CN110505095A - A method of large-scale virtual data center is built using a small amount of server - Google Patents
A method of large-scale virtual data center is built using a small amount of server Download PDFInfo
- Publication number
- CN110505095A CN110505095A CN201910793932.9A CN201910793932A CN110505095A CN 110505095 A CN110505095 A CN 110505095A CN 201910793932 A CN201910793932 A CN 201910793932A CN 110505095 A CN110505095 A CN 110505095A
- Authority
- CN
- China
- Prior art keywords
- virtual
- switch
- leaf
- virtual machine
- data packet
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 26
- 230000008569 process Effects 0.000 claims description 6
- 238000005538 encapsulation Methods 0.000 claims description 4
- 230000003068 static effect Effects 0.000 claims description 3
- 238000011161 development Methods 0.000 abstract description 2
- 230000006855 networking Effects 0.000 description 4
- 238000012795 verification Methods 0.000 description 2
- 241000037488 Coccoloba pubescens Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011981 development test Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000005641 tunneling Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/46—Interconnection of networks
- H04L12/4641—Virtual LANs, VLANs, e.g. virtual private networks [VPN]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Security & Cryptography (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention provides a kind of methods for building large-scale virtual data center using a small amount of server, belong to cloud deployment techniques field, leaf spine on each physical host is connected the leaf spine to form a logic by the present invention using VxLAN technology, all empty machines on each physical server are all connected on this logic leaf spine, to form a virtual data center.The development rate of cloud product is greatly speeded up.
Description
Technical Field
The invention relates to a cloud deployment technology, in particular to a method for building a large-scale virtual data center by using a small number of servers.
Background
Now, data centers are clouded, and how to verify the large-scale deployment of the cloud before the cloud is deployed to the data centers is a troublesome problem, mainly for the following reasons:
1) data centers require a large amount of hardware resources and few companies are available with these expensive resources to use only for experimental validation.
2) These hardware resources require a computer room, consume power, and few companies can have such conditions.
3) The hardware at the low end can only establish a simple network, and the network is not easy to change and is not flexible enough. Such as the leaf spine network, the configuration changes greatly from the L2 networking to the L3 networking.
The leaf-spine is a networking architecture of the data center, enables the network to be flat, has good horizontal expansibility, and well meets the requirements of the data center for east-west traffic.
An existing relatively inexpensive method is to rent the virtual data center according to the needs of the public cloud, and quit renting after the cloud is deployed and verified, which can greatly save the cost. But even then, the cost is burdensome for small companies or organizations.
Vxlan (virtual Extensible lan) is an overlay tunneling technique for tenant networks, encapsulating tenant L2 ethernet packets using UDP to span L3 networks.
Another approach is to connect several servers together using OpenvSwitch and VxLAN to form a large logical bridge, and virtual machines on all servers are hung on the large logical bridge, so as to form a large-scale virtual data center, but the problem is that it cannot simulate a leaf spin network architecture, and cannot configure various networking modes, such as layer two access to leaf switches, and layer three access to leaf switches and spin switches. More complex, layer three accesses to the leaf switch.
Disclosure of Invention
In order to solve the technical problems, the invention provides a method for building a large-scale virtual data center by using a small number of servers.
The technical scheme of the invention is as follows:
a method for building a large-scale virtual data center by using a small number of servers uses physical servers, each physical server can create a proper amount of virtual machines according to the number of available CPUs, each virtual machine is equivalent to one server, a leaf spine network can be simulated by using a VirtualBox network, the leaf spine networks on the physical servers are connected to form a large logical leaf spine network by using VxLAN technology, all the virtual machines on the physical servers are equivalent to be connected to a large leaf spine network to form a large virtual data center, and a cloud can be deployed to the virtual data center and can perform various experimental verifications.
Further, in the above-mentioned case,
the method mainly comprises the following steps:
1) emulating a switch/router using a virtual machine
2) Simulating an entire data center using a virtual machine
3) Building large-scale data center by using physical server
4) Cross-physical server stacking of virtual switches/routers is implemented using VxLAN technology.
