WO2015192563A1 - Method and device for implementing load balancing and load balancing service system - Google Patents

Abstract

Provided are a method and device for implementing load balancing and load balancing service system, solving the problem of low performance of load balancing service. The method comprises: creating a virtual network system and a load balancing module according to a creation request, and connecting the created load balancing module to the virtual network system, the created virtual network system comprising at least one virtual machine; storing in the load balancing module the network information of the virtual machine in different network segments with the load balancing module; when a received data packet requiring load balancing is present in a virtual load balancing processing network, forwarding the data packet to the virtual machine by the load balancing module for processing. The solution of the present application can solve the problem of low performance of load balancing service.

Description

Method, device and load balancing service system for implementing load balancing Technical field

The present invention relates to the field of cloud computing, and in particular, to a method, an apparatus, and a load balancing service system for implementing load balancing.

Background technique

With the development of the Internet, the use of cloud computing is more and more extensive. Cloud computing is an Internet-based computing method. In this way, software and hardware resources and network information can be shared according to the needs of computers or other devices. .

There are three service modes in existing cloud computing:

Software-as-a-Service (SaaS for short): Consumers use applications, but do not control the operating system, hardware, or operational network infrastructure. It is the basis of a service concept. The software service provider provides customer service with the concept of leasing instead of purchasing. The more common mode is to provide a set of account passwords.

Platform-as-a-Service (PaaS for short): Consumers use host operations applications. Consumers control the environment in which the application runs (and also have control over the host), but do not control the operating system, hardware, or operational network infrastructure. The platform is usually the application infrastructure.

Infrastructure-as-a-Service (IaaS): The service provided to consumers is the utilization of all facilities, including processing, storage, network and other basic computing resources, users can deploy and run arbitrary Software, including operating systems and applications. Consumers do not manage or control any cloud computing infrastructure, but can control the choice of operating system, storage space, deployed applications, and possibly control of restricted network components (eg, firewalls, load balancers, etc.).

FIG. 1 is a common cloud computing architecture mode in which a user creates his own cloud environment through a console, for example, a virtual machine can be created, a network is created, a mirror template is created, and the like; in the prior art, Cloud computing also expands other services, such as load balancing services, firewalls, VPNs (Virtual Private Network), and more.

Among the above-mentioned extended services, the load balancing service is a common network device, and its main function is to provide an effective and transparent method for expanding the bandwidth of the network device and the server, increasing the throughput, strengthening the network data processing capability, and improving the network. Flexibility and usability, etc. Currently, load balancing services in cloud computing are usually set up. In a virtual network of the user, that is, a virtual network belonging to the user belongs to the same layer 2 network. When the load balancing service is applied to the network device in the virtual machine network, the data can be directly forwarded without addressing. For other virtual networks that do not belong to the same Layer 2 network as the load balancing service, when load balancing services the network devices in the virtual network, the data needs to be forwarded through route addressing. In this case, when a large number of cloud computing services need to be processed, it will have a large impact on the cloud computing and load balancing system, so the performance of the load balancing processing service is reduced.

Therefore, how to improve the service performance of load balancing services in cloud computing is a key issue that needs to be solved.

Summary of the invention

The technical problem to be solved by the present invention is to provide a method, a device and a load balancing service system for implementing load balancing, which can solve the problem of low performance of the load balancing service.

In order to solve the above problem, the embodiment of the present application provides a method for implementing load balancing, including:

Creating a virtual network system and a load balancing module according to the creation request, and connecting the load balancing module to the virtual network system, where the virtual network system includes at least one virtual machine;

Storing the network information of the virtual machine that is not on the same network segment as the load balancing module in the load balancing module, and configuring the virtual machine as an application service node at the back end of the load balancing module;

The data packet requiring load balancing processing is forwarded to the virtual machine for processing by the load balancing module.

In an embodiment of the present invention, the creating a virtual network system and a load balancing module, and accessing the load balancing module to the virtual network system is specifically:

Creating a virtual subnet, the virtual subnet comprising: a first level virtual subnet and/or a multi-level virtual subnet; the first level virtual subnet comprising at least one virtual machine accessing the virtual network system through the virtual switch; The multi-level virtual subnet includes at least one virtual machine that accesses the virtual network system through the virtual switch; the virtual switch in the multi-level virtual subnet accesses the virtual network step by step through the virtual switch in the other at least one level one virtual subnet. system;

Create a load balancing module and directly connect the load balancing module to the virtual network system.

In an embodiment of the present invention, the creating a virtual subnet is specifically: creating at least one level one virtual subnet, and directly connecting the virtual switch in the level one virtual subnet to the virtual network system.

In an embodiment of the present invention, the first-level virtual sub-network implements network isolation in any one of the following manners in the virtual network system: a virtual local area network (VLAN), and a general routing package. (Generic Routing Encapsulation, GRE for short), Virtual eXtensible Local Area Network (VXLAN); the network information corresponding to the virtual machine in the first-level virtual subnet is any one of the following information: VLAN information, GRE information, and VXLAN information.

In an embodiment of the present invention, the network information of the virtual machine that is not in the same network segment as the load balancing module is stored in the load balancing module, specifically:

Obtaining at least one network information of a virtual machine that is not in the same network segment as the load balancing module;

And creating virtual network device or flow table information corresponding to each virtual machine according to the acquired network information of the virtual machine, and saving the information in the load balancing module.

In order to solve the above problem, the present application further provides an apparatus for implementing load balancing, including: a creating module, a processing module, and a forwarding module;

The creating module is configured to separately create a virtual network system and a load balancing module according to the creation request, and connect the load balancing module to the virtual network system, where the virtual network system includes at least one virtual machine;

The processing module is configured to store network information of a virtual machine that is not in the same network segment as the load balancing module in a load balancing module, and configure the virtual machine as an application service node at a back end of the load balancing module;

The forwarding module is configured to forward the data packet requiring load balancing processing to the virtual machine through the load balancing module for processing.

In an embodiment of the invention, the creating module includes a subnet creating unit and a load balancing creating unit;

The subnet creating unit is configured to create a virtual subnet, where the virtual subnet includes: a first level virtual subnet and/or a multi-level virtual subnet; and the first level virtual subnet includes at least one virtual access through a virtual switch. a virtual machine in the network system; the virtual switch in the multi-level virtual subnet accesses the virtual network system step by step through virtual switches in the other at least one level one virtual subnet;

The load balancing creation unit is configured to create a load balancing module and directly connect the load balancing module to the virtual network system.

In an embodiment of the present invention, the subnet creating unit creates a virtual subnet specifically: creating at least one level one virtual subnet, and directly connecting the virtual switch in the first level virtual subnet to the virtual network system.

In an embodiment of the present invention, the virtual subnet created by the subnet creating unit implements network isolation in any one of the following manners in the virtual network system: VLAN, GRE, VXLAN; corresponding to the first level The network information of the virtual machine in the virtual subnet is any one of the following information: VLAN information, GRE information, and VXLAN information.

In an embodiment of the invention, the processing module further includes an obtaining unit and a creating unit;

The obtaining unit is configured to acquire network information of at least one virtual machine that is not in the same network segment as the load balancing module;

The creating unit is configured to create virtual network device or flow table information corresponding to each virtual machine according to the acquired network information of the virtual machine, and save the information in the load balancing module.

In order to solve the above problem, the present application further provides a load balancing service system, including: a virtual network system and a load balancing module;

The load balancing module accesses the virtual network system and provides a load balancing service for at least one virtual machine in the virtual network system;

The load balancing module is configured to store network information of a virtual machine that is not in the same network segment as the virtual machine, and configure the virtual machine as its backend service node, and to pass the data packet requiring load balancing processing The load balancing module forwards to the virtual machine for processing.

The beneficial effects of the invention are:

The invention provides a method, a device and a load balancing service system for implementing load balancing, which solves the problem that the performance of the load balancing service is low. The content of the present invention is to create a virtual network system and a load balancing module according to the creation request, and the created load balancing module is connected to the virtual network system, and the created virtual network system includes at least one virtual machine; The network information of the virtual machine of a network segment is stored in the load balancing module; when the received data packet requiring the equalization processing exists in the virtual load balancing processing network, the data packet is forwarded to the storage device through the load balancing module. Processing is performed in the virtual machine. It can be seen that, compared with the prior art, the present application configures the load balancing module and the virtual machine into the virtual load balancing processing network, so that the load balancing module and any one of the virtual machines are in the same layer 2 network environment, when needed. When forwarding data packets, routing addressing is not required, and the data can be directly forwarded directly through the load balancing module, which enhances the performance of the load balancing processing data packet. Therefore, the solution of the load balancing service can be solved by adopting the solution in itself.

DRAWINGS

1 is a network architecture of a cloud computing system in the prior art provided by the present invention;

2 is a flowchart of a method for implementing load balancing according to a first embodiment of the present invention;

3 is a schematic diagram of a cloud computing network architecture composed of a virtual network system and a load balancing module when the virtual subnet is a primary virtual subnet provided by the first embodiment of the present invention;

4 is a schematic structural diagram of a device module for implementing load balancing according to a second embodiment of the present invention;

5 is a schematic structural diagram of a load balancing service system provided in a third embodiment of the present invention;

FIG. 6 is a flowchart of a method for implementing load balancing creation in cloud computing according to a fourth embodiment of the present invention.

detailed description

The embodiment relates to a method for implementing load balancing in cloud computing. Further, the embodiment actually relates to a method for improving load balancing service performance in cloud computing; it should be understood that although the embodiment disclosed includes improving load in cloud computing Detailed description of the balanced performance, but the technical solutions described therein

The implementation is not limited to the cloud computing environment. Rather, embodiments of the invention can be implemented in conjunction with any other type of computing environment now known or later developed.

