CN105610614A - High availability access system and high availability fault switching method - Google Patents
High availability access system and high availability fault switching method Download PDFInfo
- Publication number
- CN105610614A CN105610614A CN201510997234.2A CN201510997234A CN105610614A CN 105610614 A CN105610614 A CN 105610614A CN 201510997234 A CN201510997234 A CN 201510997234A CN 105610614 A CN105610614 A CN 105610614A
- Authority
- CN
- China
- Prior art keywords
- node
- access
- sdn
- heartbeat
- module
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/10—Active monitoring, e.g. heartbeat, ping or trace-route
-
- 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
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0654—Management of faults, events, alarms or notifications using network fault recovery
- H04L41/0668—Management of faults, events, alarms or notifications using network fault recovery by dynamic selection of recovery network elements, e.g. replacement by the most appropriate element after failure
Abstract
The invention discloses a high availability access system. The system comprises an access resource pool; the access resource pool is composed of multiple to-be-accessed nodes; the system also comprises a heartbeat monitoring module for monitoring the node states in the access resource pool through periodically sending heartbeat packets, and an SDN module for directly receiving an access request from an external network, processing the access request according to the node states provided by the heartbeat monitoring module and directly forwarding to a suitable node in the multiple to-be-accessed nodes without using other proxy hosts. The invention also provides a high availability fault switching method.
Description
Technical field
The present invention relates to high available access system and high available failure switching method.
Background technology
" high availability " (HighAvailability) refers to that a system is through special design, reaches and reduces downtime, the continual effect of maintenance service. The height of service system is available conventionally to carry out active and standby switching by multiple redundant nodes and realizes, and the abstract model of its system as shown in Figure 1.
As shown in Figure 1, service end forms a resource pool by multiple access nodes. High-availability system outside access is changed and the resource pool that leads in certain available resource node. High-availability system externally issue a unification " virtual external address " so that client conduct interviews. In the time that client conducts interviews to service system by virtual address, high available front-end proxy agent is changed the destination address to this request of access, converts inner active node address to, and by the access stream actual active node that leads.
High-availability system can carry out status monitoring for movable service node by certain mechanism, in the time detecting that active node is delayed machine, high-availability system will be chosen secondary node from access resources pond, and access stream is switched to new active node, thereby reach the effect of continual service.
The available access scheme of existing height is the solution realizing based on main frame mostly, such as Heartbeat, Corosync, Keepalived etc. Its common implementation as shown in Figure 2. High available core function realizes by the proxy of front end. In the time that the access message of client arrives, the actual address that front-end proxy agent machine is internal activity main frame by the virtual external reference address transition in message, and again access message is encapsulated, E-Packet by the network equipment subsequently; In the time of internal node returned packet, proxy machine carries out reverse address transition again and adds decapsulation operation, and message is passed to client.
Meanwhile, front-end proxy agent machine also needs timing to carry out heartbeat health monitoring to the node in resource pool, and in the time finding that host node is delayed machine, proxy machine need to select remaining guest machine as new active node, and upgrade forwarding-table item, so that the message arriving afterwards can switch in time.
Such implementation can reach the effect of high available access, but has following deficiency:
First, functional module mixes, and implementation complexity is high. High available agent main frame has been realized the function of heartbeat health monitoring and network message address transition simultaneously. And the latter belongs to the operation of pure network level, it is more suitable to realize by the professional network equipment.
Secondly there is the risk of single point failure in front-end proxy agent main frame. First all external reference messages must carry out address transition through front-end proxy agent main frame before arriving network forwarding equipment, the machine once front-end proxy agent main frame is delayed, and whole system all will quit work.
Finally, in such scheme, the turnover of each primary traffic by undertaken by proxy network message address conversion, add the operation of the network level such as decapsulation and reverse proxy. The network equipment professional than direct application carries out this generic operation, and its stability and throughput performance are short of to some extent.
Summary of the invention
For addressing the above problem, according to an aspect of the present invention, provide a kind of height available access system, described system comprises: access resources pond, described access resources pond is made up of multiple nodes to be visited; Heartbeat inspecting module, for monitoring the node state in described access resources pond by regular transmission heartbeat packet; And SDN module, for directly receiving request of access from outer net, and the node state providing according to described heartbeat inspecting module and described request of access is processed and is directly forwarded to an applicable node in multiple nodes to be visited, and not through other proxies.
In the available access system of above-mentioned height, described SDN module comprises: SDN controller, and for receiving heartbeat inspecting abnormal information from described heartbeat inspecting module, and notify SDN switch to flow switching; And described SDN switch, for directly receiving request of access from outer net, and according to the notice of described SDN controller, an applicable node in described multiple node to be visited is processed and be directly forwarded to described request of access.
In the available access system of above-mentioned height, described SDN switch directly receives request of access and carries out message processing and forwarding from outer net, described heartbeat inspecting module and described SDN controller access from bypass as management node, and for the formulation of forwarding strategy with issue, and directly do not affect the forward-path of service traffics.
In the available access system of above-mentioned height, the first node in described multiple nodes to be visited is for being set to main service node, and described heartbeat inspecting module is monitored the state of described first node by regularly sending heartbeat packet via described SDN switch.
In the available access system of above-mentioned height, in the time that described heartbeat inspecting module monitors is delayed machine to described first node, it is configured to: select Section Point in described multiple node to be visited as mobile host computers, and by SDN controller described in this message informing.
In the available access system of above-mentioned height, described SDN controller is further configured to: issue the instruction of stream table change to described SDN switch, informing network is flowed through and forwarded by Section Point.
According to another aspect of the application, provide a kind of height available failure switching method, described method comprises: by regular transmission heartbeat packet, heartbeat inspecting module monitors exists abnormal as the first node of active node in access resource pool; Heartbeat inspecting module is chosen Section Point in described access resources pond as new active node, and by message informing SDN controller; SDN controller convection current table is changed, and issues the stream table of change to SDN switch; SDN switch is forwarded to described Section Point according to the stream table of change by the request of access receiving from outer net, and not through other proxies.
In the available failure switching method of above-mentioned height, the stream table that described SDN switch issues according to described controller, the message that data plane is flowed through forwards.
Than the available implementation of existing Host Based height, the application's technical scheme at least has following advantage:
First, the application's technical scheme makes full use of the advantage of SDN framework, the open network interface providing by SDN, network function module is separated from main frame and go out to transfer to the professional network equipment to process, thereby heartbeat inspecting module and mixed-media network modules mixed-media are carried out farthest separating coupling, simplify system, strengthened reliability simultaneously.
The second, all external reference messages are directly forwarded to final service node by the network equipment, without by intermediate host, have overcome thus the front-end proxy agent single point failure problem based in main frame implementation.
The 3rd, at aspect of performance, the application's technical scheme directly by the professional network equipment carry out message address conversion, add the operations such as decapsulation and reverse proxy, forwarding, can reduce the drop probabilities in failover process and improve throughput performance, the connection that reaches better high-availability system keeps, takes over seamlessly effect.
Brief description of the drawings
After having read the specific embodiment of the present invention with reference to accompanying drawing, those skilled in the art will become apparent various aspects of the present invention. Those skilled in the art should be understood that: these accompanying drawings are only for coordinating detailed description of the invention that technical scheme of the present invention is described, and are not intended to protection scope of the present invention to be construed as limiting.
Fig. 1 is the schematic diagram of the available access system of existing height;
Fig. 2 is the available implementation schematic diagram of existing Host Based height;
Fig. 3 is SDN model support composition;
Fig. 4 is the decomposition module figure of the front-end proxy agent machine in Fig. 2;
Fig. 5 is according to the application embodiment, decomposition module figure based on the high available realization of SDN framework;
Fig. 6 is according to the frame diagram of the application embodiment, high-availability system based on SDN framework; And
Fig. 7 is according to the application embodiment, the available failover schematic diagram of height based on SDN framework.
Detailed description of the invention
What introduce below is some in multiple possibility embodiment of the present invention, aims to provide basic understanding of the present invention, is not intended to confirm key of the present invention or conclusive key element or limits claimed scope. Easily understand, according to technical scheme of the present invention, do not changing under connotation of the present invention other implementation that one of ordinary skill in the art can propose mutually to replace. Therefore, below detailed description of the invention and accompanying drawing be only the exemplary illustration to technical scheme of the present invention, and should not be considered as of the present invention all or be considered as restriction or the restriction to technical solution of the present invention.
Network technology has greatly promoted the fast development of internet. But, traditional network technology is comparatively sealing always, and core technology and equipment Market are monopolized by several manufacturers, and the application on upper strata cannot direct control to the equipment below Internet. 2009, Stamford team has formally proposed software defined network (SoftwareDefinedNetworking, SDN) concept, be intended to break the closed situation of Internet, by the PN interface of standard is externally provided, application can be carried out alternately with network better, thus the global optimization performance of Hoisting System.
As shown in Figure 3, it is made up of SDN switch and SDN controller two parts typical SDN model. SDN controller is the center control nodes of whole network, is responsible for safeguarding the forwarding strategy of whole network, and issues the instruction of stream table by the management port of direct-connected each SDN switch. Each SDN switch is being stored a network flow table (FlowTable), the stream table strategy being issued to receive controller, and the message of data plane being flowed through by the list item in coupling stream table forwards.
Communication protocol between SDN controller and switch is called as south orientation agreement, and the south orientation agreement proposing is the earliest OpenFlow, has been developed at present 1.5 versions. In OpenFlow agreement, each stream list item (FlowEntry) is made up of parts such as classification and matching territory (MatchField) and operation rules (Instruction). The matched rule that the instruction of classification and matching word territory flows into message, once the match is successful, carries out the indicated operation of Instruction part. For example, in the stream table in Fig. 3, rule 1 represents that all messages that enter from port one all forward from port 2; The message that 2 instructions of stream table rule are 0.0.0.0 by all source addresses abandons. SDN switch operates all messages of flowing through according to these rules.
Continue with reference to figure 2, existing front-end proxy agent system can be divided into following two main functional modules, as shown in Figure 4:
1) heartbeat inspecting module: by the Host Status in regular transmission heartbeat packet monitoring resource pool;
2) network function module: the access message to client carries out address transition, and be guided to real work node.
Between two modules, be loose coupling relation, only, in the time that heartbeat monitoring modular is found host node abnormal state, just can carry out alternately, notifying it to flow switching with network function module.
Different from existing technical scheme, according to the application embodiment, in the module of the available realization of height based on SDN framework, still retain heartbeat inspecting module, network function module is isolated from front end proxy simultaneously, then realized by the network equipment of SDN architecture combined specialty. The system module block diagram of new departure as shown in Figure 5.
As shown in Figure 5, in one embodiment, SDN module is made up of network controller and SDN switch two parts. For the request of access of outer net, data message will directly carry out message by SDN switch to be processed and forwards, and needn't pass through other proxies. Heartbeat inspecting module and SDN controller are only to access from bypass as management node, are mainly responsible for the formulation of forwarding strategy and issue, and directly do not have influence on the forward-path of service traffics.
System module exploded view based on Fig. 5, wherein SDN module is carried out after instantiation, and the real system of this programme is disposed can be as shown in Figure 6. In one embodiment, SDN module is made up of network controller and SDN switch two parts. In the system starting stage, the node 1 that system is chosen in resource pool is main service node, and first network controller writes SDN switch by following two stream table rules:
Table 1: the stream table of system starting stage SDN switch
The meaning of stream rule 1 is that the message repeating of extranet access to node 1 is processed, and the destination address of message is changed to (address that is converted to node 1) simultaneously. 2 of stream rules are the B-rules of stream rule 1.
As shown in Figure 6, the access message of client will be directly sent to SDN switch, without preposition equipments such as process proxy machines. Mate for all messages that enter from port 0 according to stream rule 1, SDN switch, if destination address is vip, the destination address of message is converted to the address ip-1 of node 1, and from the port one of switch, message is produced.
Carry out reverse operation by flowing rule 2 for the message returning from node 1.
Meanwhile, heartbeat inspecting module is by regularly sending the state of heartbeat message monitoring node 1, and this process and above-mentioned outside message browsing process are completely independently.
Fig. 7 shows according to the application embodiment, the available failover schematic diagram of height based on SDN framework.
In conjunction with Fig. 7, according to the application embodiment, in the time that heartbeat monitoring modular detects that node 1 is delayed machine, system will be carried out following handling process:
(1) heartbeat inspecting module detects node 1 machine of delaying, and chooses secondary node 2 as mobile host computers, and by this message informing in network controller;
(2) SDN network controller issues stream table change directive to SDN interchanger, and the stream table after renewal is as shown in the table:
Table 2: the stream table of SDN switch after active and standby switching
(3) network flow is after this flowed through after SDN interchanger and is all forwarded via port 2, and active and standby switching completes.
Thus, the height that the application's technical scheme has realized connected reference by the framework of software defined network (SDN) is available, and can obtain the systematic function better than original Host Based high available solutions and the realization of more easy system.
In sum, the application has proposed a kind of available implementation of height based on software defined network (SDN) framework. The application's technical scheme makes full use of the advantage of SDN framework, the open network interface providing by SDN, network function module is separated from main frame and go out to transfer to the professional network equipment to process, thus the each module in high-availability system is carried out to maximum decoupling zero, simplify system, strengthened reliability simultaneously.
Above, describe the specific embodiment of the present invention with reference to the accompanying drawings. But those skilled in the art can understand, without departing from the spirit and scope of the present invention in the situation that, can also do various changes and replacement to the specific embodiment of the present invention. These changes and replacement all drop in the claims in the present invention book limited range.
Claims (8)
1. a high available access system, described system comprises:
Access resources pond, described access resources pond is made up of multiple nodes to be visited;
Heartbeat inspecting module, for monitoring described access resources pond by regular transmission heartbeat packetIn node state; And
SDN module, for directly receiving request of access from outer net, and according to described heartbeat prisonThe node state that provides of module is provided and to described request of access processes be also directly forwarded to manyAn applicable node in individual node to be visited, and not through other proxies.
2. the available access system of height as claimed in claim 1, wherein, described SDN moduleComprise:
SDN controller, for receiving heartbeat inspecting abnormal information from described heartbeat inspecting module,And notify SDN switch to flow switching; And
Described SDN switch, for directly receiving request of access from outer net, and according to describedThe notice of SDN controller and to described request of access process and be directly forwarded to described manyAn applicable node in individual node to be visited.
3. the available access system of height as claimed in claim 2, wherein, described SDN exchangeMachine directly receives request of access and carries out message processing and forwarding, described heartbeat inspecting mould from outer netPiece and described SDN controller access from bypass as management node, and for forwarding strategyFormulate and issue, and directly do not affect the forward-path of service traffics.
4. high available access system as claimed in claim 2 or claim 3, wherein, described multiple treatingFirst node in access node is for being set to main service node, and described heartbeat inspecting moduleMonitor the shape of described first node by regularly sending heartbeat packet via described SDN switchState.
5. the available access system of height as claimed in claim 4, wherein, when described heartbeat inspectingWhen module monitors is delayed machine to described first node, it is configured to:
Select Section Point in described multiple node to be visited as mobile host computers, and this is disappearedBreath is notified described SDN controller.
6. the available access system of height as claimed in claim 5, wherein, described SDN controlsDevice is further configured to:
Issue the instruction of stream table change to described SDN switch, informing network is flowed through by secondNode forwards.
7. a high available failure switching method, described method comprises:
By regular transmission heartbeat packet, heartbeat inspecting module monitors is to accessing in resource pool as livingThe first node of moving node exists abnormal;
Heartbeat inspecting module is chosen Section Point in described access resources pond as new active sectionPoint, and by message informing SDN controller;
SDN controller convection current table is changed, and issues the stream table of change to SDN switch;
SDN switch is forwarded to according to the stream table of change the request of access receiving from outer netDescribed Section Point, and not through other proxies.
8. the available failure switching method of height as claimed in claim 7, wherein, described SDNThe stream table that switch issues according to described controller, the message that data plane is flowed through turnsSend out.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510997234.2A CN105610614B (en) | 2015-12-28 | 2015-12-28 | High Availabitity accesses system and High Availabitity failure switching method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510997234.2A CN105610614B (en) | 2015-12-28 | 2015-12-28 | High Availabitity accesses system and High Availabitity failure switching method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105610614A true CN105610614A (en) | 2016-05-25 |
| CN105610614B CN105610614B (en) | 2019-06-18 |
Family
ID=55990160
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510997234.2A Active CN105610614B (en) | 2015-12-28 | 2015-12-28 | High Availabitity accesses system and High Availabitity failure switching method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105610614B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107086963A (en) * | 2017-03-10 | 2017-08-22 | 中国传媒大学 | A kind of accurate video switching method of destination based on SDN |
| CN116915837A (en) * | 2023-09-12 | 2023-10-20 | 苏州浪潮智能科技有限公司 | Communication method and communication system based on software defined network |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1725702A (en) * | 2004-07-20 | 2006-01-25 | 联想网御科技(北京)有限公司 | Network safety equipment and assemblied system and method for implementing high availability |
| US20120254204A1 (en) * | 2011-03-28 | 2012-10-04 | Microsoft Corporation | Techniques to manage file conversions |
| CN103795805A (en) * | 2014-02-27 | 2014-05-14 | 中国科学技术大学苏州研究院 | Distributed server load balancing method based on SDN |
| CN103929333A (en) * | 2014-05-08 | 2014-07-16 | 陈桂芳 | Implementation method for SDN controller pool |
-
2015
- 2015-12-28 CN CN201510997234.2A patent/CN105610614B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1725702A (en) * | 2004-07-20 | 2006-01-25 | 联想网御科技(北京)有限公司 | Network safety equipment and assemblied system and method for implementing high availability |
| US20120254204A1 (en) * | 2011-03-28 | 2012-10-04 | Microsoft Corporation | Techniques to manage file conversions |
| CN103795805A (en) * | 2014-02-27 | 2014-05-14 | 中国科学技术大学苏州研究院 | Distributed server load balancing method based on SDN |
| CN103929333A (en) * | 2014-05-08 | 2014-07-16 | 陈桂芳 | Implementation method for SDN controller pool |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107086963A (en) * | 2017-03-10 | 2017-08-22 | 中国传媒大学 | A kind of accurate video switching method of destination based on SDN |
| CN107086963B (en) * | 2017-03-10 | 2018-10-09 | 中国传媒大学 | A kind of accurate video switching method of destination based on SDN |
| CN116915837A (en) * | 2023-09-12 | 2023-10-20 | 苏州浪潮智能科技有限公司 | Communication method and communication system based on software defined network |
| CN116915837B (en) * | 2023-09-12 | 2024-01-26 | 苏州浪潮智能科技有限公司 | Communication method and communication system based on software defined network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105610614B (en) | 2019-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103125102B (en) | For providing the system and method for the Ethernet virtual concentrator scalability based on infinite bandwidth in middleware machine environment | |
| US8982727B2 (en) | System and apparatus of generalized network controller for a software defined network (SDN) | |
| EP2553886B1 (en) | Aggregating data traffic from access domains | |
| CN102197627B (en) | The improvement of flux of multicast convergence | |
| US9231849B2 (en) | Apparatus and method for controlling virtual switches | |
| US8438307B2 (en) | Method and device of load-sharing in IRF stack | |
| JP5715185B2 (en) | Packet processing method, apparatus and system | |
| CN104639464A (en) | System and method for realizing cross-interchanger link aggregation on OpenFlow interchanger | |
| US9692686B2 (en) | Method and system for implementing a multi-chassis link aggregation group in a network | |
| US20140029412A1 (en) | Systems and methods for providing anycast mac addressing in an information handling system | |
| EP2608459A2 (en) | Router, virtual cluster router system and establishion method thereof | |
| US9008080B1 (en) | Systems and methods for controlling switches to monitor network traffic | |
| CN103179046A (en) | Data center flow control method and data center flow control system based on openflow | |
| CN107078974A (en) | Multicast advertisements message for the network switch in storage area network | |
| US20140185429A1 (en) | Communication system, path switching method and communication device | |
| JPWO2014077313A1 (en) | COMMUNICATION SYSTEM, CONTROL DEVICE, ITS CONTROL METHOD, AND PROGRAM | |
| JP7092813B2 (en) | Packet transmission method and equipment | |
| US20150055451A1 (en) | Network Switching Systems And Methods | |
| KR20150007623A (en) | Method and appratus for protection switching in packet transport system | |
| EP3866393A1 (en) | Data center traffic exchange method and apparatus, device and storage medium | |
| CN103067291A (en) | Method and device of up-down link correlation | |
| CN104219115B (en) | It is a kind of to enable the method and system of Ethernet ring agreement and Spanning-Tree Protocol isomery mixed networking | |
| US9166868B2 (en) | Distributed control plane for link aggregation | |
| Du et al. | Software-defined networking for real-time ethernet | |
| KR102093296B1 (en) | Data processing system changing massive path time-deterministically and operating method of the system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |