WO2023207189A1 - Load balancing method and system, computer storage medium, and electronic device - Google Patents

Abstract

The present disclosure relates to the technical field of cloud computing. Provided are a load balancing method and system, a medium, and a device. The method comprises: pre-configuring a cloud host resource pool, wherein the cloud host resource pool comprises a plurality of cloud hosts, and each cloud host is associated with a pre-allocated IP segment; in response to a load balancing instance creation request, scheduling at least two candidate cloud hosts from the cloud host resource pool; selecting an idle IP from an IP segment corresponding to each candidate cloud host, and invoking a domain name server to store a domain name corresponding to at least two idle IPs, wherein the at least two idle IPs correspond to the same domain name; returning the domain name to a client, such that the client acquires, from the domain name server, a target IP corresponding to the domain name; and creating a load balancing instance on the basis of the target IP, such that the client establishes a connection with a target cloud host by means of the load balancing instance. The present disclosure provides a load balancing solution with a wider application range.

Description

Load balancing method and system, computer storage media, electronic equipment

Cross-references to related applications

This disclosure claims priority to the Chinese patent application with application number 202210470381.4 and titled "Load Balancing Method and System, Computer Storage Media, Electronic Devices" submitted on April 28, 2022. The entire content of this Chinese patent application is incorporated by reference. incorporated in this disclosure.

Technical field

The present disclosure relates to the technical field of cloud computing, and in particular to a load balancing method, a load balancing system, a computer storage medium and an electronic device.

Background technique

Kubernetes clusters deployed on private cloud bare metal architecture (relative to cloud environments) usually cannot use LoadBalancer type services. Because Kubernetes itself does not provide an implementation of a network load balancer (a service of type LoadBalancer) for bare metal clusters. If the Kubernetes cluster is not running on the IaaS (Infrastructure as a Service, Infrastructure as a Service) platform of the public cloud, LoadBalancers will remain in the "suspended" state indefinitely when created, which means that only the public cloud vendor's own Kubernetes Support LoadBalancer.

Currently, MetalLB load balancers are generally used to solve the problem that private bare metal clusters cannot use LoadBalancer type services. MetalLB supports two methods, namely Layer2 mode and BGP (Border Gateway Protocol) mode. However, the Layer 2 mode has the problem of limited network bandwidth due to single-point forwarding. The BGP mode cannot effectively respond to node failures. When a node goes down, all nodes will stop working.

In view of this, there is an urgent need to develop a new load balancing method and system in this field.

It should be noted that the information disclosed in the above background section is only used to enhance understanding of the background of the present disclosure.

Contents of the invention

One or more embodiments of the present disclosure provide a load balancing method, a load balancing system, a computer storage medium, and an electronic device, thereby overcoming, at least to a certain extent, the problem in related technologies that a load balancing device is required to implement the load balancing function. .

Additional features and advantages of the disclosure will be apparent from the following detailed description, or, in part, may be learned by practice of the disclosure.

According to a first aspect of the present disclosure, a load balancing method is provided, which pre-configures a cloud host resource pool. The cloud host resource pool includes multiple cloud hosts, and each cloud host is associated with a pre-allocated IP segment. The method includes: in response to a load balancing instance creation request, scheduling at least two candidate cloud hosts from a cloud host resource pool; the load balancing instance is used to provide external client access; and selecting the IP segment corresponding to each of the candidate cloud hosts. Select an idle IP, call the domain name server to store at least two domain names corresponding to the idle IP; wherein, at least two of the idle IPs correspond to the same domain name; return the domain name to the client, so that the client starts from Obtain the target IP corresponding to the domain name from the domain name server; the target IP is any one of the at least two idle IPs; create the load balancing instance based on the target IP so that the client can pass the The load balancing instance establishes a connection with the target cloud host; the target IP is the IP corresponding to the load balancing instance, and the target cloud host is the candidate cloud host corresponding to the target IP.

According to a second aspect of the present disclosure, a load balancing method is provided, which is applied to a target cloud host. The method includes: receiving a business request sent by a client through a pre-created load balancing instance; based on the load balancing instance associated with the load balancing instance. Load balancing strategy forwards the service request to the specified backend service.

According to a third aspect of the present disclosure, a load balancing system is provided, including: a cloud host management module and a cloud host resource pool; the cloud host management module is used to manage a preconfigured cloud host resource pool, and the cloud host resource pool including a plurality of cloud hosts, each cloud host associated with a pre-allocated IP segment; the cloud host management module is used to respond to a load balancing instance creation request, dispatch at least two candidate cloud hosts from the cloud host resource pool ; The load balancing instance is used to provide external client access; select an idle IP from the IP segment corresponding to each of the candidate cloud hosts, and call the domain name server to store at least two domain names corresponding to the idle IP; wherein, at least The two idle IPs correspond to the same domain name; the domain name is returned to the client, so that the client obtains the target IP corresponding to the domain name from the domain name server; the target IP is the at least two idle IP addresses. Any of the IPs; create the load balancing instance based on the target IP, so that the client establishes a connection with the target cloud host through the load balancing instance; the target IP is the IP corresponding to the load balancing instance, The target cloud host is a candidate cloud host corresponding to the target IP; the target cloud host is used to receive a service request sent by the client through the load balancing instance; based on the load balancing policy associated with the load balancing instance, Forward the service request to the designated backend service.

According to a fourth aspect of the present disclosure, a computer storage medium is provided, a computer program is stored thereon, and when the computer program is executed by a processor, the load balancing method described in the first and second aspects is implemented.

According to a fifth aspect of the present disclosure, an electronic device is provided, including: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform the operation via executing the executable instructions. The load balancing method described in the first aspect and the second aspect is executed.

It can be seen from the above technical solutions that the load balancing method, load balancing system, computer storage medium and electronic device in the exemplary embodiments of the present disclosure have at least the following advantages and positive effects:

In the technical solution provided by some embodiments of the present disclosure, on the one hand, in response to a load balancing instance creation request, at least two candidate cloud hosts are scheduled from the cloud host resource pool, and one is selected from the IP segment corresponding to each cloud host. Idle IP, and call the domain name server to store the domain names corresponding to at least two idle IPs, and return the domain names to the client, which can avoid the technical problems of business interruption caused by calling a single cloud host for request forwarding each time when the cloud host fails. , improve the system’s emergency handling capabilities. Furthermore, the target IP is associated with the load balancing instance, so that the client can establish a connection with the target cloud host through the load balancing instance, providing a new solution that can realize the load balancing function without using load balancing equipment, reducing operation and maintenance. Invest.

It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit the present disclosure.

Description of drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure. Obviously, the drawings in the following description are only some embodiments of the present disclosure. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without exerting creative efforts.

Figure 1 shows a schematic flowchart of a load balancing method in an embodiment of the present disclosure;

Figure 2 shows a schematic flow chart of the cloud host management module performing fault monitoring on the target cloud host in the embodiment of the present disclosure;

Figure 3 shows a schematic flow chart of how to update the status of the backup cloud host so that the backup cloud host takes over the target cloud host in an embodiment of the present disclosure;

Figure 4 shows a schematic flow chart of adding a cloud host to the cloud host resource pool in an embodiment of the present disclosure;

Figure 5 shows a schematic flowchart of removing faulty cloud hosts from the cloud host resource pool in an embodiment of the present disclosure;

Figure 6 shows a schematic flowchart of creating a load balancing instance in an embodiment of the present disclosure;

Figure 7 shows a schematic flowchart of deleting a load balancing instance in an embodiment of the present disclosure;

Figure 8 shows a schematic flow chart of how to manage a cloud host in an embodiment of the present disclosure;

Figure 9 shows a schematic structural diagram of a load balancing system in a cloud environment in an embodiment of the present disclosure;

FIG. 10 shows a schematic structural diagram of an electronic device in an exemplary embodiment of the present disclosure.

Detailed ways

Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concepts of the example embodiments. To those skilled in the art. The described features, structures or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to provide a thorough understanding of embodiments of the disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details described, or other methods, components, devices, steps, etc. may be adopted. In other instances, well-known technical solutions have not been shown or described in detail to avoid obscuring aspects of the disclosure.

The terms "a", "an", "the" and "said" are used in this specification to indicate the existence of one or more elements/components/etc.; the terms "include" and "have" are used to indicate an open-ended Inclusive is intended and means that there may be additional elements/components/etc. in addition to those listed; the terms "first" and "second" etc. are used as labels only and do not refer to The number of its objects is limited.

Furthermore, the drawings are merely schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities.

IaaS is a cloud computing delivery model that delivers computing resources in the form of virtualized operating systems, workload management software, hardware, network and storage services. It can also include the delivery of operating systems and virtualization technologies to managed resources. A load balancer is a method that distributes network requests to available servers in a server cluster by managing incoming Web (World Wide Web, Global Wide Area Network) data traffic and increasing effective network bandwidth.

In a private cloud environment, there is usually no support for LoadBalancer (used to allocate tasks to less busy servers when the traffic is heavy) provided by the cloud environment. Therefore, Kubernetes clusters cannot use LoadBalancer type services.

Currently, MetalLB is usually used to solve the problem that bare metal Kubernetes clusters cannot use LoadBalancer type services. MetalLB is a load balancer that specifically solves the pain point of being unable to use LoadBalancer type services in bare metal Kubernetes clusters. MetalLB uses a standardized routing protocol so that external services on bare metal Kubernetes clusters also work as well as possible. That is, MetalLB can help you create LoadBalancer-type Kubernetes services in bare metal Kubernetes clusters.

MetalLB supports publishing allocated External IP in two ways, namely Layer2 mode and BGP mode. Among them, in Layer2 mode, the assigned External IP will be bound to a Node node machine network card, and the External IP will be announced through ARP (Address Discovery Protocol) to make it reachable in the local network. This method can be applied to any Ethernet environment. BGP mode can achieve load balancing across multiple nodes and can also implement fine-grained traffic control based on the BGP policy mechanism.

However, MetalLB's Layer 2 mode will have a single point problem. All traffic of the service is forwarded through a single point, and its network bandwidth may also become a bottleneck. In addition, Layer 2 mode requires the cooperation of the ARP client. When a failover occurs, MetalLB will send an ARP packet to announce changes in IP and MAC address mapping, and the address allocation process is slightly cumbersome.

MetalLB's BGP mode cannot handle failover gracefully. When a node holding an External IP goes down, all active connected clients will receive a "communication peer closes the connection" notification. In addition, the BGP router performs a simple hash on the source IP (Internet Protocol), destination IP, and protocol type of the data packet, and determines which node to send it to based on the hash value. Kubernetes cluster nodes are unstable. When a node holding an External IP goes down, most of the active links will become unavailable due to rehash.

In the embodiments of the present disclosure, a load balancing method is first provided, which overcomes, at least to a certain extent, the problem in related technologies that a load balancing device is required to implement the load balancing function.

Figure 1 shows a schematic flowchart of a load balancing method in an embodiment of the present disclosure. The execution subject of the load balancing method may be a server deploying a cloud host management module.

Referring to Figure 1, a load balancing method according to one embodiment of the present disclosure includes the following steps:

Step S110: In response to the load balancing instance creation request, schedule at least two candidate cloud hosts from the cloud host resource pool; the load balancing instance is used to provide external client access;

Step S120, select an idle IP from the IP segment corresponding to each candidate cloud host, and call the domain name server to store the domain names corresponding to at least two idle IPs; wherein, at least two idle IPs correspond to the same domain name;

Step S130, return the domain name to the client, so that the client obtains the target IP corresponding to the domain name from the domain name server; the target IP is any one of at least two idle IPs;

Step S140: Create a load balancing instance based on the target IP so that the client establishes a connection with the target cloud host through the load balancing instance; the target IP is the IP of the load balancing instance, and the target cloud host is the candidate cloud host corresponding to the target IP.

In the technical solution provided by the embodiment shown in Figure 1, on the one hand, in response to a load balancing instance creation request, at least two candidate cloud hosts are scheduled from the cloud host resource pool, and one is selected from the IP segment corresponding to each cloud host. Idle IP, and call the domain name server to store the domain names corresponding to at least two idle IPs, and return the domain names to the client, which can avoid the technical problems of business interruption caused by calling a single cloud host for request forwarding each time when the cloud host fails. , improve the system’s emergency handling capabilities. Furthermore, the target IP is associated with the load balancing instance, so that the client can establish a connection with the target cloud host through the load balancing instance, which provides a new solution that can realize the load balancing function without resorting to load balancing equipment, reducing operation and maintenance. Invest.

The specific implementation process of each step in Figure 1 is described in detail below:

Load Balance refers to balancing and allocating loads (work tasks) to multiple units to improve concurrent processing capabilities. The role of load balancing is: first, to solve concurrency pressure and improve application processing performance (increase throughput and strengthen network processing capabilities); second, to provide failover and achieve high availability; third, to provide high availability by adding or reducing the number of servers. Website scalability (expansibility); fourth, security protection.

In an exemplary embodiment of the present disclosure, a cloud host resource pool may be pre-configured. For example, multiple cloud hosts may be pre-created to form a cloud host resource pool (that is, the cloud host resource pool includes multiple cloud hosts).

For example, after each cloud host is created, an IP segment can be assigned to each cloud host (each IP segment contains multiple IP addresses, for example: from 192.168.0.1 to 192.168.255.255, this is an IP segment), And start the load balancing service on each cloud host. Further, a notification message can be sent to the IaaS so that the IssS points the next hop of the IP segment to the cloud host.

After configuring the cloud host resource pool, if a load balancing instance creation request is received, step S110 can be entered. In step S110, in response to the load balancing instance creation request, at least two candidate cloud hosts are scheduled from the cloud host resource pool. ; The load balancing instance is used to provide external client access.

In this step, when the cloud host management module receives the load balancing instance creation request, it can schedule at least two candidate cloud hosts from the cloud host resource pool. By scheduling at least two candidate cloud hosts, it can be ensured that when any candidate cloud host fails, there will be a backup cloud host to ensure the normal operation of the business.

Among them, the load balancing instance is a running load balancing service that is used to provide external client access.

For example, at least two candidate cloud hosts can be scheduled from the cloud host resource pool according to a random algorithm, or at least two candidate cloud hosts can be scheduled from the cloud host resource pool using a preset round robin scheduling algorithm. The situation is set by oneself, and this disclosure does not impose special restrictions on this.

In step S120, an idle IP is selected from the IP segments corresponding to each candidate cloud host, and the domain name server is called to store domain names corresponding to at least two idle IPs; wherein at least two idle IPs correspond to the same domain name.

In this step, after scheduling at least two candidate cloud hosts from the cloud host resource pool, the cloud host management module may select an idle IP from the IP segment corresponding to each candidate cloud host. Furthermore, the above-mentioned idle IP can be bound to the loopback address of the candidate cloud host. The loopback address is the IP inside the host IP stack. It is mainly used for network software testing and local machine inter-process communication, regardless of the program. , once data is sent using the loopback address, the protocol software returns it immediately without any network transmission.

After selecting the above-mentioned idle IP, the domain name corresponding to the above-mentioned idle IP can be generated according to the domain name rules (all idle IPs correspond to the same domain name). Furthermore, the DNS domain name server can be called to store the mapping relationship between the above-mentioned idle IP and the domain name, and will generate The domain name is returned to the Kubernetes cluster. For example, the selected free IPs are IP_1 and IP_2, and the generated domain name is dn1. Refer to Table 1. Table 1 shows the mapping relationship between the free IPs and domain names stored in the DNS domain name server.

Table 1

Idle IP	domain name
IP_1	dn1
IP_2	dn1

By using the DNS domain name server to store the mapping relationship between free IPs and domain names, users only need to know the easy-to-remember domain name addresses without having to memorize a long list of IPs (which is not intuitive and very inconvenient for users to remember). For example, when a user enters a domain name in a browser, the browser can forward the domain name to the DNS domain name server so that the DNS domain name server determines the IP corresponding to the domain name. When the DNS domain name server returns the IP to the browser, the browser Requests can be made based on IP.

For example, a domain name can be registered for the above-mentioned idle IP based on the following naming rules: $lbname.$lbnamespace.$suffix. Among them, lbname is the load balancing identifier, for example, it can be a serial number or nickname; lbnamespace is the group to which the load balancing belongs. For example, the cloud hosts in the cloud host resource pool can be divided into multiple groups, and each group corresponds to a group identifier, then lbnamespace can be the group identifier to which the candidate cloud host belongs; suffix is the private network subdomain name agreed by the user, which can be set according to the actual situation. This disclosure does not impose special restrictions on this.

In step S130, the domain name is returned to the client, so that the client obtains the target IP corresponding to the domain name from the domain name server; the target IP is any one of at least two idle IPs.

In this step, after registering the domain name, the registered domain name can be returned to the client, and then the client can send a domain name resolution request to the DNS domain name server to obtain the target IP corresponding to the above domain name from the DNS domain name server. For example, after receiving the above domain name, the domain name server can select a target IP from the idle IPs corresponding to the above domain name in a random or polling manner (for example: select IP_1 from IP_1 and IP_2 in Table 1 as the target IP) , and return the above target IP to the above client.

In step S140, a load balancing instance is created based on the target IP, so that the client establishes a connection with the target cloud host through the load balancing instance; the target cloud host is a candidate cloud host corresponding to the target IP.

In this step, after the target IP is determined, a load balancing instance can be created based on the above target IP (since the load balancing instance can provide external client access, when the target IP corresponding to the domain name is determined, it is equivalent to determining the target IP corresponding to the domain name. The load balancing instance corresponding to the domain name), and then the client can establish a connection with the target cloud host corresponding to the target IP through the load balancing instance.

After establishing a connection with the target cloud host, the client can send a service request to the above target IP. After receiving the above-mentioned business request, the target cloud host can forward the above-mentioned business request to the designated back-end service according to the load balancing policy contained in the above-mentioned load balancing instance to implement the load balancing function.

Among them, the above-mentioned business request is issued by the client and is used to request access to the Internet or request network resources. The business request can also be called a service request message, an access request message, an access request packet, a service request packet, and an access request. Requests, etc. can be set according to the actual situation, and this disclosure does not impose special restrictions on this.

For example, the above load balancing strategy can be any of the following:

①Polling: Poll all back-end services to send business requests;

② Least Connections and Least Weighted Connections: Least Connections is an algorithm for communicating with the server that handles the least number of connections (number of sessions) among multiple servers. Even if the processing capabilities of each server are different and the processing volume of each business is also different, the load of the server can be reduced to a certain extent. Weighted Least Connection is an algorithm that adds a weight to each server in the least connection algorithm. This algorithm allocates the number of connections to each server in advance and transfers business requests to the server with the smallest number of connections.

③ Random algorithm: Random, set random probability according to weight. The probability of collision on a section is high, but the larger the call volume, the more even the distribution will be, and the weight will be more even after using the weight according to probability, which is conducive to dynamically adjusting the provider weight.

④IP hashing: An algorithm that uniformly forwards packets from the same sender (or packets sent to the same destination) to the same server by managing the hashes of the sender IP and destination IP. When a client has a series of services to process and must communicate repeatedly with a server, this algorithm can use flow (session) as a unit to ensure that communications from the same client can always be processed in the same server.

⑤URL (Uniform Resource Locator, Uniform Resource Locator) hash: An algorithm that forwards requests sent to the same URL to the same server by managing the hash of the client request URL information.

In an exemplary embodiment of the present disclosure, the above-mentioned target cloud host can also periodically monitor the running status of each back-end service. For example, the CPU (Central Processing Unit, central processing) occupancy rate of each back-end service can be obtained. Parameters such as memory usage and disk usage, among which the above parameters can be obtained through the task manager that comes with the backend service. Furthermore, the running status of the backend service can be determined based on the above parameters. For example, when the CPU occupancy/memory occupancy/disk occupancy is greater than or equal to the preset threshold, it can be determined that the running status of the above backend service is abnormal. When the above CPU occupancy/memory occupancy/disk occupancy are all less than When the above preset threshold is reached, it can be determined that the running status of the above backend service is normal.

Therefore, when the abnormal running status of the above-mentioned back-end service is detected, the data stored on the above-mentioned back-end service can be backed up, and then the back-end service with abnormal running status can be eliminated.

In an exemplary embodiment of the present disclosure, the above-mentioned cloud host management module can also perform fault monitoring on the target cloud host. Optionally, refer to Figure 2. Figure 2 shows the cloud host management module in the embodiment of the present disclosure monitoring the target cloud host. A schematic flow chart for fault monitoring, including steps S201 to S203:

In step S201, the operating status of the target cloud host is periodically monitored.

In this step, the cloud host management module can periodically monitor the operating status of the target cloud host. For example, the cloud host management module can send a heartbeat detection message to the target cloud host every 2 seconds (the specific duration can be set according to the actual situation, and this disclosure does not specifically limit this). The heartbeat response message of the cloud host is used to determine whether the operating status of the target cloud host is normal. For example, if the heartbeat response message returned by the target cloud host is received within a preset time period, it can be determined that the operating status of the target cloud host is normal; otherwise, it can be determined that the operating status of the target cloud host is abnormal.

In step S202, in response to detecting abnormal operating status of the target cloud host, the target cloud host is marked as unschedulable.

In this step, if the abnormal running status of the target cloud host is detected, it can be determined whether the IP segments bound to the target cloud host have been released (that is, none of the multiple IPs included in the IP segment at the current moment have been released). occupied), if so, the following step S203 can be performed, otherwise, the waiting logic can be executed, and after each IP in the above IP segment is released, the following step S203 can be performed.

In step S203, a backup cloud host of the target cloud host is selected from the cloud host resource pool to replace the target cloud host with the backup cloud host.

In this step, the cloud host management module may select a backup cloud host of the target cloud host from the cloud host resource pool to replace the target cloud host through the backup cloud host.

For example, the cloud host management module can select the above-mentioned backup cloud host from the cloud host resource pool according to the following conditions: first, filter out all cloud hosts in normal operating status (hereinafter referred to as the first set), and then determine the first set of cloud hosts. The number of idle IPs of each cloud host is selected from the first set to select cloud hosts whose number of idle IPs meets the preset number conditions (hereinafter referred to as the second set). Furthermore, a cloud host can be randomly selected from the second set. As the above backup cloud host.

After selecting the above-mentioned backup cloud host, you can first mark the backup cloud host as unschedulable, that is, lock the backup cloud host to avoid the backup cloud host from being used during the preparation period before the backup cloud host officially takes over the target cloud host. The host's free IP is occupied, resulting in a shortage of free IPs.

After selecting the above-mentioned backup cloud host and marking the backup cloud host as unschedulable, the above-mentioned cloud host management module can also update the status of the backup cloud host after migrating the data on the target cloud host to the backup cloud host, thereby controlling The backup cloud host takes over the above target cloud host. Optionally, refer to Figure 3. Figure 3 shows a schematic flow chart of how to update the status of the backup cloud host so that the backup cloud host takes over the target cloud host in an embodiment of the present disclosure, including steps S301 to S303:

In step S301, data stored on the target cloud host is migrated to the backup cloud host.

In this step, the data stored on the target cloud host can be migrated to the above backup cloud host.

In step S302, select an idle IP from the IP segment associated with the backup cloud host, and send the idle IP to the domain name server, so that the domain name server changes the mapping relationship between the idle IP of the target cloud host and the domain name to that of the backup cloud host. Mapping relationship between IP and domain name.

In this step, you can randomly select an idle IP (for example: IP_3) from the above-mentioned backup cloud host, and send the idle IP to the domain name server, so that the domain name server changes the mapping relationship between the idle IP and the domain name of the target cloud host to the backup one. The mapping relationship between the free IP of the cloud host and the domain name.

Therefore, referring to Table 2, the mapping relationship between the idle IP and the domain name stored in the above domain name server changes to:

Table 2

Idle IP	domain name
IP_2	dn1
IP_3	dn1

Therefore, when the target cloud host fails, the present disclosure can take over the target cloud host through the backup cloud host without the client being aware of it, so that the client can still connect to the backup cloud host through the domain name, that is, the client's access The form remains unchanged.

In step S303, the backup cloud host is marked as a schedulable state so that the backup cloud host can take over the target cloud host.

In this step, after the domain name server updates the mapping relationship between the idle IP and the domain name stored therein, the backup cloud host can be marked as schedulable, so that the backup cloud host can take over the target cloud host.

For example, after marking the backup cloud host as schedulable, an alarm notification can be sent to the manager of the application. For example, the alarm content can be "The target cloud host has failed, and the backup cloud host has taken over the target." Cloud Host" to explain the fault and fault response to the above-mentioned managers. It should be noted that the specific alarm content can be set according to the actual situation, and this disclosure does not impose special restrictions on this.

Referring to Figure 4, Figure 4 shows a schematic flow chart of adding a cloud host to a cloud host resource pool in an embodiment of the present disclosure, including steps S401 to S405:

In step S401, the cloud host management module creates a cloud host;

In step S402, IPAM (IP Management Module) obtains it from Etcd (Etcd is a highly consistent distributed key-value store that provides a reliable way to store data that needs to be accessed by a distributed system or machine cluster) The unbound IP segment is sent to the cloud host management module, so that the cloud host management module binds an IP segment to the cloud host; and stores the corresponding relationship between the cloud host and the IP segment in Etcd;

In step S403, the cloud host management module initializes the load balancing service;

In step S404, the cloud host management module adds the created cloud host to the cloud host resource pool;

In step S405, the cloud host management module sends a notification message to IaaS, so that IaaS uses the newly created cloud host as the next hop of the IP segment to which it is bound.

