CN101867521A - Multilink accessing and flow load dispatching managing method - Google Patents
Multilink accessing and flow load dispatching managing method Download PDFInfo
- Publication number
- CN101867521A CN101867521A CN201010200000A CN201010200000A CN101867521A CN 101867521 A CN101867521 A CN 101867521A CN 201010200000 A CN201010200000 A CN 201010200000A CN 201010200000 A CN201010200000 A CN 201010200000A CN 101867521 A CN101867521 A CN 101867521A
- Authority
- CN
- China
- Prior art keywords
- link
- flow
- inserts
- module
- insert
- 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.)
- Pending
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides a network multilink accessing and flow load dispatching managing method. The method combinatively comprises the following steps that: 1) a link accessing module de-encapsulates a received packet and provides an interface to upper application; 2) a link accessing and load balancing module parallelizes the user link accessing and flow dispatching management in a distributing way; and 3) a link flow controlling module controls the flow. The method is realized by the link accessing module, the link accessing and load balancing module and the ink flow controlling module. The method supports large-scale link accessing, can effectively reduce the link access time delay of data, improves the smoothness of the flow of the link and the use ratio of the whole bandwidth of the system, and is suitable for controlling the flow of the accessed link of high-performance wide band.
Description
Technical field
The invention belongs to broadband network inserts and the flow control technique field.
Background technology
The broadband network engineering is embodied in service layer and not only requires broadband access network but also the high network bandwidth should be provided, Networks such as voice-bearer, video, data simultaneously, and have and can provide the ability of differentiated service according to client and application demand.From development of internet technology, the growth of the network bandwidth makes the exchanges data of server end and control become a bottleneck, and can effectively alleviate the loading problem of control gateway node and network egress link by flow scheduling and bandwidth control.
Existing broadband network hardware access product does not have personalized flow control and the allocated bandwidth function based on the role, therefore in the real network management, lack effective management technical means, domestic each big operator even ferment the problems such as illegal parallel operation, abuse Internet resources and internet worm that the charge on traffic scheme becomes increasingly conspicuous with reply.The hardware construction is only paid attention in the networking of China, the operating scheme of often having ignored whole system, after the hardware construction is intact, still there are many problems, all types campus network and commercialization sub-district ubiquity the small part user and are abused Internet resources, seriously expend the phenomenon of the network bandwidth, seriously blocked network line, cause exception of network traffic to increase, thereby influenced other Internet users' proper network visit.
Along with broadband networks inserts user's growth and becoming increasingly abundant of Internet application, its flow is also increasing sharply, the network congestion phenomenon is more and more serious, therefore network operation presses for and rationally utilizes Internet resources, and in network, carry out accurate flow control, could eliminate network congestion and bottleneck, but have only to each Internet user carry out different flow controls just can reach to broadband network can management, the purpose of may command and O﹠M.
Along with increasing user's accessing to wide band network, user's the load that must cause access server that increases increases.Communication optimization between the client can alleviate the load of tunnel server to a certain extent, but the main composition of server load is user and other data between nodes stream, and it is very limited therefore to utilize client communication optimization to alleviate the effect of server load.In addition, by upgrading hardware, increase the network bandwidth or adopt centralized load-balancing mechanism can improve the disposal ability of single access server, but the extensibility of these modes is limited, still can not satisfy system to the ever-increasing demand of load disposal ability.
Summary of the invention
The objective of the invention is to overcome the deficiencies in the prior art, provide a kind of network multilink to insert and the flow load schedule management method.
In order to realize the foregoing invention purpose, the technical scheme of employing is as follows:
A kind of multilink inserts and the flow load schedule management method, and this method is by one or more following steps realization that combines:
1) inserts processing module by link the UDP message bag that receives is carried out the decapsulation processing, and provide interface to upper layer application;
2) insert load balancing module to user link access and the distributed parallel processing of flow scheduling management carrying out by link;
3) carry out flow control by the link flow control module.
Described link inserts processing module and realizes that with the form that virtual network interface drives it loads automatically at kernel spacing when os starting.
Described link inserts processing module and comprises to the interface that upper layer application provides: link configuration order, client interactive program and link server interactive program insert processing module by I/O control channel to link and operate.
The present invention adopts the link configuration order to increase by the link access I/O control channel that processing module provided or deletes link, and revises or the demonstration link parameter.
Described client interactive program or link server interactive program insert the I/O control channel that processing module provided by link and carry out the establishment of client address configuration and link.
Described link inserts load balancing module and comprises a plurality of processing nodes, and according to processing node and link load situation, can expand to any a plurality of processing node, the link access of each processing node and flow scheduling are by Control Node and Redundant Control node control, and every user link can be connected on any one processing node.It is the processing procedure that client and processing node connect by authentication that link inserts, and after this process completes successfully, will set up a user link between client and processing node.Processing node is responsible for setting up user link with client, and Control Node is responsible for controlling certain bar link and is linked certain link and insert on processing node and carry out flow scheduling.
Described link inserts load balancing module and also is provided with a scheduler, realizes user's access request is dispatched, and for the user selects a load lighter access server relatively, thereby realizes the load balancing between a plurality of access servers.
Described link flow control module comprises formation adjuster and shared token Buffer Pool, described formation adjuster is formation of each link data flow distribution, when data arrive, be introduced into and carry out traffic shaping and queuing in the formation adjuster, be admitted to the shared token Buffer Pool again, described shared token Buffer Pool is according to the number of first come first served basis and available tokens, for each enters the data allocations token of Buffer Pool, the data that will obtain token again are sent to outlet, to adjust the bandwidth of each link data stream.Because it is the processing procedure that client and processing node connect by authentication that link inserts, and after this process completes successfully, will set up a user link between client and processing node.So carrying out data transmit-receive, user side can on this user link, produce link data stream.
Realize that method of the present invention inserts processing module, link access load balancing module, the realization of link flow control module by link, the present invention supports extensive link to insert, can reduce data effectively delays time at link-access, improve flatness, the entire system bandwidth availability ratio of link flow, be suitable for the application of high performance wideband access link flow control.
Description of drawings
Fig. 1 is for realizing system architecture diagram of the present invention;
Fig. 2 realizes schematic diagram for link inserts;
Fig. 3 is the extensive link access model of distributed parallel figure;
Fig. 4 inserts load-balancing method figure for multiprocessing node scheduling link;
Fig. 5 is bandwidth controlling model figure under the multilink environment.
Embodiment
The present invention is described further below in conjunction with accompanying drawing.
Realize system architecture diagram of the present invention as shown in Figure 1, form by link access processing module, link access load balancing module, link flow control module.
As shown in Figure 2, link inserts processing module two kinds of implementations, realizes in the system kernel space or realizes at user's space.The former characteristics are that it and upper level applications are separate, and chain circuit function can be by all programs rather than by certain specific routine call, and these programs do not need to make any modification for this reason, i.e. the existence of application programs link is transparent.In this manner, the LI(link interface) that application program employing system provides communicates, and sends message and encapsulates automatically through LI(link interface), sends after becoming a message again; The latter is used for data forwarding, at first set up raw socket and receive the link layer data, if receive a packed packet, obtain a packet after then carrying out decapsulation, this realization exists as the part of application program, and it can not be that other application programs are used that encapsulation function is conciliate in its encapsulation.From improving the link server forwarding performance, realize that at kernel spacing chain circuit function is obviously than realizing more effective at user's space.Therefore, on the data link layer ppp protocol, add one deck processing procedure, protocol data is transmitted, operations such as the foundation of data link, dismounting are finished by ppp protocol, function such as data forwarding and Route Selection then realizes in core stage, and develops one and monitor finger daemon, realized the driver and the socket of core schema at kernel, by using the socket interface to read and send packet, realize that the large-scale user of single device support inserts.
Shown in accompanying drawing 3 and accompanying drawing 4, adopt distributed multiserver structure to solve the scalability problem that system causes because of the user increases.Disposing a plurality of access servers in network, is that state is irrelevant between each server, can be independently for the user provide access service, and these servers are positioned at different places, have the distributed nature on the geographical position.Under this network configuration, can satisfy the access demand that the user increases day by day by disposing the whole volume that new access server improves system.But because access server is towards the user who disperses, the number of users of selecting certain server to insert is not fixed, and unlike the roughly number of common access server that works estimating user, and then determines requirement to the server key property.Under the uncertain situation of user's access number, increase a load dispatch device and solve distributed multiserver structure and cause system the problem of load imbalance to occur easily, it is overweight to solve some server load, and the phenomenon of other resource surplus.
As shown in Figure 5, bandwidth control implementation method is the critical field according to packets such as source/destination interface, source/destination addresses under the multilink environment, the network bandwidth is classified, limit the speed that packet sends in each classification, thereby reach the Control Network flow, ensure the purpose of key business bandwidth.Bandwidth control is made up of queue discipline module, sort module and filtering module.
All bind a queue discipline on each user link logic interfacing, be used to control data flow through this logic interfacing.CBQ, HTB can classify in linux kernel, and they can be for each logical links be provided with priority, and what priority was high sends earlier; Can also realize link sharing, when a classificatory bandwidth is idle, can use classification to other; Can also control burst flow, limiting speed etc.
At first create a trunk classification in this structure, it has whole bandwidth, creates a plurality of classification again in the trunk classification, and each classification can be created their subclass again.Their main effect is that data flow is classified according to various conditions, is convenient to they are carried out different control.Each classification at each Different Logic interface has all distributed certain bandwidth value, and the bandwidth summation that belongs to all subclasses of same parent can not surpass the bandwidth value of parent.Classification just looks like to be to have set up several channels for packet, and the flow direction of packet is then decided by filtering rule.
Therefore can authenticate and property control by radius server, must add following content in/etc/ppp/options the inside:
plugin/usr/lib/pppd/2.4.4/radius.so
plugin/usr/lib/pppd/2.4.4/radattr.so
In radius server, at first to define user's upstream and downstream bandwidth attribute, here be respectively Download and Upload, just can control by revising this property value the online bandwidth of different user, can check by radius server and return to this attribute of user at/var/run/radattr.ppp* after the subscriber dialing successful connection, specific as follows:
Framed-IP-Address?192.168.17.231
Framed-MTU?576
Service-Type?Framed-User
Framed-Protocol?PPP
Framed-Compression?Van-Jacobson-TCP-IP
Download?2048
Upload?1024
Burst?4096
Idle-Timeout?60
Session-Timeout?6350086
Editor/etc/ppp/ip-up.local file then, allow each dial-links when setting up, carry out below based on the flow control and the allocated bandwidth script of user role, specific procedure is as follows:
#！/bin/sh
#$1＝interface-name
#$2＝tty-device
#$3＝speed
#$4＝local-IP-address
#$5＝remote-IP-address
#$6＝ipparam
DOWN＝`cat/var/run/radattr.$1|grep’Download’|cut-d”-f?2`
UP＝`cat/var/run/radattr.$1|grep’Upload’|cut-d”-f?2`
Buff＝`cat/var/run/radattr.$1|grep’Burst’|cut-d”-f?2`
#Limit?Download?Bandwidth
/sbin/tc?qdisc?add?dev$1?root?handle?1:htb?default?1
/sbin/tc?class?add?dev$1?parent?1:classid?1:1?htb?rate${DOWN}kbitceil?${DOWN}kbit?burst?${Buff}k
#Limit?Upload?Bandwidth
/sbin/tc?qdisc?add?dev?$1?handle?ffff:ingress
/sbin/tc?filter?add?dev?$1?parent?ffff:protocol?ip?prio?50?u32?matchip?src?$5/32?police?rate?${UP}kbit?burst?${Buff}k?drop?flowid:1
Claims (8)
1. a multilink inserts and the flow load schedule management method, it is characterized in that this method is by one or more following steps realization that combines:
1) inserts processing module by link the packet that receives is carried out the decapsulation processing, and provide interface to upper layer application;
2) insert load balancing module to user link access and the distributed parallel processing of flow scheduling management carrying out by link;
3) carry out flow control by the link flow control module.
2. multilink according to claim 1 inserts and the flow load schedule management method, it is characterized in that described link inserts processing module and realizes that with the form that virtual network interface drives it loads automatically at kernel spacing when os starting.
3. multilink according to claim 1 and 2 inserts and the flow load schedule management method, and it is characterized in that described link inserts processing module and comprises to the interface that upper layer application provides: link configuration order, client interactive program and link server interactive program insert processing module by I/O control channel to link and operate.
4. multilink according to claim 3 inserts and the flow load schedule management method, it is characterized in that adopting the link configuration order to insert I/O control channel that processing module provided by link increases or delete link, and revises or the demonstration link parameter.
5. multilink according to claim 3 inserts and the flow load schedule management method, it is characterized in that described client interactive program or link server interactive program insert the I/O control channel that processing module provided by link and carry out the establishment of client address configuration and link.
6. multilink according to claim 1 inserts and the flow load schedule management method, it is characterized in that described link inserts load balancing module and comprises that a plurality of links insert processing node, and according to link access processing node loading condition, can expand to any a plurality of link and insert processing node, Control Node is responsible for controlling certain bar link and is linked certain link and insert on processing node and carry out flow scheduling, by the control of Control Node, link can be connected to the lighter link of certain load and insert on the processing node.
7. insert and the flow load schedule management method according to claim 1 or 6 described multilinks, it is characterized in that described link inserts load balancing module and also is provided with a scheduler, realization is dispatched user's access request, for the user selects a load lighter access server relatively, thereby realize the load balancing between a plurality of access servers.
8. multilink according to claim 1 inserts and the flow load schedule management method, it is characterized in that described link flow control module comprises formation adjuster and shared token Buffer Pool, described formation adjuster is formation of each link data flow distribution, when data arrive, be introduced into and carry out traffic shaping and queuing in the formation adjuster, be admitted to the shared token Buffer Pool again, described shared token Buffer Pool is according to the number of first come first served basis and available tokens, for each enters the data allocations token of Buffer Pool, the data that will obtain token again are sent to outlet, to adjust the bandwidth of each link data stream.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010200000A CN101867521A (en) | 2010-06-11 | 2010-06-11 | Multilink accessing and flow load dispatching managing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010200000A CN101867521A (en) | 2010-06-11 | 2010-06-11 | Multilink accessing and flow load dispatching managing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101867521A true CN101867521A (en) | 2010-10-20 |
Family
ID=42959091
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010200000A Pending CN101867521A (en) | 2010-06-11 | 2010-06-11 | Multilink accessing and flow load dispatching managing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101867521A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103247003A (en) * | 2013-05-15 | 2013-08-14 | 上海大智慧股份有限公司 | Distributed program trading system oriented to event processing |
| CN104735724A (en) * | 2013-12-20 | 2015-06-24 | 中兴通讯股份有限公司 | Load balancing processing method and device, wireless access node and user equipment |
| CN105743999A (en) * | 2016-04-05 | 2016-07-06 | 广东石油化工学院 | Distributed link access bandwidth control system |
| CN106982170A (en) * | 2017-04-19 | 2017-07-25 | 上海海加网络科技有限公司 | Multi-link load balance method and device based on linux system |
| CN109547517A (en) * | 2017-09-22 | 2019-03-29 | 贵州白山云科技股份有限公司 | A kind of bandwidth scheduling method and apparatus |
| WO2021114930A1 (en) * | 2019-12-12 | 2021-06-17 | 中兴通讯股份有限公司 | Network slice message transmission method, electronic device and storage medium |
| CN113726473A (en) * | 2020-05-25 | 2021-11-30 | 华为技术有限公司 | Method for interacting signaling information in wireless local area network and communication device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1553691A (en) * | 2003-05-26 | 2004-12-08 | ��Ϊ��������˾ | High-capacity wide-band inserting method and system |
| EP1551137A1 (en) * | 2003-12-31 | 2005-07-06 | Alcatel | Parallel data link layer controllers in a network switching device |
-
2010
- 2010-06-11 CN CN201010200000A patent/CN101867521A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1553691A (en) * | 2003-05-26 | 2004-12-08 | ��Ϊ��������˾ | High-capacity wide-band inserting method and system |
| EP1551137A1 (en) * | 2003-12-31 | 2005-07-06 | Alcatel | Parallel data link layer controllers in a network switching device |
Non-Patent Citations (1)
| Title |
|---|
| 《微电子学与计算机》 20090405 梁根 等 分布式并行链路接入及流量负载调度模型 第26卷, 第4期 2 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103247003B (en) * | 2013-05-15 | 2017-07-07 | 上海大智慧股份有限公司 | Towards the distributed program transaction system of event handling |
| CN103247003A (en) * | 2013-05-15 | 2013-08-14 | 上海大智慧股份有限公司 | Distributed program trading system oriented to event processing |
| CN104735724B (en) * | 2013-12-20 | 2019-07-02 | 中兴通讯股份有限公司 | Load balance process method, apparatus, radio access node and user equipment |
| CN104735724A (en) * | 2013-12-20 | 2015-06-24 | 中兴通讯股份有限公司 | Load balancing processing method and device, wireless access node and user equipment |
| CN105743999A (en) * | 2016-04-05 | 2016-07-06 | 广东石油化工学院 | Distributed link access bandwidth control system |
| CN106982170A (en) * | 2017-04-19 | 2017-07-25 | 上海海加网络科技有限公司 | Multi-link load balance method and device based on linux system |
| CN109547517A (en) * | 2017-09-22 | 2019-03-29 | 贵州白山云科技股份有限公司 | A kind of bandwidth scheduling method and apparatus |
| CN109547517B (en) * | 2017-09-22 | 2021-11-09 | 贵州白山云科技股份有限公司 | Method and device for scheduling bandwidth resources |
| WO2021114930A1 (en) * | 2019-12-12 | 2021-06-17 | 中兴通讯股份有限公司 | Network slice message transmission method, electronic device and storage medium |
| CN112994908A (en) * | 2019-12-12 | 2021-06-18 | 中兴通讯股份有限公司 | Network slice message transmission method, electronic equipment and storage medium |
| CN112994908B (en) * | 2019-12-12 | 2024-01-23 | 中兴通讯股份有限公司 | Network slice message transmission method, electronic equipment and storage medium |
| CN113726473A (en) * | 2020-05-25 | 2021-11-30 | 华为技术有限公司 | Method for interacting signaling information in wireless local area network and communication device |
| WO2021238578A1 (en) * | 2020-05-25 | 2021-12-02 | 华为技术有限公司 | Method for signaling information exchange in wireless local area network, and communication apparatus |
| CN113726473B (en) * | 2020-05-25 | 2023-09-08 | 华为技术有限公司 | Method for interacting signaling information in wireless local area network and communication device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101867521A (en) | Multilink accessing and flow load dispatching managing method | |
| CN1206600C (en) | Full distribution type aggregation network servicer system | |
| US20210243247A1 (en) | Service mesh offload to network devices | |
| US7751409B1 (en) | Logical service domains for enabling network mobility | |
| JP6188093B2 (en) | Communication traffic processing architecture and method | |
| US20020188732A1 (en) | System and method for allocating bandwidth across a network | |
| WO2014082538A1 (en) | Business scheduling method and apparatus and convergence device | |
| CN100420249C (en) | Method for guarantee service quality of radio local network | |
| US20120198032A1 (en) | Communication system | |
| CN106059821A (en) | Software-defined network (SDN)-based data center service quality assurance method | |
| CN102170396A (en) | QoS control method of cloud storage system based on differentiated service | |
| CN111355658B (en) | SDN cross-domain cooperation method based on distributed service framework | |
| CN1674576A (en) | Method for transmitting strategic information inter-network equipment | |
| USRE44798E1 (en) | Cross-layer architecture for a network device | |
| CN107919982A (en) | A kind of DCI management platforms and its management method | |
| CN105871751A (en) | Method and system for guaranteeing network bandwidth of data center | |
| CN103747274A (en) | Video data center with additionally-arranged cache cluster and cached resource scheduling method thereof | |
| CN105262703A (en) | Routing bandwidth allocation method based on double priorities | |
| CN1284094C (en) | Distributed parallel scheduling wide band network server system | |
| CN103685065A (en) | Communication scheduling method based on time | |
| CN110177056B (en) | Automatic adaptive bandwidth control method | |
| CN1518296A (en) | Method of implementing integrated queue scheduling for supporting multi service | |
| CN100536434C (en) | A QOS-based self-adapted scheduling method based on IP DSLAM | |
| CN110213334A (en) | A kind of transmission method and device of shared file | |
| CN100591024C (en) | Mobile access controller, mobile locak area network and metropolitan area network, and access method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20101020 |