Cloud computing is a service delivery model for convenient, on-demand network access to shared configurable computing resource pools. A configurable computing resource is a resource that can be quickly deployed and released with minimal administrative cost or minimal interaction with a service provider, such as network, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and service. Such a cloud model may include at least five features, at least three service models (already described in the background), and at least four deployment models.

Features include:

On-demand self-service: Cloud consumers can unilaterally and automatically deploy computing power such as server time and network storage on-demand, without the need for human interaction with service providers.

Extensive network access: Computational capabilities can be obtained on the network through standard mechanisms that facilitate the adoption of different types of thin client platforms or thick client platforms (eg mobile phones, laptops, personal digital assistant PDAs) ) the use of the cloud.

Resource pool: The provider's computing resources are grouped into resource pools and serve multiple consumers through a multi-tenant model, where different physical and virtual resources are dynamically allocated and reallocated as needed. In general, consumers cannot control or even know the exact location of the resources provided, but can specify locations (such as countries, states, or data centers) at a higher level of abstraction and are therefore location-independent.

Rapid resiliency: Ability to deploy computing power quickly, flexibly (and sometimes automatically) for rapid expansion and rapid release for rapid shrinking. In the consumer's view, the computing power available for deployment tends to be infinite and can capture any amount of computing power at any time.

Measurable Services: Cloud systems automatically control and optimize resource utility by leveraging some level of abstraction that is appropriate for the type of service (eg, storage, processing, bandwidth, and active consumer accounts). Resource usage can be monitored, controlled and reported to provide transparency to both providers and consumers of the services used.

The deployment model is as follows:

Private cloud: The cloud infrastructure runs separately for an organization. The cloud infrastructure can be managed by the organization or a third party and can exist inside or outside the organization.

Community Cloud: The cloud infrastructure is shared by several organizations and supports a specific community of shared interests such as mission mandates, security requirements, policies, and compliance considerations. The community cloud can be managed by multiple organizations or third parties within the community and can exist inside or outside the community.

Public cloud: The cloud infrastructure is provided to the public or large industry groups and is owned by organizations that sell cloud services.

Hybrid Cloud: A cloud infrastructure consists of two or more deployment models of clouds (private, community, or public) that remain unique entities, but with standardized or proprietary technologies that enable data and applications to be ported. (For example, cloud burst traffic sharing technology for load balancing between clouds) is bundled together.

The cloud computing environment is service-oriented, with features focused on statelessness, low-complexity, modularity, and semantic interoperability. At the heart of cloud computing is the infrastructure that encompasses the network of interconnected nodes.

In this embodiment, the virtual network system and the load balancing module are created according to the creation request, and the created load balancing module is connected to the virtual network system, and the created virtual network system includes at least one virtual machine; The network information of the virtual machine of a network segment is stored in the load balancing module, and the virtual machine is configured as an application service node at the back end of the load balancing module; when the received data packet requiring equalization processing is received, the data packet is passed. The load balancing module forwards to the virtual machine for processing. In an embodiment of the present invention, a specific manner of creating a virtual network system and a load balancing module is: creating a virtual subnet, where the virtual subnet includes: a first-level virtual subnet and/or a multi-level virtual subnet; Level virtual subnets include at least one connected through a virtual switch a virtual machine that enters a virtual network; the virtual switch in the multi-level virtual subnet accesses the virtual network step by step through virtual switches in at least one first-level virtual subnet; creates a load balancing module, and directly connects the load balancing module In the cloud computing network architecture, the virtual machine that is not in the same network segment as the load balancing module is configured as the application service node of the load balancing module backend, and the purpose of constructing the cloud computing network architecture is to Even if the load balancing module and any one of the virtual machines are not in the same network segment (Layer 2 network), the packet forwarding can be performed without routing. The solution in this embodiment is used to avoid the problem that the data forwarding is performed when the virtual machine that needs to be served is not in the same network segment as the load balancing system in the prior art. Therefore, in this embodiment, The method can solve the problem of low performance of the load balancing service.

In order to better understand the present embodiment, the content of the present application will be further described below in conjunction with specific embodiments:

In the present embodiment, the terminology used is for the purpose of describing particular embodiments, and is not intended to limit the invention. The singular forms "a", "the", and "the" Furthermore, the use of the terms "a", "an", etc. does not denote a limitation of the quantity, but rather the presence of at least one of the referenced items. It is further understood that the term "comprising", used in the <RTI ID=0.0> </RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; The presence or addition of a combination thereof.

Embodiment 1:

FIG. 2 is a flowchart of a method for implementing load balancing according to an embodiment of the present disclosure, including:

Step 101: Create a virtual network system and a load balancing module including a virtual machine according to the creation request.

Step 102: The network information of the virtual machine that is not in the same network segment as the load balancing module is stored in the load balancing module, and the virtual machine is configured as an application service node at the back end of the load balancing module.

Step 103: The data packet to be forwarded is forwarded to the required virtual machine through the load balancing module for processing.

In this embodiment, the entity that initiates the creation request includes a user, and the user initiates a creation request through the console. After receiving the creation request initiated by the user, the system creates a virtual network architecture required by the user according to the content of the creation request. In the example, the virtual network system created according to the user-initiated virtual network system creation request includes at least: a virtual network, a gateway, a route, a virtual machine, and the like, and the created gateway, the route, and the virtual machine are added to the virtual network; The load balancing service requests to create a load balancing module and will create The virtual configuration is the application service node of the load balancing module backend. In this embodiment, the created load balancing module can provide a higher performance load balancing service for the user.

In the step 101, the virtual network system and the load balancing module are separately created according to the user request, and the created virtual network system includes at least one virtual machine. In this embodiment, when the virtual machine is created, the virtual machine is more than two. The VMs can belong to the same Layer 2 network or belong to different Layer 2 networks. In this embodiment, after the virtual network system and the load balancing module are created, the created load balancing module is connected to the created virtual network system to ensure that the load balancing module can be interconnected with the virtual network system network.

In this embodiment, the virtual network system refers to a virtual layer 3 network provided for users in a cloud computing service, in which a user can create a network device such as a subnet, a route, a gateway, or the like, or Different networks perform interworking settings, and the performance of the network is exactly the same as the physical network.

In step 102, acquiring network information of the virtual machine that is not in the same network segment as the load balancing module (the network information is previously configured network information), and storing the network information in the load balancing module, and simultaneously The virtual machine is configured as an application service node at the back end of the load balancing module. In this embodiment, the virtual machine and the load balancing module are not in the same network segment, and the virtual and load balancing modules are not in the same Layer 2 network.

In this embodiment, the load balancing module can directly communicate with the virtual machine, and the load balancing module does not need to be in a network segment with the virtual machine.

Preferably, in this embodiment, when the load balancing module is created according to the creation request, the load balancing algorithm, the health check and the session hold function of the load balancing module are further included:

The load balancing algorithm includes a general load balancing algorithm such as polling, ratio, priority, minimum connection number, and fastest response time, and includes a hash algorithm, policy-based load balancing, packet-based content distribution, and the like. And other advanced algorithms.

The health check refers to checking the available status of various services opened by the server. The load balancing device is generally configured with various health check methods, such as Ping, Transmission Control Protocol (TCP), User Data Protocol (UDP), and Hypertext Transfer Protocol (HTTP). ), File Transfer Protocol (FTP), Domain Name System (DNS), etc. Ping belongs to the third-level health check, which is used to check the connectivity of the server IP. TCP/UDP belongs to the fourth-level health check, which is used to check the UP/DOWN of the service port. If you want to check the more accurate, you need to use Go to a 7-layer based health check, such as creating an HTTP health check, getting a page back, and checking if the page content contains a specified string, if it is included, then The service is UP. If the page is not included or cannot be retrieved, the server's web service is considered to be unavailable (DOWN).

The session retention refers to maintaining the continuity and consistency of the session. Since it is difficult for the servers to synchronize the user access information in real time, it is required to keep the user's front and back access sessions to a server for processing. The load balancing device generally configures some session-maintaining options by default, such as session keeping of the source address, keeping the cookie session, etc., depending on different applications, configuring different session holdings, otherwise the load may be unbalanced or even accessed abnormally.

In step 103, when receiving the data packet requiring load balancing processing, the data packet is preferably forwarded to the virtual machine for processing according to the network information in the load balancing module.

Preferably, in this embodiment, the load balancing service provided by the load balancing module may be implemented by using hardware or by software. The specific implementation manner is not limited in this embodiment.

In this embodiment, the load balancing module is configured to provide an external address for the data packet that is subjected to the load balancing process, where the external address may be the address of the virtual network created by the user, or may be directly the public network address of the external service. It is not further limited in this embodiment.

Further, in this embodiment, a specific manner of creating a virtual network and a load balancing module according to the creation request includes:

Creating a virtual subnet, the virtual subnet being created includes a first-level virtual subnet and/or a multi-level virtual subnet; preferably, in the embodiment, the created virtual subnet further includes creating a virtual machine, a gateway, or At least one of the routes.

Create a load balancing module and directly connect the created load balancing module to the virtual network system.

In this embodiment, the first-level virtual subnet includes at least one virtual machine that accesses the virtual network system through the virtual switch. It can be understood that only the first-level virtual switch in the first-level virtual subnet controls at least one virtual Specifically, when the load balancing module forwards the received data packet, the first-level virtual switch forwards the received data packet to the virtual machine; the multi-level virtual subnet includes at least one virtual network access system through the virtual switch. The virtual switch in the multi-level virtual subnet accesses the virtual network system step by step through the virtual switch in the other at least one level one virtual subnet, that is, the multi-level virtual subnet passes through at least two Step-by-step control of a virtual switch by a virtual switch, which can be understood as a first-level virtual switch controlling at least one virtual machine and at least one second-level virtual switch, and the second-level virtual switch simultaneously controls multiple virtual Machine; of course, can also include more levels of virtual switches and virtual machines; specifically when the load balancing module forwards packets First grade through the first virtual switch forwards the data, if the virtual need is not found at this level forwarding In the case of a virtual machine, the data is forwarded to the required virtual switch for processing, and of course, more levels of virtual switches and virtual machines can be included.

In this embodiment, the created virtual subnet is at least one level one virtual subnet, and the created at least one level one virtual subnet is directly accessed to the virtual network system through the virtual switch. In this embodiment, The first-level virtual subnet refers to a virtual switch in the network that controls or connects at least one virtual machine, and the virtual switch and at least one virtual mechanism controlled or connected become a first-level virtual subnet; Accessing a virtual network system can also be understood as a network interworking relationship between data forwarding between a virtual machine and a virtual network.

For further explanation of the embodiment, FIG. 3 is a schematic diagram of a cloud computing network architecture formed by a virtual network system and a load balancing module when the virtual subnet created in the embodiment is a first-level virtual subnet; FIG. 3 is the same embodiment. Specifically, the virtual machine in the virtual subnet is not in the same network segment as the load balancing module. Of course, in a virtual network system, the primary virtual subnet and the multi-level virtual subnet may be included at the same time.

When a data packet enters, the specific manner of processing the data packet by the cloud computing network architecture of FIG. 3 is:

In FIG. 3, after receiving the data packet requiring load balancing processing, the gateway sends the data packet to the load balancing module. For example, the data packet requiring the load balancing service needs to be A and B respectively. Processing on the virtual machine, the load balancing module can search for the corresponding virtual MAC and IP addresses respectively according to the previously stored network information or the network information corresponding to each virtual machine obtained in real time according to the content of the data packet. The packet is forwarded directly to the virtual switch in the virtual subnet, and then the virtual switch chooses to forward the packet to the virtual machine whose MAC and IP addresses are A and B respectively. As can be seen from FIG. 3, in the embodiment, load balancing is implemented on the data, and no route addressing is required, and only data transmission across the network segment can be implemented according to the acquired network information.

In this embodiment, preferably, before the load balancing module of FIG. 3 forwards the data packet to the virtual machine for processing, it is not only required to be based on previously stored or currently acquired network information corresponding to each virtual machine, but preferably also needs to be combined. The load balancing algorithm or the like forwards the data packet to the virtual machine for processing.

Further, in this embodiment, the network information of the virtual machine that is not on the same network segment as the load balancing module is stored in the load balancing module, specifically:

Obtaining at least one network information of the virtual machine that is not in the same network segment as the load balancing module. In this embodiment, preferably, acquiring the network information may be obtained before receiving the load balancing service data, or may be received. After the load balancing service data is required, it is acquired in real time according to requirements. In this embodiment, it is preferably acquired before receiving the load balancing service data.

After the network information is obtained, the virtual network device or the flow table information corresponding to each virtual machine is generated and stored in the load balancing module according to the obtained network information. In this embodiment, preferably, the generated The virtual network device or the flow table information is the address list information corresponding to each virtual machine. In this embodiment, it may also be another information list. By using the information list, the data packet can be directly received without addressing. The implementation manner of the load balancing sent to the corresponding virtual machine is within the scope of the claimed application.

Further, in this embodiment, the virtual network system preferably implements network isolation of the virtual machine by any one of the following methods: VLAN, GRE, and VXLAN; of course, other possible network isolation modes; GRE and VXLAN are extensions of the isolated VLAN. Because of the development of cloud computing, the VLAN ID in the VLAN isolation mode is limited and cannot satisfy a larger range of cloud computing. Therefore, it is extended on the basis of VLAN. GRE or VXLAN isolation is one of the extended isolation methods;

In this embodiment, the obtained network information is the address isolation information for isolating the virtual machine. In this embodiment, the network information of the virtual machine is obtained as any one of the following information: VLAN information, GRE information, and VXLAN. Information, wherein the VLAN information includes at least VLAN ID information, specifically including physical address information or network address information of the virtual machine; the GRE information also includes at least physical address information and/or network address information of the virtual machine; the VXLAN information needs a physical machine. At least one of network address information, physical address information of the virtual machine, and network address information.

In this embodiment, the network information of the virtual machine that is not in the same network segment as the load balancing module is obtained in the cloud computing network structure, and the corresponding virtual network device or flow table information is generated according to the network information, and is stored. In the load balancing module, the purpose of doing this is to configure a virtual machine that is not on the same network segment as the load balancing module as a service node at the back end of the load balancing module; when the data needs load balancing processing, only the acquired network needs to be obtained. The information is either the stored virtual network device or the flow table information, and the load balancing module can directly forward the data packet to the corresponding virtual machine service node without directly routing the route, so the data load balancing service is accelerated. At the same time, the performance of the load balancing service is also improved.

Embodiment 2:

4 is a schematic structural diagram of an apparatus for implementing load balancing according to an embodiment of the present disclosure, where the apparatus includes: a creating module 401, a processing module 402, and a forwarding module 403;

The creating module 401 separately creates a virtual network system and a load balancing module according to the user request, and the created virtual network system includes at least one virtual machine. In this embodiment, when the created virtual machine is more than two, the virtual The machines can belong to the same Layer 2 network or belong to different Layer 2 networks. In this embodiment, in After the virtual network system and the load balancing module are created, the created load balancing module is connected to the created virtual network system to ensure that the load balancing module can be interconnected with the virtual network system network.

In this embodiment, the virtual network system refers to a virtual layer 3 network provided for users in a cloud computing service, in which a user can create a network device such as a subnet, a route, a gateway, or the like, or Different networks perform interworking settings, and the performance of the network is exactly the same as the physical network.

The processing module 402 is configured to obtain network information of a virtual machine that is not in the same network segment as the load balancing module, and store the network information in the load balancing module, and configure the virtual machine as a service node at the back end of the load balancing module. In this embodiment, the virtual machine and the load balancing module are not in the same network segment, and the virtual and load balancing modules are not in the same Layer 2 network.

In this embodiment, the load balancing module can directly communicate with the virtual machine, and the load balancing module does not need to be in a network segment with the virtual machine.

Preferably, in this embodiment, when the load balancing module is created according to the creation request, the method further includes: configuring the load balancing module to provide a load balancing algorithm, a health check function, and a session maintaining function:

The forwarding module 403 is configured to forward the received data packet to the virtual machine for processing according to the network information in the load balancing module when receiving the data packet that needs to be subjected to load balancing processing.

Preferably, in this embodiment, the load balancing service provided by the load balancing module may be implemented by using hardware or by software. The specific implementation manner is not limited in this embodiment.

In this embodiment, the load balancing module preferably provides an external address for the data packet that needs to be subjected to load balancing processing, where the external address may be the address of the virtual network created by the user, or may be directly used as the public network address of the external service. It is not further limited in this embodiment.

Further, in this embodiment, the creating module 401 includes a subnet creating unit 4011 and a load balancing creating unit 4012;

The subnet creating unit 4011 is configured to create a virtual subnet. In this embodiment, the created virtual subnet further includes at least one of creating a virtual machine, a gateway, or a route.

The created virtual subnet includes: a first-level virtual sub-network and/or a multi-level virtual sub-network; the first-level virtual sub-network includes at least one virtual machine that accesses the virtual network through the virtual switch, which can be understood as Only the first level virtual switch controls at least one virtual machine in the first level virtual subnet; the multi-level virtual subnet includes at least one virtual switch accessing the virtual network step by step through the virtual switch in the other at least one level one virtual subnet, Understand For the multi-level virtual subnet, the step-by-step control of the virtual machine by at least two virtual switches can be understood as the first-level virtual switch controlling at least one virtual machine and at least one second-level virtual switch. The second level virtual switch controls multiple virtual machines at the same time.

The load balancing creation unit 4012 is configured to create a load balancing module and directly connect the created load balancing module to the virtual network system.

Further, in this embodiment, the processing module 402 includes an obtaining unit 4021; the obtaining unit 4021 is configured to acquire network information of at least one virtual machine that is not in the same network segment as the load balancing module, where the network information is performed on the virtual machine. The information of the network is isolated. In this embodiment, the obtaining unit 4021 may acquire the network information before receiving the data that needs the load balancing service, or may acquire the data according to the data content after receiving the data that needs the load balancing processing. Internet Information.

Preferably, in the embodiment, the virtual subnet created by the subnet creating unit 4011 is preferably, but not limited to, network isolation in the virtual network system by using any one of the following manners: VLAN, GRE and VXLAN; the network information of the virtual machine in the corresponding virtual subnet includes any one of the following information: VLAN information, GRE information, VXLAN information; GRE and VXLAN in the above isolation mode are extensions to the isolated mode VLAN because The VLAN ID in the VLAN isolation mode is limited, and cannot satisfy a larger range of cloud computing. Therefore, the VLAN is extended based on the VLAN, and the GRE or VXLAN isolation mode is one of the extended isolation modes.

In the embodiment, the network information acquired by the obtaining unit 4021 is: the VLAN information includes at least the VLAN ID information, and specifically includes the physical address information and/or the network address information of the virtual machine; the GRE information also includes at least the physical address of the virtual machine. Information and/or network address information; VXLAN information requires at least one of network address information of a physical machine, physical address information of a virtual machine, and network address information.

In this embodiment, the processing module 402 further includes an information creating unit 4022;

The information creating unit 4022 creates virtual network device or flow table information corresponding to each virtual machine according to the network information acquired by the obtaining unit 4021, and saves it in the load balancing module. In this embodiment, the virtual network device or the flow table information created by the information creating unit 4022 is address information corresponding to each virtual machine, and the address information enables the load balancing module to implement cross-network segment without reasoning addressing. data transmission.

In this embodiment, when the subnet creating unit 4011 creates a level one virtual subnet, the obtaining unit 4021 and the information creating unit 4022 have created virtual network devices or flow table information in the load balancing module according to the acquired network information. The virtual network device or the flow table information includes the VLAN information, the GRE information, or the VXLAN information acquired by the obtaining unit 4021; after receiving the data packet requiring the load balancing processing, the data packet is sent. And sent to the load balancing module, the load balancing module determines, according to the content of the data packet, which service node (virtual machine) needs to be processed on the back end of the data packet, and then searches for the service node corresponding to the virtual network device or the flow table information. Address, and finally the packet is transmitted to the corresponding virtual machine through the virtual switch for processing.

The device for implementing load balancing in the embodiment can enable the load balancing module to implement the cross-network segment without addressing based on the acquired network information (virtual network device or address information in the flow table information). Data transfer improves service performance for load balancing.

Embodiment 3:

In this embodiment, a load balancing service system is provided. Referring to FIG. 5, the system includes a virtual network system 501 and a load balancing module 502.

The load balancing module 502 accesses the virtual network system and provides load balancing services for at least one virtual machine in the virtual network system.

In this embodiment, the load balancing module 502 stores network information of a virtual machine that is not in the same network segment as the network segment. The purpose of storing the network information is to configure the virtual machine as a service node at the back end. In this embodiment, the network information is preferably related information in an isolation manner of performing network isolation on the virtual machine. For example, when the network is isolated by the VLAN in the VLAN mode, the obtained network information includes the VLAN ID. When the virtual machine network is isolated by using the GRE, the obtained network information includes the tag ID.

Certainly, the isolation mode used is not limited to the two types mentioned in the embodiment, and may also be a VXLAN isolation mode, and the corresponding network information includes VXLAN information, including network address information of the physical machine, physical address information and network of the virtual machine. Address information. In the meantime, in this embodiment, the acquired network information is not limited to the ones listed in the embodiment, and includes other possible and related address information or identification information.

In this embodiment, when there is a data packet to be forwarded in the load balancing module 502, the load balancing module 502 forwards the data packet to be forwarded directly to the service node (virtual machine) of the back end for processing according to the stored network information. The load balancing service system in this embodiment stores the network information of the virtual machine that is not in the same network segment, and configures the virtual machine as the service node of the back end, so that the data that needs the load balancing service is received. After the packet, the packet needs to be processed by the corresponding backend service node (virtual machine) according to the stored network information, and no addressing is needed. Therefore, the performance of the load balancing service is improved.

Embodiment 4:

In order to further explain the application, in this embodiment, a method for implementing load balancing creation in cloud computing is proposed. Referring to FIG. 6, the method includes:

Step 601: The user logs in through the console.

Step 602: Create a virtual machine first through the console.

Step 603: Create a virtual network through the console.

Step 604: Add the virtual machine created in step 602 to the virtual network created in step 603.

Step 605: Create a load balancing module (adding a virtual network system during the creation process);

Step 606: When the virtual machine is isolated by using the VLAN, the load balancing module obtains the VLAN information.

Step 607: Create flow table information according to the obtained VLAN information, and store it in the load balancing module.

In step 607, load balancing creates flow table information according to the obtained VLAN information, and configures the virtual machine as a service node of the back end, and the flow table information can ensure direct communication with the virtual machine, so that the load balancing module When a packet is received for forwarding, the packet can be directly forwarded to the corresponding virtual machine, and the routing must be forwarded because the virtual machine is in multiple network devices.

Through the above-mentioned load balancing creation method, after the user creates the load balancing, the configuration of various virtual networks can be completed at the same time, and the data link connection from the load balancing module node to the user virtual machine node is realized, thereby improving the load balancing service and the entire cloud. Calculate the performance of the network and enhance the user experience.

The above is a further detailed description of the present invention in connection with the specific embodiments, and the specific embodiments of the present invention are not limited to the description. It will be apparent to those skilled in the art that the present invention may be made without departing from the spirit and scope of the invention.

Industrial applicability

The virtual network system and the load balancing module are created according to the creation request, and the created load balancing module is connected to the virtual network system, and the created virtual network system includes at least one virtual machine; The network information of the virtual machine that is not on the same network segment as the load balancing module is stored in the load balancing module; when the received data packet requiring equalization processing exists in the virtual load balancing processing network, the data packet is passed through the The load balancing module forwards to the virtual machine for processing. It can be seen that, compared with the prior art, the present application configures the load balancing module and the virtual machine into the virtual load balancing processing network, so that the load balancing module and any one of the virtual machines are in the same layer 2 network environment, when needed. Carry out packet transfer When sending, there is no need for routing addressing, and the data can be directly forwarded directly through the load balancing module, which enhances the performance of the load balancing processing packet and can solve the problem of low performance of the load balancing service.

Claims (11)

A method of implementing load balancing, including: Creating a virtual network system and a load balancing module according to the creation request, and connecting the load balancing module to the virtual network system, where the virtual network system includes at least one virtual machine; Storing the network information of the virtual machine that is not on the same network segment as the load balancing module in the load balancing module, and configuring the virtual machine as an application service node at the back end of the load balancing module; The data packet requiring load balancing processing is forwarded to the virtual machine for processing by the load balancing module. The method for implementing load balancing according to claim 1, wherein the creating a virtual network system and a load balancing module, and accessing the load balancing module to the virtual network system is specifically: Creating a virtual subnet, the virtual subnet comprising: a first level virtual subnet and/or a multi-level virtual subnet; the first level virtual subnet comprising at least one virtual machine accessing the virtual network system through the virtual switch; The multi-level virtual subnet includes at least one virtual machine that accesses the virtual network system through the virtual switch; the virtual switch in the multi-level virtual subnet accesses the virtual network step by step through the virtual switch in the other at least one level one virtual subnet. system; Create a load balancing module and directly connect the load balancing module to the virtual network system. The method for implementing load balancing according to claim 2, wherein the creating the virtual subnet is specifically: creating at least one level one virtual subnet, and directly connecting the virtual switch in the level one virtual subnet to the virtual network system. The method for implementing load balancing according to claim 3, wherein the primary virtual subnet implements network isolation in the virtual network system by any one of the following methods: a virtual local area network VLAN, a general routing encapsulation GRE, The virtual scalable local area network (VXLAN); the network information corresponding to the virtual machine in the first-level virtual subnet is any one of the following information: VLAN information, GRE information, and VXLAN information. The method for implementing load balancing according to any one of claims 1-4, wherein the network information of the virtual machine that is not in the same network segment as the load balancing module is stored in the load balancing module, specifically: Obtaining at least one network information of a virtual machine that is not in the same network segment as the load balancing module; And creating virtual network device or flow table information corresponding to each virtual machine according to the acquired network information of the virtual machine, and saving the information in the load balancing module. An apparatus for implementing load balancing, comprising: creating a module, a processing module, and a forwarding module; The creating module is configured to separately create a virtual network system and a load balancing module according to the creation request, and connect the load balancing module to the virtual network system, where the virtual network system includes at least one virtual machine; The processing module is configured to store network information of a virtual machine that is not in the same network segment as the load balancing module in a load balancing module, and configure the virtual machine as an application service node at a back end of the load balancing module; The forwarding module is configured to forward the data packet requiring load balancing processing to the virtual machine through the load balancing module for processing. The device for implementing load balancing according to claim 6, wherein the creation module comprises a subnet creation unit and a load balancing creation unit; The subnet creating unit is configured to create a virtual subnet, where the virtual subnet includes: a first level virtual subnet and/or a multi-level virtual subnet; and the first level virtual subnet includes at least one virtual access through a virtual switch. a virtual machine in the network system; the virtual switch in the multi-level virtual subnet accesses the virtual network system step by step through virtual switches in the other at least one level one virtual subnet; The load balancing creation unit is configured to create a load balancing module and directly connect the load balancing module to the virtual network system. The apparatus for implementing load balancing according to claim 7, wherein the subnet creating unit creates a virtual subnet specifically: creating at least one level one virtual subnet, directly accessing the virtual switch in the level one virtual subnet Virtual network system. The apparatus for implementing load balancing according to claim 8, wherein the virtual subnet created by the subnet creating unit implements network isolation in the virtual network system by any one of the following modes: VLAN, GRE, virtual The scalable local area network VXLAN; the network information corresponding to the virtual machine in the first-level virtual subnet is any one of the following information: VLAN information, GRE information, and VXLAN information. The device for implementing load balancing according to any one of claims 6-9, wherein the processing module further comprises an obtaining unit and a creating unit; The obtaining unit is configured to acquire network information of at least one virtual machine that is not in the same network segment as the load balancing module; The creating unit is configured to create virtual network device or flow table information corresponding to each virtual machine according to the acquired network information of the virtual machine, and save the information in the load balancing module. A load balancing service system includes: a virtual network system and a load balancing module; The load balancing module accesses the virtual network system and provides a load balancing service for at least one virtual machine in the virtual network system; The load balancing module is configured to store network information of a virtual machine that is not in the same network segment as the virtual machine, and configure the virtual machine as its backend service node, and to pass the data packet requiring load balancing processing The load balancing module forwards to the virtual machine for processing.

Images