After the cloud host is created, if the operating status of the cloud host is abnormal, the cloud host can be removed from the cloud host resource pool. Referring to Figure 5, Figure 5 shows a schematic flowchart of removing faulty cloud hosts from the cloud host resource pool in an embodiment of the present disclosure, including steps S501 to S505:

In step S501, start;

In step S502, the cloud host management module sends a message to IPAM, so that IPAM reads Etcd to determine whether the IP segment bound to the cloud host is released; if not, the waiting logic is executed;

In step S503, if the cloud host management module determines that the IP segment is released, it terminates the operation of the cloud host;

In step S504, the cloud host management module terminates the load balancing service;

In step S505, the cloud host management module updates the cloud host to an unschedulable state;

In step S506, the cloud host management module sends a notification message to IaaS, so that IaaS revokes the next hop configuration of the IP segment bound to the cloud host.

Based on the above implementation, the present disclosure can quickly remove a certain cloud host when it fails, and use the backup cloud host to replace it, ensuring stable load balancing performance.

Referring to Figure 6, Figure 6 shows a schematic flowchart of creating a load balancing instance in an embodiment of the present disclosure, including steps S601 to S606:

In step S601, the scheduler traverses the cloud host resource pool (sends a message to the cloud host management module so that the cloud host management module obtains cloud host information);

In step S602, the scheduler selects at least two cloud hosts with the most idle IPs and two candidate cloud hosts;

In step S603, the scheduler selects an idle IP from the IP segments bound to each cloud host (sends a message to the cloud host management module to determine the unoccupied idle IP);

In step S604, the scheduler sends a status update notification to the cloud host management module, so that the cloud host management module updates the status of the cloud host to be schedulable;

In step S605, the scheduler creates a load balancing instance and stores it in the load balancing instance pool;

In step S606, the scheduler calls the DNS domain name server to register a domain name for the idle IP.

After creating a load balancing instance, you can add the load balancing instance, bind the idle IP to the local loopback address, and synchronize the load balancing policy associated with the load balancing instance to the load balancing configuration, and then the load balancing service The management module updates the load balancing configuration to reload the load balancing service.

Referring to Figure 7, Figure 7 shows a schematic flowchart of deleting a load balancing instance in an embodiment of the present disclosure, including steps S701 to S705:

In step S701, the scheduler updates the status of the load balancing instance to delete;

In step S702, the scheduler cancels the domain name;

In step S703, the scheduler releases what is bound to the cloud host (sends a message to IPAM to cause IPAM to release what is bound to the cloud host);

In step S704, the scheduler deletes the load balancing instance;

In step S705, the scheduler sends a status update notification to the cloud host management module, so that the cloud host management module updates the status of the cloud host to unschedulable.

After deleting the load balancing instance, the load balancing service management module can unbind the idle IP from the local loopback address, and then the load balancing service management module can update the load balancing configuration to reload the load balancing service.

Referring to Figure 8, Figure 8 shows a schematic flow chart of how to manage a cloud host in an embodiment of the present disclosure, including steps S801 to S810:

In step S801, the cloud host management module periodically traverses each cloud host;

In step S802, the cloud host management module detects whether each cloud host is in a normal state;

If the cloud host is in a normal state, proceed to step S803, and the cloud host management module detects whether the load balancing service is in a normal state; if the load balancing service is in a normal state, proceed to step S804, with no action; otherwise, proceed to step S805, Trigger the mechanism to eliminate cloud hosts;

In step S806, the cloud host management module traverses the cloud hosts and obtains cloud host information through the scheduler to determine the backup cloud host of the faulty cloud host;

In step S807, the cloud host management module locks the backup cloud host;

In step S808, the cloud host management module migrates the load balancing instance corresponding to the failed target cloud host through the scheduler;

In step S809, the cloud host management module marks the backup cloud host as schedulable;

In step S810, the cloud host management module sends an alarm notification to relevant managers. The alarm notification includes information about the faulty cloud host, the backup cloud host, and the backup load balancing instance.

Based on the above technical solution, the present disclosure provides a universally applicable load balancing solution. By preconfiguring a cloud host resource pool, a kubernetes cluster is deployed in the cloud host resource pool, and the client can communicate with the cloud host resource pool through the load balancing instance. Establish a connection with the target cloud host, and then forward the business request through the target cloud host. It can be seen that this solution provides a new solution that can access the kubernetes cluster from the outside to achieve the load balancing function without resorting to relevant load balancing equipment. It is suitable for the network environment of all IaaS vendors and reduces operation and maintenance investment.

The present disclosure also provides a load balancing system. Figure 9 shows a schematic structural diagram of the load balancing system in an embodiment of the present disclosure. As shown in Figure 9, the load balancing system 900 may include a cloud host management module 910 and a cloud host resource pool 920. , the cloud host resource pool includes multiple cloud hosts 9201, and each cloud host is associated with a pre-allocated IP segment. in:

The cloud host management module 910 is used to respond to a load balancing instance creation request and schedule at least two candidate cloud hosts from the cloud host resource pool; the load balancing instance is used to provide external client access; and the IP corresponding to each candidate cloud host is obtained from Select an idle IP in the segment, call the domain name server to store the domain names corresponding to at least two idle IPs; among them, at least two idle IPs correspond to the same domain name; return the domain name to the client, so that the client obtains the domain name corresponding to the domain name server The target IP; the target IP is any one of at least two idle IPs; create a load balancing instance based on the target IP so that the client can establish a connection with the target cloud host through the load balancing instance; the target cloud host is the candidate cloud corresponding to the target IP host;

Target cloud host 9201 is used to receive business requests sent by the client through the load balancing instance; based on the load balancing policy associated with the load balancing instance, forward the business request to the specified back-end service.

In an exemplary embodiment of the present disclosure, the cloud host management module 910 is configured as:

Periodically monitor the operating status of the target cloud host;

In response to detecting abnormal operating status of the target cloud host, mark the target cloud host as unschedulable;

Select the backup cloud host of the target cloud host from the cloud host resource pool to take over the target cloud host through the backup cloud host.

In an exemplary embodiment of the present disclosure, the backup cloud host satisfies the following conditions:

The running status of the backup cloud host is normal, and the number of idle IPs in the backup cloud host meets the preset number conditions.

In an exemplary embodiment of the present disclosure, the cloud host management module 910 is configured as:

Migrate the data stored on the target cloud host to the backup cloud host;

Select an idle IP from the IP segment associated with the backup cloud host, and send the idle IP to the domain name server, so that the domain name server changes the mapping relationship between the idle IP and the domain name of the target cloud host to the mapping relationship between the idle IP and the domain name of the backup cloud host. ;

Mark the standby cloud host as schedulable so that the standby cloud host can take over the target cloud host.

In an exemplary embodiment of the present disclosure, the target cloud host 9201 is configured as:

Periodically monitor the running status of back-end services;

In response to detecting abnormal running status of the backend service, the data stored on the backend service is deleted and the backend service is marked as unschedulable.

In an exemplary embodiment of the present disclosure, the running status of the backend service is determined by one or more of the following parameters: CPU occupancy, memory occupancy, and disk occupancy.

The specific details of each module in the above load balancing system have been described in detail in the corresponding load balancing method, so they will not be described again here.

It should be noted that although several modules or units of equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to embodiments of the present disclosure, the features and functions of two or more modules or units described above may be embodied in one module or unit. Conversely, the features and functions of one module or unit described above may be further divided into being embodied by multiple modules or units.

Furthermore, although various steps of the methods of the present disclosure are depicted in the drawings in a specific order, this does not require or imply that the steps must be performed in that specific order, or that all of the illustrated steps must be performed to achieve the desired results. result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step for execution, and/or one step may be decomposed into multiple steps for execution, etc.

Through the above description of the embodiments, those skilled in the art can easily understand that the example embodiments described here can be implemented by software, or can be implemented by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, U disk, mobile hard disk, etc.) or on the network , including several instructions to cause a computing device (which may be a personal computer, a server, a mobile terminal, a network device, etc.) to execute a method according to an embodiment of the present disclosure.

This application also provides a computer-readable storage medium. The computer-readable storage medium may be included in the electronic device described in the above embodiments; it may also exist independently without being assembled into the electronic device.

The computer-readable storage medium may be, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or any combination thereof. More specific examples of computer readable storage media may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard drive, random access memory (RAM), read only memory (ROM), removable Programmed read-only memory (EPROM or flash memory), fiber optics, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above. In this disclosure, a computer-readable storage medium may be any tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device.

Computer-readable storage media may transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

The computer-readable storage medium carries one or more programs. When the one or more programs are executed by an electronic device, the electronic device implements the method described in the above embodiments.

In addition, an electronic device capable of implementing the above method is also provided in an embodiment of the present disclosure.

Those skilled in the art will understand that various aspects of the present disclosure may be implemented as systems, methods, or program products. Therefore, various aspects of the present disclosure may be embodied in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which may be collectively referred to herein as "Circuit", "Module" or "System".

An electronic device 1000 according to this embodiment of the present disclosure is described below with reference to FIG. 10 . The electronic device 1000 shown in FIG. 10 is only an example and should not bring any limitations to the functions and usage scope of the embodiments of the present disclosure.

As shown in Figure 10, electronic device 1000 is embodied in the form of a general computing device. The components of the electronic device 1000 may include, but are not limited to: the above-mentioned at least one processing unit 1010, the above-mentioned at least one storage unit 1020, a bus 1030 connecting different system components (including the storage unit 1020 and the processing unit 1010), and the display unit 1040.

The storage unit stores program code, which can be executed by the processing unit 1010, so that the processing unit 1010 performs the steps according to various exemplary embodiments of the present disclosure described in the "Exemplary Method" section of this specification. For example, the processing unit 1010 can perform as shown in Figure 1: Step S110, in response to a load balancing instance creation request, schedule at least two candidate cloud hosts from the cloud host resource pool; the load balancing instance is used to provide external client access Entrance; Step S120, select an idle IP from the IP segment corresponding to each candidate cloud host, and call the domain name server to store the domain names corresponding to at least two idle IPs; among which, at least two idle IPs correspond to the same domain name; Step S130, return the domain name to the client, so that the client obtains the target IP corresponding to the domain name from the domain name server; the target IP is any one of at least two idle IPs; step S140, create a load balancing instance based on the target IP so that the client can pass the load The balancing instance establishes a connection with the target cloud host; the target IP is the IP corresponding to the load balancing instance, and the target cloud host is the candidate cloud host corresponding to the target IP.

The storage unit 1020 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 10201 and/or a cache storage unit 10202, and may further include a read-only storage unit (ROM) 10203.

Storage unit 1020 may also include a program/utility 10204 having a set of (at least one) program modules 10205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, Each of these examples, or some combination, may include the implementation of a network environment.

Bus 1030 may be a local area representing one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, a graphics acceleration port, a processing unit, or using any of a variety of bus structures. bus.

Electronic device 1000 may also communicate with one or more external devices 1100 (e.g., keyboard, pointing device, Bluetooth device, etc.), may also communicate with one or more devices that enable a user to interact with electronic device 1000, and/or with Any device that enables the electronic device 1000 to communicate with one or more other computing devices (eg, router, modem, etc.). This communication may occur through input/output (I/O) interface 1050. Furthermore, the electronic device 1000 may also communicate with one or more networks (eg, a local area network (LAN), a wide area network (WAN), and/or a public network, such as the Internet) through the network adapter 1060. As shown, network adapter 1060 communicates with other modules of electronic device 1000 via bus 1030. It should be understood that, although not shown in the figures, other hardware and/or software modules may be used in conjunction with electronic device 1000, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives And data backup storage system, etc.

Other embodiments of the disclosure will be readily apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure that follow the general principles of the disclosure and include common knowledge or customary technical means in the technical field that are not disclosed in the disclosure. . It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

Claims (10)

1. A load balancing method that pre-configures a cloud host resource pool. The cloud host resource pool includes multiple cloud hosts, and each cloud host is associated with a pre-allocated IP segment. The method includes:

   In response to the load balancing instance creation request, schedule at least two candidate cloud hosts from the cloud host resource pool; the load balancing instance is used to provide external access to the client;

   Select an idle IP from the IP segment corresponding to each candidate cloud host, and call the domain name server to store the domain names corresponding to at least two of the idle IPs; wherein, at least two of the idle IPs correspond to the same domain name;

   Return the domain name to the client, so that the client obtains the target IP corresponding to the domain name from the domain name server; the target IP is any one of the at least two idle IPs;

   Create the load balancing instance based on the target IP, so that the client establishes a connection with the target cloud host through the load balancing instance; the target IP is the IP corresponding to the load balancing instance, and the target cloud host is the candidate cloud host corresponding to the target IP.
2. The method of claim 1, further comprising:

   Periodically monitor the operating status of the target cloud host;

   In response to detecting abnormal operating status of the target cloud host, marking the target cloud host as unschedulable;

   Select a backup cloud host of the target cloud host from the cloud host resource pool to take over the target cloud host through the backup cloud host.
3. The method according to claim 2, wherein the backup cloud host meets the following conditions:

   The running status of the backup cloud host is normal, and the number of idle IPs in the backup cloud host meets the preset number condition.
4. The method of claim 3, further comprising:

   Migrate the data stored on the target cloud host to the backup cloud host;

   Select an idle IP from the IP segment associated with the backup cloud host, and send the idle IP to the domain name server, so that the domain name server changes the mapping relationship between the idle IP of the target cloud host and the domain name to the desired value. The mapping relationship between the idle IP of the backup cloud host and the domain name;

   Mark the backup cloud host as a schedulable state so that the backup cloud host can take over the target cloud host.
5. A load balancing method, applied to the target cloud host, the method includes:

   Receive business requests sent by clients through pre-created load balancing instances;

   Based on the load balancing policy associated with the load balancing instance, the service request is forwarded to the designated backend service.
6. The method of claim 5, further comprising:

   Periodically monitor the running status of the backend service;

   In response to monitoring the abnormal running status of the back-end service, delete the data stored on the back-end service and mark the back-end service as unschedulable.
7. The method according to claim 6, wherein the running status of the back-end service is determined by one or more of the following parameters: CPU occupancy, memory occupancy, and disk occupancy.
8. A load balancing system including:

   Cloud host management module and cloud host resource pool; the cloud host management module is used to manage a preconfigured cloud host resource pool. The cloud host resource pool includes multiple cloud hosts, and each cloud host is associated with a pre-allocated IP segment;

   The cloud host management module is used to dispatch at least two candidate cloud hosts from the cloud host resource pool in response to a load balancing instance creation request; the load balancing instance is used to provide external client access; from each of the candidates Select an idle IP from the IP segment corresponding to the cloud host, call the domain name server to store at least two domain names corresponding to the idle IP; wherein, at least two of the idle IPs correspond to the same domain name; return the domain name to the client to The client obtains the target IP corresponding to the domain name from the domain name server; the target IP is any one of the at least two idle IPs; and the load balancing instance is created based on the target IP so that all The client establishes a connection with the target cloud host through the load balancing instance; the target IP is the IP corresponding to the load balancing instance, and the target cloud host is the candidate cloud host corresponding to the target IP;

   The target cloud host is used to receive the service request sent by the client through the load balancing instance; and forward the service request to the designated back-end service based on the load balancing policy associated with the load balancing instance.
9. A computer storage medium on which a computer program is stored. When the computer program is executed by a processor, the load balancing method according to any one of claims 1 to 8 is implemented.
10. An electronic device including:

    processor; and

    memory for storing executable instructions for the processor;

    Wherein, the processor is configured to execute the load balancing method according to any one of claims 1 to 8 by executing the executable instructions.

Images