WO2017173952A1

WO2017173952A1 - Method, device, and system for centralizing management of virtual machines and implementing communications between virtual machines

Info

Publication number: WO2017173952A1
Application number: PCT/CN2017/078834
Authority: WO
Inventors: 沈世元; 袁俊; 叶松青; 孙月新; 孙文颖
Original assignee: 中兴通讯股份有限公司
Priority date: 2016-04-08
Filing date: 2017-03-30
Publication date: 2017-10-12
Also published as: CN107276783A; CN107276783B

Abstract

A method, device, and system for centralizing management of virtual machines and implementing communications between virtual machines. The method comprises: creating a first flow table and transmitting the same to a first switch for forwarding first virtual machine transceiving information, the first flow table comprising mapping between a first virtual machine and a virtual network; and creating a second flow table and transmitting the same to a second switch for forwarding second virtual machine transceiving information, the second flow table comprising mapping between a second virtual machine and a virtual network. The embodiment realizes centralized management for different types of virtual machines, increasing a service capability, and reducing operation costs. Furthermore, the forwarding process between the first virtual machine and the first switch is performed according to the first flow table, and the forwarding process between the second virtual machine and the second switch is performed according to the second flow table, implementing communication between different types of virtual machine, expanding a network scale in a virtual data center, and enabling the virtual data center to manage a larger network scale and range.

Description

Method, device and system for realizing unified management and intercommunication of virtual machines

Technical field

This document relates to, but is not limited to, a communication network, and in particular, to a method, device and system for implementing unified management and interworking of virtual machines.

Background technique

Virtual Data Center (VDC) is a new type of data center that applies the concept of cloud computing to the Internet Data Center (IDC). A unified and innovative VDC operation management system built by combining traditional IDC services and cloud computing technologies can be applied to technologies such as virtualization and automated deployment to build a scalable virtualized infrastructure, using centralized management and distributed service modes to users. Provide basic IT facilities solutions and services for acceptance and network-wide services. The main difference between VDC and traditional IDC is that infrastructure is provided as a service; abstraction of physical resources is abstracted through virtualization technology to enhance service capabilities; resource utilization and service reliability are improved through dynamic resource allocation and scheduling; and automated services are provided. Open capacity, reduce operation and maintenance costs, and provide a convenient user experience; provide more security mechanisms and reliability mechanisms to meet the security standards of enterprise applications.

Software Defined Network (SDN), whose core technology OpenFlow separates the control plane of the network device from the data plane, thus achieving flexible control of network traffic and providing a good platform for innovation of core networks and applications. Currently, in the carrier network, the pure SDN can start with a relatively closed data center and other scenarios to build a new SDN enhanced VDC. By introducing an SDN controller on the network control side, centralized control of the virtual switch (vSwitch) and the OF protocol enhanced TOR (Top of Rack) hardware switch built on the computing resources becomes a traditional one. The complex network topology of the data center is a large Layer 2 network architecture.

In the related art, the virtual data center involves a large number of network elements, including Openstack virtual platform, VMware ESXi virtual platform, SDN, DVS, ToR switch and other network elements, but the virtual data center management portal cannot simultaneously manage different types of resource pools. For example, you can't manage VMware and KVM resource pools at the same time, if the management portal can manage VMware and KVM at the same time. Resource pool, then these two computing resources can meet the requirements of flexible networking, that is, you can retain the original VMware ESXi virtual platform and integrate it with the Openstack virtual platform to enhance service capabilities and reduce operating costs. The network scale and scope that can be managed by the virtual data center system based on SDN is larger. In order to clearly understand the above, the following descriptions are made through relevant definitions, including:

Openstack Virtual Platform: OpenStack is an open source cloud computing management platform project that combines several major components to accomplish specific tasks. OpenStack supports almost all types of cloud environments. The goal of the project is to provide a cloud computing management platform that is simple to implement, scalable, rich, and standardized.

VMware ESXi Virtual Platform: VMware virtual machine software, a server and desktop virtualization solution, is an enterprise-class virtual platform that runs directly on hardware.

SDN controller: What SDN does is to separate the control rights on the network device and manage it by the centralized controller. It does not need to rely on the underlying network devices (routers, switches, firewalls), and shields the differences from the underlying network devices.

VDC management system management portal sub-module: resource management system, which provides centralized, flexible, and highly reliable computing, storage, network and other resources for VDC, unified management, on-demand distribution, and rental services.

TOR switch: Rack top switch, usually used for physical server access, needs to provide stacking/clustering capability, stack/cluster switch supports OpenFlow protocol, and supports port aggregation (same or cross-device) and hairpin (hairpin) Forward.

DVS: A virtual switch that can be used to form a virtual network. It also needs to provide a hypervisor agent (a proxy virtual machine located in ESXi) to register or vlan log off the VMware virtual machine virtual LAN (vlan).

Summary of invention

The following is an overview of the topics detailed in this document. This Summary is not intended to limit the scope of the claims.

The embodiments of the present invention provide a method, a device, and a system for implementing unified management and interworking of a virtual machine, which can simultaneously manage and manage virtual machines of different types and implement virtual machines of different types. through.

An embodiment of the present invention provides a method for implementing unified management of a virtual machine, including: constructing a first flow table, and sending the first flow table to a first switch configured to forward and receive information of the first virtual machine, the first flow table Include the mapping information between the first virtual machine and the virtual network; construct a second flow table, and send the second flow table to a second switch configured to forward the second virtual machine to send and receive information, the second flow The table contains mapping information between the second virtual machine and the virtual network.

Optionally, the constructing the first flow table includes: sending a packet that includes the first virtual machine identification information to a controller for physical address learning; and after completing the physical address learning, the controller includes the first virtual machine The information of the mapping relationship between the physical address and the virtual network is encapsulated into the first flow table.

Optionally, the constructing the second flow table includes: receiving an association message forwarded by the second switch, and setting a mapping between the port of the second virtual machine and the virtual network according to the associated message; The information of the mapping relationship between the port and the virtual network is encapsulated into the second flow table.

Optionally, before the constructing the second flow table, the method further includes: registering the second virtual machine with the second switch and the controller, and performing the second virtual machine and the virtual local area network. The mapping between port identifiers.

The embodiment of the present invention further provides a method for implementing virtual machine interworking, including: sending a first packet to a first switch or a second switch, where the first packet includes a network protocol data packet and virtual local area network information; Determining, by the first flow table, forwarding processing between the first virtual machine and the first switch, and the second flow table performing forwarding processing between the second virtual machine and the second switch, Implementing virtual machine interworking; wherein the first flow table includes mapping information between the first virtual machine and the virtual network, and the second flow table includes mapping information between the second virtual machine and the virtual network.

Optionally, the implementation of the virtual machine interworking includes: when the first virtual machine and the second virtual machine are in the same network segment, implementing Layer 2 interworking between the first virtual machine and the second virtual machine; When the first virtual machine and the second virtual machine are in different network segments, the Layer 2 and/or Layer 3 interworking between the first virtual machine and the second virtual machine is implemented.

The embodiment of the invention further provides an apparatus for implementing unified management of a virtual machine, including: first processing a unit and a second processing unit, the first processing unit is configured to construct a first flow table, and send the first flow table to a first switch configured to forward and receive information of the first virtual machine, where the first flow table includes Mapping information between the first virtual machine and the virtual network; the second processing unit is configured to construct a second flow table, and send the second flow table to the second switch configured to forward the second virtual machine to send and receive information The second flow table includes mapping information between the second virtual machine and the virtual network.

Optionally, the first processing unit includes a first processing subunit and a second processing subunit, where the first processing subunit is configured to send a packet that includes the first virtual machine identification information to a controller. The second processing sub-unit is configured to: after the controller completes the physical address learning, encapsulate information including a mapping relationship between the physical address of the first virtual machine and the virtual network as the first flow table.

Optionally, the second processing unit includes a third processing subunit and a fourth processing subunit, where the third processing subunit is configured to: receive an association message forwarded by the second switch, according to the The association message sets a mapping between the port of the second virtual machine and the virtual network; the fourth processing subunit is configured to: encapsulate information including a mapping relationship between the port and the virtual network as the first Second-rate table.

An embodiment of the present invention further provides an apparatus for implementing virtual machine interworking, a third processing unit, a fourth processing unit, and a fifth processing unit.

The third processing unit is configured to: send a first packet to the first switch or the second switch, where the first packet includes a network protocol data packet and virtual local area network information;

The fourth processing unit is configured to: perform forwarding processing according to the first packet;

The fifth processing unit is configured to: perform forwarding processing between the first virtual machine and the first switch by using the first flow table, and perform forwarding processing between the second virtual machine and the second switch by using the second flow table, Implement virtual machine interworking;

The first flow table includes mapping information between the first virtual machine and a virtual network;

The second flow table includes mapping information between the second virtual machine and the virtual network.

Optionally, the implementing the virtual machine interworking includes:

When the first virtual machine and the second virtual machine are in the same network segment, implementing the first virtual Layer 2 interworking between the virtual machine and the second virtual machine;

When the first virtual machine and the second virtual machine are in different network segments, the Layer 2 and/or Layer 3 interworking between the first virtual machine and the second virtual machine is implemented.

The embodiment of the invention further provides a system for implementing unified management and interworking of a virtual machine, comprising: a controller, a first switch and a second switch,

The controller is configured to: construct a first flow table and a second flow table, and send the first flow table to the first switch, and send the second flow table to the second switch; the first flow table includes Mapping information between a virtual machine and a virtual network, the second flow table including mapping information between the second virtual machine and the virtual network;

The first switch is configured to receive the first flow table sent by the controller, perform forwarding processing with the first virtual machine and the second switch by using the first flow table, and receive the first a message; the first message includes a network protocol data packet and virtual local area network information.

The second switch is configured to receive the second flow table sent by the controller, perform forwarding processing with the second virtual machine and the first switch by using the second flow table, and receive First message.

Compared with the related art, the technical solution provided by the embodiment of the present invention includes: constructing a first flow table, and sending the first flow table to a first switch for forwarding information sent and received by the first virtual machine, where the first flow table includes the first virtual Mapping information between the machine and the virtual network; constructing a second flow table, and sending the second flow table to the second switch for forwarding the second virtual machine to send and receive information, where the second flow table includes the second virtual machine and the virtual network Mapping information. It realizes unified management of different types of virtual machines, enhances service capabilities, and reduces operating costs.

In addition, the first packet is sent to the first switch or the second switch, where the first packet includes the network protocol data packet and the virtual local area network information; the forwarding process is determined according to the first packet; and the first virtual machine is performed by using the first flow table. Forwarding processing with the first switch; forwarding processing between the second virtual machine and the second switch by the second flow table. The interoperability between different types of virtual machines is realized, and the network scale in the virtual data center is expanded, so that the scale and scope of the network that the virtual data center system can manage is larger.

Other aspects will be apparent upon reading and understanding the drawings and detailed description.

BRIEF abstract

1 is a flowchart of a method for implementing unified management of a virtual machine according to Embodiment 1 of the present invention;

2 is a flowchart of a method for implementing virtual machine interworking in Embodiment 2 of the present invention;

3 is a schematic diagram of an apparatus for implementing unified management of a virtual machine according to Embodiment 3 of the present invention;

Figure 4 is a schematic view of the first processing unit of Figure 3;

Figure 5 is a schematic view of the second processing unit of Figure 3;

6 is a schematic diagram of an apparatus for implementing virtual machine interworking in Embodiment 4 of the present invention;

FIG. 7 is a schematic diagram of a system for implementing unified management and interworking of a virtual machine according to Embodiment 5 of the present invention; FIG.

8 is a schematic diagram of a system for implementing unified management and interworking of KVM and VMware virtual machines according to Embodiment 5 of the present invention;

FIG. 9 is another schematic diagram of a system for implementing unified management and interworking of KVM and VMware virtual machines according to Embodiment 5 of the present invention; FIG.

FIG. 10 is a flowchart of a method for implementing unified management and interworking of KVM and VMware virtual machine according to Embodiment 5 of the present invention.

Detailed

Embodiments of the present application will be described in detail below with reference to the accompanying drawings. It should be noted that, in the case of no conflict, the features in the embodiments and the embodiments in the present application may be arbitrarily combined with each other.

The technical solutions in the embodiments of the present invention will be clearly and completely described in the following with reference to the accompanying drawings. For example, not all of the embodiments; it should be noted that the embodiments of the present application and the features of the embodiments may be combined with each other without conflict. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of the present invention.

Embodiment 1:

This embodiment provides a method for implementing unified management of a virtual machine. Referring to FIG. 1, the method includes:

Step 101: Construct a first flow table, and send the first flow table to a first switch configured to forward and receive information of the first virtual machine, where the first flow table includes a mapping information between the first virtual machine and the virtual network. interest;

Step 102: Construct a second flow table, and send the second flow table to a second switch configured to forward the second virtual machine to send and receive information, where the second flow table includes mapping information between the second virtual machine and the virtual network.

For the step 101, the first flow table is configured to send the message including the first virtual machine identification information to the controller for learning. After the learning ends, the information including the mapping relationship between the physical address of the first virtual machine and the virtual network is encapsulated as First flow table. Optionally, the virtual platform corresponding to the first virtual machine and the second virtual machine is configured on the management interface. After the configuration is complete, the first virtual machine is created on the management interface, and the first switch sends the packet of the first virtual machine. After the packet is sent to the controller, the virtual machine MAC (physical address) is learned. After the learning is completed, the controller sends the first flow table to the first switch where the first virtual machine is located. The first flow table includes VNET (virtual network). The entry of the +SRC MAC (the physical NIC information of the source MAC source virtual machine), the traffic sent by the subsequent first virtual machine can be forwarded according to the first flow table. Optionally, the first virtual machine herein may be a KVM virtual machine, and the corresponding first switch is a DVS, where the controller may be an SDN controller. It should be noted that the first virtual machine may include one type of virtual machine different from the second virtual machine instead of one virtual machine. In addition, the first virtual machine is not limited to the KVM virtual machine, and the first switch is not limited to The DVS, the controller is not limited to the SDN controller, and any module that can implement the content of the above step 1 is within the protection scope of the embodiment, and can be correspondingly selected and replaced as needed.

For step 102, constructing the second flow table may include receiving an association message forwarded by the second switch, setting a mapping between the second virtual machine port and the virtual network according to the association message, and mapping the port including the second virtual machine to the virtual network. The information of the relationship is encapsulated into a second flow table. Before the second flow table is constructed, the embodiment of the present invention further includes: registering the second virtual machine with the second switch and the controller, and performing mapping between the second virtual machine and the virtual local area network and the port identifier. Optionally, the second virtual machine is created on the management interface, including {"port_uuid", "virtual network_id", "name", "address (ip_address)", "physical address" (mac)", "security group (security_groups)"} and other information. Then, the association message is sent to perform port_uuid. Optionally, the associated message includes {port_uuid, vid}; after the second virtual machine is created and the port_uuid is completed, the second switch and the controller are injected through the associated message. The registration information includes the port_uuid and the vlan, and performs mapping between the VM and the vlan and the port_uuid. When the virtual machine is deleted, the association is eliminated to the second switch and the controller, and the controller is notified that the virtual machine has been deleted. After the registration is completed, the second switch sends the associated message to the controller according to the second flow table sent by the controller in advance, and the controller first determines the access of the second virtual machine according to the port and dpid sent to the packet-in. And then, according to the port_uuid, confirm the information of the second virtual machine associated with the associated message; then, according to the in_port of the vlan and the OpenFlow, configure the vlan sub-interface to the second virtual machine port, and add the vlan sub-interface to the virtual network, that is, set the port- And vnet mapping, according to the foregoing process, generating a second flow table; the second flow table includes a mapping relationship between the second virtual machine port and the VNET; and then sending the second flow table to the second switch. Optionally, the second virtual machine may be a VMware virtual machine, the corresponding second switch is a ToR switch, the controller may be an SDN controller, the management interface is a VDC management interface, and the VDC management is performed when the VMware virtual machine is created. The interface can be used to create a VMware virtual machine by using the northbound interface of the SDN controller. The corresponding associated message may be a VDP (VSI Discovery Protocol) association message. The second flow table may be a VDP protocol flow table.

It should be understood that when the first virtual machine is a KVM virtual machine and the second virtual machine is a VMware virtual machine, the virtual platform configured on the management interface of the VDC may be an Openstack virtual platform and a VMware virtual platform, and the pre-delivered stream. The table can be (Ethertype: 0x8940). The first virtual machine and the second virtual machine include two types of virtual machines of different types, instead of one virtual machine. In addition, it should be understood that any module that can implement the foregoing content is within the protection scope of the embodiment. It can be selected and replaced as needed.

The method for implementing the unified management of the virtual machine provided in this embodiment, by constructing the first flow table, and sending the first flow table to the first switch for forwarding the information of the first virtual machine to send and receive, the first flow table includes the first virtual machine and the virtual Mapping information between the networks; constructing a second flow table, and sending the second flow table to the second switch for forwarding the second virtual machine to send and receive information, where the second flow table includes mapping information between the second virtual machine and the virtual network . The unified management of the virtual machine by the management interface is realized, the service capability of the system is enhanced, and the operation cost is reduced.

The embodiment of the invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the foregoing method for realizing unified management of the virtual machine. law.

The embodiment of the invention further provides an apparatus for implementing unified management of a virtual machine, comprising: a memory and a processor; wherein

The processor is configured to execute program instructions in the memory;

Program instructions perform the following operations on the processor read:

Constructing a first flow table, and sending the first flow table to the first switch configured to forward the first virtual machine to send and receive information, where the first flow table includes mapping information between the first virtual machine and the virtual network;

Constructing a second flow table, and sending the second flow table to the second switch configured to forward the second virtual machine to send and receive information, where the second flow table includes mapping information between the second virtual machine and the virtual network.

Embodiment 2:

This embodiment provides a method for implementing virtual machine interworking. Referring to FIG. 2, the method includes:

Step 201: Send a first packet to the first switch, where the first packet includes a network protocol data packet and virtual local area network information.

Step 202: Determine, according to the first packet, that forwarding processing is performed.

Step 203: Perform forwarding processing between the first virtual machine and the first switch by using the first flow table. The first flow table includes mapping information between the first virtual machine and the virtual network.

Step 204: Perform switch forwarding processing between the second virtual machine and the second switch by using the second flow table. The second flow table includes mapping information between the second virtual machine and the virtual network.

The embodiment of the present invention performs forwarding processing between the first virtual machine and the first switch by using the first flow table, and the second flow table performs forwarding processing between the second virtual machine and the second switch, so that virtual machine interworking can be implemented;

The interworking between the virtual machines is performed by the foregoing steps: when the first virtual machine and the second virtual machine are in the same network segment, the Layer 2 forwarding interworking between the first virtual machine and the second virtual machine is implemented; when the first virtual machine and the first virtual machine When the second virtual machine is in different network segments, Layer 2 and/or Layer 3 forwarding interworking between the VMware virtual machine and the KVM virtual machine is implemented.

Optionally, the network protocol data packet (IP packet) may be encapsulated into a first packet by using a vlan (virtual local area network) information, and then forwarded to the second switch, and the second switch performs forwarding processing according to the first flow table sent by the controller. Therefore, the two-layer and three-layer forwarding interworking between virtual machines is realized. If the first virtual machine and the second virtual machine are the same network segment, and the first virtual machine is used as the destination end and the second virtual machine is used as the source end, the hypervisor adds the vlan of the second virtual machine to the vlan and forwards the IP packet. To the second switch, the second switch performs forwarding processing according to the first flow table sent by the controller, and converts the message vlan pop according to vlan and inport, and converts it into metadata (virtual network (VNET), that is, converts to Metadata of the virtual network; determining whether host learning is required according to the metadata (VNET) and the source (src) MAC; performing Layer 2 forwarding according to the metadata (VNET) and the dst MAC, and the first switch at the destination receives the request packet. Forwarding to the first virtual machine, the response packet of the first virtual machine is also sent to the second switch according to the first flow table on the first switch, and the second switch is forwarded to the second virtual machine, thereby implementing the first virtual machine and the first virtual machine. The second virtual machine and the second virtual machine are in different network segments. If the first virtual machine and the second virtual machine are on different network segments, the hypervisor adds the vlan1 of the second virtual machine to the second switch, and the second switch presses the second virtual switch. According to the flow table sent by the controller, the packet is forwarded according to vlan and inport, and converted into metadata (VNET); according to metadata (VNET) and src MAC, it is determined whether host learning is required; according to metadata (VNET) And the dst MAC, determine to perform Layer 2 forwarding; according to metadata (virtual routing forwarding table (VRF)) and dst IP, that is, according to the virtual routing forwarding table metadata and dst IP for Layer 3 forwarding; the destination first switch receives After the request packet is forwarded to the first virtual machine, the response packet of the first virtual machine is sent to the second switch according to the first flow table on the first switch, and the second switch is forwarded to the second virtual machine, thereby implementing the first Three layers of interworking between the virtual machine and the second virtual machine.

Optionally, the first virtual machine may include a KVM virtual machine, the second virtual machine may include a VMware virtual machine, and the controller may include an SDN controller. Correspondingly, the first switch is a DVS, and the second switch is a ToR switch. The first virtual machine and the second virtual machine include two types of virtual machines of different types instead of one virtual machine. In addition, it should be understood that any module that can implement the foregoing steps is in this embodiment. Within the scope of protection, you can make the appropriate selection and replacement as needed.

In this embodiment, the first packet is sent to the first switch or the second switch, where the first packet includes Network protocol data packet and virtual local area network information; performing forwarding processing according to the first packet; performing forwarding processing between the first virtual machine and the first switch by using the first flow table, where the first flow table includes the first virtual machine and the virtual network Mapping information; performing forwarding processing between the second virtual machine and the second switch by using the second flow table. The second flow table includes mapping information between the second virtual machine and the virtual network. The interoperability between different types of virtual machines is realized, and the network scale in the virtual data center is expanded, so that the scale and scope of the network that the virtual data center system can manage is larger.

The embodiment of the invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the foregoing method for implementing virtual machine interworking.

An embodiment of the present invention further provides an apparatus for implementing virtual machine interworking, including: a memory and a processor; wherein

The processor is configured to execute program instructions in the memory;

Program instructions perform the following operations on the processor read:

Sending a first packet to the first switch or the second switch, where the first packet includes a network protocol data packet and virtual local area network information;

Performing forwarding processing according to the first packet;

The forwarding process between the first virtual machine and the first switch is performed by using the first flow table, and the second flow table performs forwarding processing between the second virtual machine and the second switch to implement virtual machine interworking;

The first flow table includes mapping information between the first virtual machine and the virtual network;

Embodiment 3:

This embodiment provides an apparatus for implementing unified management of a virtual machine. Referring to FIG. 3, the first processing unit 11 and the second processing unit 12 are configured. The first processing unit 11 is configured to construct a first flow table, and send the first flow table. For the first switch configured to forward the first virtual machine to send and receive information, the first flow table includes mapping information between the first virtual machine and the virtual network; and the second processing unit 12 is configured to build The second flow table is sent to the second switch configured to forward the second virtual machine to send and receive information, and the second flow table includes mapping information between the second virtual machine and the virtual network.

Optionally, referring to FIG. 4, the first processing unit 11 includes a first processing sub-unit 111 and a second processing sub-unit 112, and the first processing sub-unit 111 is configured to send a packet including the first virtual machine identification information to The controller performs physical address learning. The second processing sub-unit 112 is configured to encapsulate, after the physical address learning, the information including the mapping relationship between the physical address of the first virtual machine and the virtual network as the first flow table. Optionally, the virtual platform corresponding to the first virtual machine and the second virtual machine is configured on the management interface. After the configuration is complete, the first virtual machine is created on the management interface, and the first switch sends the packet of the first virtual machine. After the packet is sent to the controller, the virtual machine MAC (physical address) is learned. After the learning is completed, the controller sends the first flow table to the first switch where the first virtual machine is located. The first flow table includes VNET (virtual network). The entry of the +SRC MAC (the physical NIC information of the source MAC source virtual machine), the traffic sent by the subsequent first virtual machine can be forwarded according to the first flow table. Optionally, the first virtual machine herein may include a KVM virtual machine, and the corresponding first switch may be a DVS, where the controller may be an SDN controller. It should be understood that the first virtual machine includes a virtual machine that is different from the second virtual machine, rather than a single virtual machine. In addition, the first virtual machine is not limited to the KVM virtual machine, and the first switch is not limited to the DVS. The controller is not limited to the SDN controller, and any module that can implement the above content is within the protection scope of the embodiment, and can be correspondingly selected and replaced as needed.

Optionally, referring to FIG. 5, the second processing unit 12 includes a third processing sub-unit 121 and a fourth processing sub-unit 122. The third processing sub-unit 121 is configured to: receive an association message forwarded by the second switch, according to the association. The message sets a mapping between the second virtual machine port and the virtual network. The fourth processing sub-unit 122 is configured to encapsulate the information including the mapping relationship between the port of the second virtual machine and the virtual network as the second flow table. It is to be understood that before the second flow table is constructed, the second virtual machine is registered with the second switch and the controller, and the mapping between the second virtual machine and the virtual local area network and the port identifier is performed.

Optionally, the second virtual machine is created on the management interface, including {"port_uuid", "virtual network_id", "name", "address (ip_address)", "physical address" (mac)", "security group (security_groups)"} and other information. Then send Associate the message with port_uuid. Optionally, the associated message includes {port_uuid, vid}. After the second virtual machine and port_uuid are created, the second switch and the controller are registered by the associated message, and the registration information includes port_uuid and vlan. The mapping between the VM and the vlan and the port_uuid is performed. When the virtual machine is deleted, the association is eliminated to the second switch and the controller, and the controller is notified that the virtual machine has been deleted. After the registration is completed, the second switch sends the associated message to the controller according to the second flow table sent by the controller in advance, and the controller first determines the access of the second virtual machine according to the port and dpid sent to the packet-in. And then, according to the port_uuid, confirm the information of the second virtual machine associated with the associated message; then, according to the in_port of the vlan and the OpenFlow, configure the vlan sub-interface to the second virtual machine port, and add the vlan sub-interface to the virtual network, that is, set the port- And vnet mapping, according to the foregoing process, generating a second flow table; the second flow table includes a mapping relationship between the second virtual machine port and the VNET; and then sending the second flow table to the second switch. Optionally, the second virtual machine may be a VMware virtual machine, the corresponding second switch is a ToR switch, the controller may be an SDN controller, the management interface is a VDC management interface, and the VDC management is performed when the VMware virtual machine is created. The interface can be used to create a VMware virtual machine by using the northbound interface of the SDN controller. The corresponding associated message may be a VDP (VSI Discovery Protocol) association message. The second flow table may be a VDP protocol flow table.

It should be understood that when the first virtual machine is a KVM virtual machine and the second virtual machine is a VMware virtual machine, the virtual platform configured on the management interface of the VDC may be an Openstack virtual platform and a VMware virtual platform, and the pre-delivered stream. The table can be (Ethertype: 0x8940). The first virtual machine and the second virtual machine include two types of virtual machines of different types instead of one virtual machine. In addition, it should be understood that any module that can implement the contents of steps 1 and 2 above is Within the scope of protection of this embodiment, it can be correspondingly selected and replaced as needed.

The method for implementing the unified management of the virtual machine provided by the embodiment includes: constructing the first flow table, and sending the first flow table to the first switch configured to forward the information of the first virtual machine to send and receive information, where the first flow table includes the first virtual machine and the virtual Mapping information between networks; constructing a second flow table, and sending the second flow table to a second switch configured to forward and receive information of the second virtual machine, where the second flow table includes mapping information between the second virtual machine and the virtual network . The unified management of the virtual machine by the management interface is realized, the service capability of the system is enhanced, and the operation cost is reduced.

Embodiment 4:

This embodiment provides an apparatus for implementing virtual machine interworking. Referring to FIG. 6, a third processing unit 13, a fourth processing unit 14, and a fifth processing unit 15 are provided. The third processing unit 13 is configured to: Or the second switch sends the first packet, where the first packet includes the network protocol data packet and the virtual local area network information; the fourth processing unit 14 is configured to: perform forwarding processing according to the first packet, and the fifth processing unit 15 is configured to: The first flow table performs forwarding processing between the first virtual machine and the first switch, and the second flow table performs forwarding processing between the second virtual machine and the second switch to implement virtual machine interworking; wherein the first flow table includes the first Mapping information between the virtual machine and the virtual network; the second flow table includes mapping information between the second virtual machine and the virtual network.

Through the foregoing steps, the interworking between the virtual machines is implemented, including: when the first virtual machine and the second virtual machine are in the same network segment, implementing Layer 2 forwarding interworking between the first virtual machine and the second virtual machine; When the second virtual machine is in different network segments, the Layer 2 and Layer 3 forwarding interworking between the first virtual machine and the second virtual machine is implemented.

Optionally, the IP packet (network protocol data packet) may be added with a vlan (virtual local area network) information, and the first packet is forwarded to the second switch, and the second switch forwards the packet according to the first flow table sent by the controller. Processing, so as to realize the forwarding and interworking between Layer 2 and Layer 3 of the virtual machine. If the first virtual machine and the second virtual machine are the same network segment, and the first virtual machine is used as the destination end and the second virtual machine is used as the source end, the hypervisor adds the IP packet of the second virtual machine to the vlan and forwards the packet to the second switch. The second switch performs forwarding processing according to the first flow table sent by the controller, converts the message vlan pop according to vlan, inport, and converts it into metadata (VNET); determines whether the host needs to be performed according to the metadata (VNET) and the src MAC. Learning; determining the Layer 2 forwarding according to the metadata (VNET) and the dst MAC, the first switch of the destination end receives the request packet and forwards the request packet to the first virtual machine, and the response packet of the first virtual machine is also according to the first switch. The first-class table is sent to the second switch, and the second switch is forwarded to the second virtual machine to implement Layer 2 interworking between the first virtual machine and the second virtual machine. If the first virtual machine and the second virtual machine are on different network segments, the hypervisor adds the vlan1 of the second virtual machine to the second switch, and the second switch performs forwarding processing according to the flow table sent by the controller. According to the vlan, Inport, the message vlan pop, and converted to metadata (VNET); according to metadata (VNET) and src MAC, determine whether host learning is required; according to metadata (VNET) and dst MAC, determine to carry out Layer 2 forwarding; according to metadata ( VRF) and dst IP, perform Layer 3 forwarding; the destination switch forwards the request packet after receiving the request packet. The first virtual machine sends the response packet of the first virtual machine to the second switch according to the first flow table on the first switch, and the second switch forwards the second virtual machine to the second virtual machine, thereby implementing the first virtual machine and the second virtual machine. Three layers of interworking between virtual machines.

Optionally, the first virtual machine may be a KVM virtual machine, the second virtual machine may be a VMware virtual machine, and the controller may be an SDN controller; correspondingly, the first switch is a DVS, and the second switch is a ToR switch. The first virtual machine and the second virtual machine refer to two types of virtual machines of different types, instead of one virtual machine. In addition, it should be understood that any module that can implement the foregoing steps is in this embodiment. Within the scope of protection, it can be selected and replaced as needed.

In this embodiment, the first packet is sent to the first switch or the second switch, where the first packet includes the network protocol data packet and the virtual local area network information; the forwarding process is determined according to the first packet; and the first virtual table is used to perform the first virtual process. Forwarding processing between the machine and the first switch, the first flow table includes mapping information between the first virtual machine and the virtual network; and the second flow table performs forwarding processing between the second virtual machine and the second switch, where the second flow table includes Mapping information between the second virtual machine and the virtual network. The interoperability between different types of virtual machines is realized, and the network scale in the virtual data center is expanded, so that the scale and scope of the network that the virtual data center system can manage is larger.

Embodiment 5:

This embodiment provides a system for implementing unified management and interworking of a virtual machine. Referring to FIG. 7, the system includes: a controller 1, a first switch 2, and a second switch 3. The controller 1 is configured to: construct a first flow table and a second Flowing the table, and sending the first flow table to the first switch 2, and sending the second flow table to the second switch 3; the first flow table includes mapping information between the first virtual machine and the virtual network, and the second flow table includes the second Mapping information between the virtual machine and the virtual network; the first switch 2 is configured to: receive the first flow table sent by the controller, perform forwarding processing with the first virtual machine and the second switch by using the first flow table; and the second switch 3 The method is configured to: receive a second flow table sent by the controller, perform forwarding processing with the second virtual machine and the first switch by using the second flow table; and the first switch and the second switch are further configured to receive the first packet, The first message contains network protocol packets and virtual local area network information.

Optionally, the first virtual machine may be a KVM virtual machine, the second virtual machine may be a VMware virtual machine, and the controller may be an SDN controller. Correspondingly, the first switch is a DVS, and the second switch is a ToR switch. , corresponding to the KVM virtual machine and VMware virtual machine virtual platform It is the Openstack virtual platform and the VMware ESXI virtual platform.

Optionally, referring to FIG. 8, the system for implementing unified management and interworking of the KVM virtual machine and the VMware virtual machine includes an Openstack virtual platform 21, a VMware ESXI virtual platform 22, an SDN controller 23, a ToR switch 24, and a DVS 25, optionally, The Openstack virtual platform 21 and the VMware ESXI virtual platform 22 are respectively configured to create a KVM virtual machine and create a VMware virtual machine on the management interface; the SDN controller 23 is configured to construct the first flow table and the second flow table, and send the first flow table to The DVS 25 sends the second flow table to the TOR switch 24; the first flow table includes mapping information between the physical address of the KVM virtual machine and the VNET, and the second flow table includes mapping information between the port of the VMware virtual machine and the VNET; the DVS 25 is set to Receiving a first flow table sent by the SDN controller, performing forwarding processing with the KVM virtual machine and the TOR switch by using the first flow table; the TOR switch is configured to receive the second flow table sent by the SDN controller 23, and pass the second flow table Perform forwarding processing with the VMware virtual machine and DVS25; also set to receive packets sent by the Hypervisor.

Referring to FIG. 9, FIG. 9 is another schematic diagram of a system for implementing unified management and interworking of a VMware virtual machine and a KVM virtual machine according to the embodiment. In addition to the module shown in FIG. 8, the virtual center (Vcenter)/VDP26 is also included. VM1 and VM2, optionally, DVS25 can be ZXDVS, VM1 is a VMware virtual machine, and VM2 is a KVM virtual machine.

Referring to FIG. 10, the following provides a method for implementing unified management and interworking of VMware and KVM virtual machines provided in this embodiment, including:

Step 301: Configure an Openstack virtual platform and a VMware virtual platform on the management interface of the VDC, and create a KVM virtual machine on the management interface to obtain a first flow table.

Optionally, the DVS sends the virtual machine packet to the SDN controller through the Packet-in Message to learn the MAC address of the virtual machine. After the learning is completed, the SDN controller sends the VNET+SRC MAC entry to the DVS where the KVM virtual machine is located. , that is, the first flow table, that is, the controller associates the virtual machine vid (virtual network id) to the VLAN sub-interface, and the traffic sent by the subsequent virtual machine can be forwarded according to the first flow table.

Step 302: The VDC management interface invokes the northbound interface of the SDN controller through OpenStack to create a VMware virtual machine.

Optionally, pass {"port_uuid", "virtualnetwork_id", "name", "ip_address", "mac (physical address)", "security_groups" (security group) )") to the SDN controller, the SDN controller saves the virtual machine's information to the local database;

Step 303: An existing agent virtual machine located in the ESXi sends a VDP association message, including {port_uuid, vid};

Step 304: After the VMware virtual machine and the port_uuid are created, the agent registers with the TOR and the SDN controller through the VDP message.

Optionally, the registration information includes port_uuid and vlan, and mapping between the VM and the vlan and the port_uuid. When the virtual machine is deleted, the agent removes the association between the TOR and the SDN controller to notify the controller that the virtual machine has been deleted.

Step 305, encapsulating the second flow table; that is, setting a port-vnet mapping to obtain a second flow table;

Optionally, the ToR switch sends the VDP message to the SDN controller according to the VDP flow table (Ethertype: 0x8940) pre-delivered by the SDN controller, and the controller determines the access location according to the port and dpid sent to the packet-in. According to the port_uuid, the virtual machine information associated with the VDP message is confirmed. According to the in_port of the vlan and the OpenFlow, the vlan sub-interface is configured on the port, and the vlan sub-interface is added to the virtual network, that is, the port-vnet mapping is set. After the processing ends, the switch is sent to the ToR switch. Sending a second flow table, where the second flow table is a VDP protocol flow table;

Step 306: Perform forwarding processing between the KVM virtual machine and the DVS through the first flow table, and perform forwarding processing between the VMware virtual machine and the TOR through the second flow table.

Optionally, if the VMware virtual machine and the KVM virtual machine are on the same network segment, the Hypervisor adds the vlan of the VMware virtual machine to the ToR switch, and the ToR performs forwarding processing according to the flow table sent by the SDN controller. According to the vlan, Inport, converts the message vlan pop and converts it to metadata (VNET); determines whether host learning is required according to metadata (VNET) and src MAC; performs Layer 2 forwarding according to metadata (VNET) and dst MAC, and destination DVS receives After the request message is forwarded to the KVM virtual machine, the response message of the KVM virtual machine is also sent to the ToR switch according to the flow table on the DVS, and the ToR switch is forwarded to the VMware virtual machine, thereby implementing the VMware virtual machine and the KVM virtual machine. Inter-layer two-way interworking;

If the VMware VM and the KVM VM are on different network segments, the Hypervisor will add the IP packet of the VMware VM to vlan1 and forward it to the ToR switch. The ToR will forward the packet according to the flow table sent by the SDN controller. According to vlan and inport, the packet will be forwarded. Vlan pop, and converted to metadata (VNET); according to metadata (VNET) and src MAC, determine whether host learning is required; according to metadata (VNET) and dst MAC, determine to carry out Layer 2 forwarding; according to metadata (VRF) and dst IP The destination DVS receives the request packet and forwards it to the KVM virtual machine. The response message of the KVM virtual machine is also sent to the TOR switch according to the flow table on the DVS, and the ToR switch forwards it to the VMware virtual machine. Implement three levels of interworking between VMware virtual machines and KVM virtual machines.

The system for implementing unified management and interworking of VMware and KVM virtual machines provided by this embodiment is based on the principle of “not requiring VMware to open a special interface and does not require deep bundling and authentication with VMware”, and moves the VTEP point to the ToR switch. The above system realizes the unified management and intercommunication of VMware virtual machine and KVM virtual machine, enhances the service capability of the system, reduces the operation cost, expands the network scale in the virtual data center based on SDN, and enables the virtual data center system to be managed. The size and scope of the network is larger.

It should be noted that the VTEP is moved from the vSwitch to the ToR switch. The ToR switch is on the same port. Each VM has one VLAN. The VLAN does not distinguish between the network and only the host (two hosts on the same network, and the VLANs are different). Under different ports of ToR, VLANs can be reused, which can simplify the Layer 2 and Layer 3 interworking processes between VMware virtual machines and KVM virtual machines. It can also manage VMware virtual machines and set advanced functions such as security. Group, meter, flow mirroring, redirection strategy, etc., and can expand the network scale in the virtual data center based on SDN, so that the virtual data center system can manage a larger network scale and scope, which can be connected to the openstack virtual platform. Docking VMware virtual platforms can also manage both and integrate.

One of ordinary skill in the art will appreciate that all or a portion of the above steps may be performed by a program to instruct related hardware, such as a processor, which may be stored in a computer readable storage medium, such as a read only memory, disk or optical disk. Wait. Alternatively, all or part of the steps of the above embodiments may also be implemented using one or more integrated circuits. Accordingly, each of the above embodiments The modules/units may be implemented in the form of hardware, for example, by an integrated circuit to implement their respective functions, or in the form of software functional modules, for example, by executing a program/instruction stored in the memory by the processor to perform its respective functions. The invention is not limited to any specific form of combination of hardware and software.

The embodiments disclosed in the present application are as described above, but the descriptions are only for the purpose of understanding the present application, and are not intended to limit the present application, such as the specific implementation method in the embodiments of the present invention. Any modifications and changes in the form and details of the embodiments may be made by those skilled in the art without departing from the spirit and scope of the disclosure. The scope defined by the appended claims shall prevail.

Industrial applicability

The foregoing technical solution realizes unified management of different types of virtual machines and interworking between different types of virtual machines, enhances service capabilities, and expands the network scale in the virtual data center.

Claims

A method for implementing unified management of a virtual machine, comprising:

Constructing a first flow table, and sending the first flow table to the first switch configured to forward the first virtual machine to send and receive information, where the first flow table includes mapping information between the first virtual machine and the virtual network;

Constructing a second flow table, and sending the second flow table to a second switch configured to forward the second virtual machine to send and receive information, where the second flow table includes mapping information between the second virtual machine and the virtual network .
The method of claim 1 wherein said constructing said first flow table comprises:

Sending the packet containing the first virtual machine identification information to the controller for physical address learning;

After the controller completes the physical address learning, the information including the mapping relationship between the physical address of the first virtual machine and the virtual network is encapsulated into the first flow table.
The method of claim 1 or 2, wherein said constructing the second flow table comprises:

Receiving an association message forwarded by the second switch, and setting a mapping between a port of the second virtual machine and the virtual network according to the associated message;

Encapsulating information including a mapping relationship between the port and the virtual network is the second flow table.
The method of claim 3, before the constructing the second flow table, the method further comprises:

Registering the second virtual machine with the second switch and the controller, and performing mapping between the second virtual machine and the virtual local area network and the port identifier.
A method for implementing virtual machine interworking, including:

Sending a first packet to the first switch or the second switch, where the first packet includes a network protocol data packet and virtual local area network information;

Determining, according to the first packet, forwarding processing;

The forwarding process between the first virtual machine and the first switch is performed by using the first flow table, and the second flow table performs forwarding processing between the second virtual machine and the second switch to implement virtual machine interworking;

The first flow table includes mapping information between the first virtual machine and a virtual network;

The second flow table includes mapping information between the second virtual machine and the virtual network.
The method of claim 5 wherein said implementing virtual machine interworking comprises:

When the first virtual machine and the second virtual machine are in the same network segment, implementing Layer 2 interworking between the first virtual machine and the second virtual machine;

When the first virtual machine and the second virtual machine are in different network segments, the Layer 2 and/or Layer 3 interworking between the first virtual machine and the second virtual machine is implemented.
An apparatus for implementing unified management of a virtual machine includes: a first processing unit and a second processing unit,

The first processing unit is configured to construct a first flow table, and send the first flow table to a first switch configured to forward and receive information of the first virtual machine, where the first flow table includes the first virtual machine and the virtual Mapping information between networks;

The second processing unit is configured to construct a second flow table, and send the second flow table to a second switch configured to forward the second virtual machine to send and receive information, where the second flow table includes the second virtual Mapping information between the machine and the virtual network.
The apparatus of claim 7, the first processing unit comprising a first processing subunit and a second processing subunit; wherein

The first processing sub-unit is configured to: send a packet that includes the first virtual machine identification information to a controller for physical address learning;

The second processing sub-unit is configured to: after the controller completes the physical address learning, encapsulate information including a mapping relationship between the physical address of the first virtual machine and the virtual network as the first flow table.
The apparatus according to claim 7 or 8, wherein said second processing unit comprises a third processing subunit and a fourth processing subunit; wherein

The third processing sub-unit is configured to: receive an association message forwarded by the second switch, and set a mapping between a port of the second virtual machine and the virtual network according to the association message;

The fourth processing subunit is configured to encapsulate information including a mapping relationship between the port and the virtual network as the second flow table.
An apparatus for implementing virtual machine interworking, comprising a third processing unit, a fourth processing unit, and a fifth processing unit,

The third processing unit is configured to: send a first packet to the first switch or the second switch, where the first packet includes a network protocol data packet and virtual local area network information;

The fourth processing unit is configured to: perform forwarding processing according to the first packet;

The fifth processing unit is configured to: perform forwarding processing between the first virtual machine and the first switch by using the first flow table, and perform forwarding processing between the second virtual machine and the second switch by using the second flow table, Implement virtual machine interworking;

The first flow table includes mapping information between the first virtual machine and a virtual network;

The second flow table includes mapping information between the second virtual machine and the virtual network.
The apparatus of claim 10 wherein said implementing virtual machine interworking comprises:

When the first virtual machine and the second virtual machine are in the same network segment, implementing Layer 2 interworking between the first virtual machine and the second virtual machine;

When the first virtual machine and the second virtual machine are in different network segments, the Layer 2 and/or Layer 3 interworking between the first virtual machine and the second virtual machine is implemented.
A system for implementing unified management and interworking of virtual machines, comprising: a controller, a first switch, and a second switch,

The controller is configured to: construct a first flow table and a second flow table, and send the first flow table to the first switch, and send the second flow table to the second switch; the first flow table includes Mapping information between a virtual machine and a virtual network, the second flow table including mapping information between the second virtual machine and the virtual network;

The first switch is configured to receive the first flow table sent by the controller, perform forwarding processing with the first virtual machine and the second switch by using the first flow table, and receive the first a packet; the first packet includes a network protocol packet and virtual local area network information;

The second switch is configured to receive the second flow table sent by the controller, perform forwarding processing with the second virtual machine and the first switch by using the second flow table, and receive First message.