further includes a response receiving unit, where the response receiving unit is configured to:

and sending the first response message to the gateway node through the tunnel.

Optionally, the response receiving unit is specifically configured to:

Optionally, the response receiving unit is specifically configured to include:

obtaining configuration information of the tunnel;

storing the configuration information of the tunnel;

A sixth embodiment of the present application further provides an electronic device for implementing the load balancing method provided in the second embodiment, and fig. 8 is a schematic diagram illustrating an electronic device provided in the sixth embodiment of the present application.

The electronic device according to the sixth embodiment of the present application is described more simply, and please refer to the corresponding description of the first embodiment for related parts.

The electronic device shown in fig. 8 includes:

a memory 801, and a processor 802;

the memory 801 is configured to store computer-executable instructions, and the processor 802 is configured to execute the computer-executable instructions to:

Optionally, the processor 802 is further configured to execute the following computer-executable instructions:

modifying the indication information into the tunnel IP of the second tunnel.

obtaining configuration information of the first tunnel;

saving the configuration information of the first tunnel;

A seventh embodiment of the present application further provides an electronic device for implementing the load balancing method provided in the third embodiment, and a schematic structural diagram of the electronic device is similar to that in fig. 8.

The electronic device according to the seventh embodiment of the present application is described more simply, and please refer to the corresponding description of the third embodiment for related parts.

An electronic device provided in a seventh embodiment of the present application includes:

a memory, and a processor;

generating a second access message according to the access data message;

and sending the second access message to a back-end server.

Optionally, the processor is further configured to execute the following computer-executable instructions:

and sending the first response message to the gateway node through the tunnel.

obtaining configuration information of the tunnel;

storing the configuration information of the tunnel;

The eighth embodiment of the present application further provides a storage device for use in the load balancing method provided in the second embodiment, which is described more simply, and please refer to the corresponding description of the second embodiment for relevant portions.

An eighth embodiment of the present application provides a storage device, storing instructions, the instructions being capable of being loaded by a processor and performing the following steps:

A ninth embodiment of the present application further provides a storage device for use in the load balancing method provided in the third embodiment, which is described more simply, and please refer to the corresponding description of the third embodiment for relevant portions.

A ninth embodiment of the present application provides a storage device storing instructions that can be loaded by a processor and perform the steps of:

generating a second access message according to the access data message;

and sending the second access message to a back-end server.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

1. Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include non-transitory computer readable media (transmyedia), such as modulated data signals and carrier waves.

2. As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

Although the present application has been described with reference to the preferred embodiments, it is not intended to limit the present application, and those skilled in the art can make variations and modifications without departing from the spirit and scope of the present application, therefore, the scope of the present application should be determined by the claims that follow.

Claims

1. A load balancing system, comprising:

2. The system of claim 1, further comprising: the second load balancing node cluster and the second back-end server are positioned in different routing areas from the first load balancing node cluster;

3. The system according to claim 2, wherein the gateway node is specifically configured to:

4. The system according to claim 1, wherein the gateway node is specifically configured to:

5. The system of claim 1, wherein the first load-balancing node cluster is specifically configured to:

6. The system of claim 5, wherein the first back-end server is further configured to: generating a second response message, and sending the second response message to the first load balancing node cluster; the destination address of the second response message is the load balancing node IP of the first load balancing node cluster, and the source address of the second response message is the first back-end server IP;

the first load balancing node cluster is further configured to:

the gateway node is further configured to:

7. The system of claim 6, wherein the first load balancing node cluster is specifically configured to:

8. The system of claim 6, wherein the first load balancing node cluster is specifically configured to:

9. The system of claim 1, wherein the first cluster of load balancing nodes is further configured to:

correspondingly, the first backend server is further configured to:

10. A method of load balancing, comprising:

11. The method of claim 10, further comprising:

12. The method of claim 10, further comprising:

13. The method of claim 12, further comprising:

modifying the indication information into the tunnel IP of the second tunnel.

14. The method of claim 10, wherein encapsulating the access data packet as a first access packet using a protocol of a first tunnel between a gateway node and the first cluster of load balancing nodes comprises:

15. The method of claim 10, further comprising:

16. The method of claim 10, further comprising:

obtaining configuration information of the first tunnel;

saving the configuration information of the first tunnel;

17. A method of load balancing, comprising:

generating a second access message according to the access data message;

and sending the second access message to a back-end server.

18. The method of claim 17, wherein generating a second access packet based on the access data packet comprises:

19. The method of claim 18, further comprising:

and sending the first response message to the gateway node through the tunnel.

20. The method of claim 19, wherein generating a response packet for the access data packet from the second response packet comprises:

21. The method of claim 20, wherein encapsulating the response packet to the access data packet as a first response packet using the tunneled protocol comprises:

22. The method of claim 17, further comprising:

obtaining configuration information of the tunnel;

storing the configuration information of the tunnel;

23. A load balancing apparatus, comprising:

24. A load balancing apparatus, comprising:

25. An electronic device, comprising:

a memory, and a processor;

26. An electronic device, comprising:

a memory, and a processor;

generating a second access message according to the access data message;

and sending the second access message to a back-end server.

27. A storage device having stored thereon instructions capable of being loaded by a processor and performing the steps of:

28. A storage device having stored thereon instructions capable of being loaded by a processor and performing the steps of:

generating a second access message according to the access data message;

and sending the second access message to a back-end server.