Wherein,
the leaf switch is a virtual machine with 16 virtual network ports, and a Linux system is used as a switch operating system to process data packet switching. The ports corresponding to all the virtual machines hung on the ports belong to the same VirtualBox private network, the port for connecting the spin switch and the port on the corresponding spin switch belong to another VirtualBox private network, the trunk port of the leaf switch belongs to a VirtualBox public network, the switch needs to forward the data packet of the target non-local virtual machine to the spin switch and the trunk port, the trunk port encapsulates the data packet by using VxLAN and sends the encapsulated data packet to the spin switch of the opposite end, and the trunk port also decapsulates the data packet from the spin switch of the opposite end and then forwards the decapsulated data packet to the corresponding virtual machine.
The spine switch is a virtual machine with 8 virtual network interfaces, and a Linux system is used as a switch operating system to process data packet switching.
The network ports of two connected local leaf switches belong to the same virtual box private network, trunk ports belong to the same virtual box public network as the leaf switches, a spine switch needs to forward a data packet received by any port to other ports corresponding to a destination, the trunk ports need to be encapsulated, decapsulated and then forwarded, the spine switch needs to know which destination leaf switch the encapsulated data packet is sent to, and destination IP addresses of different destination encapsulations are different.
Each virtual machine is a virtual machine with only one network port, and the virtual machine is connected to one leaf switch.
The IP addresses of the virtual machines are statically configured, on which the Linux system runs, the cloud (Openstack) will treat it as one physical host/physical server, each virtual machine is assigned a different static IP, and the IP addresses of any two virtual machines cannot conflict.
The invention has the advantages that
1) The method develops a green channel for developers and small and medium-sized companies, and provides a cheap and convenient scheme for building a large-scale cloud data center.
2) The network and each resource node are all realized by using virtualization technology and software, so that the network and each resource node have great flexibility, can be changed at any time, and is time-saving and labor-saving.
3) And because a development test verification environment can be built locally and can be used anytime and anywhere, the development speed of cloud products is greatly increased, and the time for market promotion is shortened.
Drawings
FIG. 1 is a schematic layout of a virtual data center of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer and more complete, the technical solutions in the embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention, and based on the embodiments of the present invention, all other embodiments obtained by a person of ordinary skill in the art without creative efforts belong to the scope of the present invention.
The leaf spine on each physical host is connected using VxLAN technology to form a large logical leaf spine to which all virtual machines on each physical server are connected, thereby forming a large virtual data center. Fig. 1 visually shows this large virtual data center:
1) the leaf switch is a virtual machine with 16 virtual network ports, a Linux system is used as a switch operating system to process data packet switching, wherein ports corresponding to all the virtual machines hung on the leaf switch belong to the same VirtualBox private network, a port connecting the spine switch and a port on the corresponding spine switch belong to another VirtualBox private network, a trunk port of the leaf switch belongs to VirtualBox public network, the switch needs to forward data packets of a target non-local virtual machine to the spine switch and the trunk port, the trunk port encapsulates the data packets by using VxLAN and sends the data packets to the spine switch at an opposite end, and the trunk port also decapsulates the data packets from the spine switch at the opposite end and forwards the decapsulated data packets to the corresponding virtual machine. This Linux-based switch operating system can use a free version of Cumulus VX available from Cumulus to simplify the management configuration.
2) The spine switch is a virtual machine with 8 virtual network interfaces, a Linux system is used as a switch operating system to process data packet switching, network ports of two connected local leaf switches belong to the same virtual box private network, trunk ports belong to the same virtual box public network as the leaf switches, the spine switch needs to forward a data packet received by any one port to other ports corresponding to a destination, the trunk ports need to be encapsulated, decapsulated and then forwarded, the spine switch needs to know which destination leaf switch to send to after encapsulation, and destination IP addresses of encapsulation of different destinations are different. This Linux-based switch operating system can use a free version of Cumulus VX available from Cumulus to simplify the management configuration.
3) Each virtual machine is a virtual machine with only one portal (the management portal is a separate network and is not shown, but is not considered here), and the virtual machine is connected to a leaf switch. The IP addresses of the virtual machines are statically configured, on which the Linux system runs, the cloud (Openstack) will treat it as one physical host/physical server, each virtual machine is assigned a different static IP, and the IP addresses of any two virtual machines cannot conflict. The virtual machine on the physical server 2 must be configured with a different IP address than the virtual machine on the physical server 1.
Therefore, all virtual machines on the two physical servers form a large virtual data center, and the virtual data center can be expanded to more servers, so that the larger virtual data center is realized. For example, 20 virtual machines are started by one physical server, and a virtual data center with 100 nodes can be built by 5 physical servers. If a higher configuration physical server is used, if 8 paths of CPUs exist, each path of CPU 28 cores has 224 CPU cores, and then the hyper-thread is opened to start 200 virtual machines without problems, so that 5 physical servers can build a virtual data center with 1000 nodes.
The above description is only a preferred embodiment of the present invention, and is only used to illustrate the technical solutions of the present invention, and not to limit the protection scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (8)
1. A method for building a large-scale virtual data center by using a small number of servers is characterized in that,
the leaf spine on each physical host is connected by VxLAN technology to form a logical leaf spine, and all virtual machines on each physical server are connected to the logical leaf spine, so that a virtual data center is formed.
2. The method of claim 1,
the method mainly comprises the following steps:
1) emulating a switch/router using a virtual machine
2) Simulating an entire data center using a virtual machine
3) Building large-scale data center by using physical server
4) Cross-physical server stacking of virtual switches/routers is implemented using VxLAN technology.
3. The method of claim 1,
the leaf switch is a virtual machine with 16 virtual network ports, and a Linux system is used as a switch operating system to process data packet switching.
4. The method of claim 3,
the ports corresponding to all the virtual machines hung on the ports belong to the same VirtualBox private network, the port for connecting the spin switch and the port on the corresponding spin switch belong to another VirtualBox private network, the trunk port of the leaf switch belongs to a VirtualBox public network, the switch needs to forward the data packet of the target non-local virtual machine to the spin switch and the trunk port, the trunk port encapsulates the data packet by using VxLAN and sends the data packet to the spin switch of the opposite end, and the trunk port also decapsulates the data packet from the spin switch of the opposite end and then forwards the data packet to the corresponding virtual machine.
5. The method of claim 4,
the spine switch is a virtual machine with 8 virtual network interfaces, and a Linux system is used as a switch operating system to process data packet switching.
6. The method of claim 5,
the network ports of two connected local leaf switches belong to the same virtual box private network, trunk ports belong to the same virtual box public network as the leaf switches, a spine switch needs to forward a data packet received by any port to other ports corresponding to a destination, the trunk ports need to be encapsulated, decapsulated and then forwarded, the spine switch needs to know which destination leaf switch the encapsulated data packet is sent to, and destination IP addresses of different destination encapsulations are different.
7. The method of claim 6,
each virtual machine is a virtual machine with only one network port, and the virtual machine is connected to one leaf switch.
8. The method of claim 1,
the IP addresses of the virtual machines are statically configured, a Linux system is operated on the virtual machines, the cloud treats the virtual machines as a physical host/physical server, each virtual machine is allocated with a different static IP, and the IP addresses of any two virtual machines cannot conflict.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910793932.9A CN110505095B (en) | 2019-08-27 | 2019-08-27 | Method for building large-scale virtual data center by using small number of servers |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910793932.9A CN110505095B (en) | 2019-08-27 | 2019-08-27 | Method for building large-scale virtual data center by using small number of servers |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110505095A true CN110505095A (en) | 2019-11-26 |
| CN110505095B CN110505095B (en) | 2022-04-08 |
Family
ID=68589890
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910793932.9A Active CN110505095B (en) | 2019-08-27 | 2019-08-27 | Method for building large-scale virtual data center by using small number of servers |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110505095B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111030926A (en) * | 2019-12-20 | 2020-04-17 | 苏州浪潮智能科技有限公司 | Method and device for improving high availability of network |
| CN111597013A (en) * | 2020-05-20 | 2020-08-28 | 山东汇贸电子口岸有限公司 | Method for building large-scale virtual data center by using containers |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150103692A1 (en) * | 2013-10-15 | 2015-04-16 | Cisco Technology, Inc. | Host Traffic Driven Network Orchestration within Data Center Fabric |
| CN109218053A (en) * | 2017-07-03 | 2019-01-15 | 中兴通讯股份有限公司 | Implementation method, system and the storage medium of virtual data center |
| CN208939985U (en) * | 2018-09-26 | 2019-06-04 | 中国海洋石油集团有限公司 | CNOOC cloud Virtual Networking System |
-
2019
- 2019-08-27 CN CN201910793932.9A patent/CN110505095B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150103692A1 (en) * | 2013-10-15 | 2015-04-16 | Cisco Technology, Inc. | Host Traffic Driven Network Orchestration within Data Center Fabric |
| CN109218053A (en) * | 2017-07-03 | 2019-01-15 | 中兴通讯股份有限公司 | Implementation method, system and the storage medium of virtual data center |
| CN208939985U (en) * | 2018-09-26 | 2019-06-04 | 中国海洋石油集团有限公司 | CNOOC cloud Virtual Networking System |
Non-Patent Citations (2)
| Title |
|---|
| WEIXIN_41158656: "数据中心网络架构浅谈(三)", 《CSDN》 * |
| 肖永钦;卓柳迎: "SDN和OVERLAY技术在高校云数据中心的应用", 《武夷学院学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111030926A (en) * | 2019-12-20 | 2020-04-17 | 苏州浪潮智能科技有限公司 | Method and device for improving high availability of network |
| CN111597013A (en) * | 2020-05-20 | 2020-08-28 | 山东汇贸电子口岸有限公司 | Method for building large-scale virtual data center by using containers |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110505095B (en) | 2022-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11838395B2 (en) | Private ethernet overlay networks over a shared ethernet in a virtual environment | |
| US10476699B2 (en) | VLAN to VXLAN translation using VLAN-aware virtual machines | |
| CN114363021B (en) | Network target range system, virtual network implementation method and device of network target range system | |
| US11716285B2 (en) | Method and system for virtual port communications | |
| US9426060B2 (en) | Software defined network (SDN) switch clusters having layer-3 distributed router functionality | |
| US9178828B2 (en) | Architecture for agentless service insertion | |
| CN106685787B (en) | PowerVM (virtual machine) virtualized network management method and device based on OpenStack | |
| US20140096183A1 (en) | Providing services to virtual overlay network traffic | |
| CN106712988B (en) | A kind of virtual network management method and device | |
| CN109617995B (en) | Management system and method for VPC (virtual private network) internal container of tenant cluster and electronic equipment | |
| CN111064649B (en) | Method and device for realizing binding of layered ports, control equipment and storage medium | |
| CN104350467A (en) | Elastic enforcement layer for cloud security using SDN | |
| WO2015149253A1 (en) | Data center system and virtual network management method of data center | |
| CN108616386A (en) | A kind of construction method and SDN virtual network environments of SDN virtual network environments | |
| CN109547349A (en) | Flow managing method, device, terminal and storage medium based on virtual flow-line | |
| CN114389905B (en) | Network traffic statistics method, related device and medium | |
| US11652717B2 (en) | Simulation-based cross-cloud connectivity checks | |
| CN110505095B (en) | Method for building large-scale virtual data center by using small number of servers | |
| CN110311860A (en) | Multi-link load balance method and device under VXLAN | |
| CN111294268B (en) | Method and device for avoiding IP address conflict | |
| CN109756419A (en) | Routing iinformation distribution method, device and RR | |
| WO2023116268A1 (en) | Network isolation method and system, and proxy device | |
| CN113596192A (en) | Communication method, device, equipment and medium based on network gate networking | |
| CN111597013A (en) | Method for building large-scale virtual data center by using containers | |
| Hu et al. | NVLAN: a novel VLAN technology for scalable multi-tenant datacenter networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 250100 No. 1036 Tidal Road, Jinan High-tech Zone, Shandong Province, S01 Building, Tidal Science Park Applicant after: Inspur cloud Information Technology Co.,Ltd. Address before: 250100 No. 1036 Tidal Road, Jinan High-tech Zone, Shandong Province, S01 Building, Tidal Science Park Applicant before: Tidal Cloud Information Technology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |