CN102075444A - Network system and method for guaranteeing multi-type service quality - Google Patents
Network system and method for guaranteeing multi-type service quality Download PDFInfo
- Publication number
- CN102075444A CN102075444A CN201110047722.9A CN201110047722A CN102075444A CN 102075444 A CN102075444 A CN 102075444A CN 201110047722 A CN201110047722 A CN 201110047722A CN 102075444 A CN102075444 A CN 102075444A
- Authority
- CN
- China
- Prior art keywords
- priority
- virtual link
- network
- link
- virtual
- 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
Images
Abstract
The invention discloses a network system and a method for guaranteeing multi-type service quality. A service perception module and a pre-labeling module are arranged in a source host and used for perceiving sent service data packets and pre-labeling the grouped data packets with different semantic priorities for sending according to the perception information when a network layer is encapsulated. An edge router and a core router are provided with a multi-service agent respectively. The edge router is also provided with an acceptance controller for mapping the semantic priorities of the service data packets into network priorities and distributing the service data packets to the corresponding service agents. The homogeneous service agents in all the routers are interconnected to form virtual topologies, and each virtual topology is provided with an independent routing table for routing selection of the service data packets of different priorities on the corresponding virtual topologies respectively. Service routing of high priority is guaranteed, and influence on low priority services is reduced as much as possible; and the invention also provides a method for setting link weights, which distributes the network traffic to each link more uniformly and improves the throughput rate.
Description
Technical field
The present invention relates to a kind of network system and method that ensures the polymorphic type QoS, exactly, relate to and a kind of the network routing node is abstracted into a plurality of service agent nodes according to type of service, make service agent node of the same type constitute virtual topology, and utilize and a kind ofly brand-new calculate network system and the method that link metric in the virtual topological structure and link available bandwidth and link utilization thereof carry out route based on context perception method, belong to Internet technical field.
Background technology
Shortest path Routing Protocol (as OSPF, IS-IS) is a Routing Protocol in the present prevailing territory, is widely used in the ISP backbone network.In these Routing Protocols, the weight of link all is that network manager is predefined, and remains unchanged in network operation process usually.Network traffics are according to the shortest path route from the source node to the destination node.Because the weight relative fixed of link, no matter how network traffics change, and the path between same source node and the destination node is fixed.Cisco has proposed a kind of scheme that ospf protocol link weight is set, and the weight of every link is set to the inverse of link bandwidth capacity.Intermediate System to Intermediate System is all link weights to be reseted be changed to 10.The weight set-up mode of this simple static can not be made adaptive adjustment according to dynamic network topology, flow rate mode and flow distribution, causes network congestion easily.The simple modifications scheme of the default fixed weight plan of establishment is: it is proportional that link weight is reseted the flow distribution that is set to link.The restriction of this scheme is that prioritization scheme is based on fixing network topology and fixing traffic demand matrix, and the weight of link still remains unchanged in the running of networking.As everyone knows, the flow height of Internet is dynamic, and traffic demand and network topology all constantly change, so this class scheme does not have generality.
Dynamically the self adaptation route technology (referring to shown in Figure 1) of adjusting the link weight can be adjusted the link weight according to link load in real time adaptively, thereby the Control Network flow distribution prevents network congestion, improves network performance.The variation of flow distribution can cause link load to change again, and the link weight is got back and adjusted accordingly, thereby forms feedback loop.
The key of dynamically adjusting the self adaptation route of link weight is the suitable weight mapping function of design, the variation of network traffics is reflected to the setting of link weight.Usually the excessive variation of link weight can cause the changes in flow rate on the link, and the feedback effect that is produced causes network traffics to be vibrated between different paths, seriously reduces network performance; Yet, too small if the link weight changes, can cause to network traffics change insensitive, the adaptability of agreement descends.Therefore must mapping scheme reasonable in design, to a certain degree compromise of demand makes this method both can reflect the dynamic of network traffics, can not cause route flapping again simultaneously.
MDAR (Minimal Delay Adaptive Routing) is the early stage a kind of adaptability routing algorithm of realizing at Advanced Research Projects Agency Network that proposes: the proportional weight calculation formula of link weighted value and link delay is set is: link weight w=w
0+ D=w
0+ d (u), in the formula, w
0Be constant, D is the function d (u) of link utilization u; Because of the lag characteristic of link, function d (u) is a nonlinear function.Experimental result shows that when offered load was light, the flow in the network was more stable; When offered load was heavier, the link weight changed greatly, caused the route flapping phenomenon serious.LSAR (Load Sensitive Adaptive Routing) proposes a kind of agreement of using the Linear Mapping scheme, introduces damping factor, to allow burst flow and to prevent route flapping.When link utilization is low, introduce the non-habitual zone, reduce the expense of network routing update.
The method of the calculating link route of above-mentioned several schemes is not all distinguished the type of service on the link, just the service traffics of all priority on the link is done as a whole the processing.When this may cause the big and weights link that the high-priority service flow is less of certain bar low priority traffice flow to change, the service quality of high-priority service is caused excessive influence.For fear of this thing happens, when calculating link metric, should consider the priority difference of various different business.At present, academia has launched extensive studies for how link metric is set effectively, because effectively being set, link metric carries out route, can make full use of Internet resources, thereby improve the transmission quality of Business Stream and the QoS of network, this problem has become the current urgent problem of scientific and technical personnel in the industry.
Summary of the invention
In view of this, the purpose of this invention is to provide a kind of route system and method that ensures the service quality of polymorphic type business.In routing procedure, this system carries out abstract to the physics routing node of network according to business classification, each physics routing node is abstracted into a plurality of service agent nodes by type of service, and each service agent node is responsible for handling a class business.The property line of reasoning is formed separately virtual topology by the similar service agent node in the node, such professional just in the enterprising walking along the street of this topology by selection.For the business of different priorities, in the professional route that ensures high priority, reduce influence to low priority traffice as far as possible.The present invention also proposes a kind of mode that link metric is set, and allows network traffics be allocated in each link more equably, improves network throughput.
In order to achieve the above object, the invention provides a kind of network system that ensures the polymorphic type QoS, comprise: send and receive the source end main frame and the end host of business data packet, and edge router and the core router of bearing transfer of data and forwarding; It is characterized in that: in the described system,
Source end main frame, set up professional sensing module and preliminary making module, the former is used for the business data packet that sends is carried out perception,, sends after according to perception information business data packet being labeled as different semantic priority in advance when the network layer encapsulation business data packet for the latter;
Edge router, set up and admit controller and a class or multiclass service agent, the semantic priority mapping that the former is used for the business data packet that will receive is a network priority, and with the allocation of packets of heterogeneous networks priority to the service agent of handling the respective priority business; Latter's kind depends on the number of degrees of the network priority that this system supports, every class service agent is responsible for handling a class priority service, and the same line of business in each edge router and core router agency interconnects and forms separately virtual topological structure, each virtual topological structure is provided with separately independently routing table, is carrying out route in the corresponding virtual topological structure separately respectively for the business data packet of heterogeneous networks priority; Transmit the core router of business data packet and only set up a class or multiclass service agent, its function is identical with edge router.
In order to achieve the above object, the present invention also provides a kind of method that adopts network system of the present invention to ensure the polymorphic type QoS, it is characterized in that: source end main frame is according to the semantic priority of application layer semantic differentiation business data packet, and carries out corresponding preliminary making when network layer encapsulates; When business data packet arrives Differentiated Services DiffServ network, the ingress edge router is network priority according to setting strategy with the semantic priority mapping of the preliminary making of each packet grouping, again according to its network priority select to comprise to traditional Best-Effort service packet carry out forwarded hop-by-hop with to the different routing policies of the high-priority service packet calculating optimum route that QoS ensures are arranged: edge router and core router are also collected the business data packet contextual information respectively, select corresponding business agency and routing policy to send it to end host according to the network priority and the current network resources of business data packet.
The present invention ensures that the network system of polymorphic type QoS and method thereof have three advantages:
System of the present invention is abstracted into some agent nodes with physical router, the network of forming virtual topological structure by the same line of business agent node in each router, again the traffic assignments of different priority levels is transmitted in corresponding virtual topological structure, with convenient professional management.
The method for routing of this system is based on context-aware, contextual information according to business data packet is divided network priority with each grouping of business data packet according to the various application messages of class of service, each grouping relative position and each grouping size, take into full account the importance of each grouping for whole service transmission quality, in the service quality that ensures high-priority service simultaneously, do not influence low priority traffice service quality as far as possible.
Another advantage is that the present invention adopts a kind of new link metric setting means, make network traffics can rationally be distributed on each bar virtual link, can either improve network resource utilization like this, can ensure the service quality of polymorphic type business again, when ensureing the high-priority service transmission performance, reduce its influence to low priority traffice as far as possible.
Key problem in technology innovation part of the present invention is:
Introduce the virtual network topology structure, adopt virtual link weights set-up mode to guarantee the service quality of multiple priority service.Wherein core technology is: each router or switching node in network system of the present invention are introduced all kinds of service agents according to different service types, and form virtual topological structure by each same line of business agency, the business datum of same kind wraps in route in the virtual topological structure that this business belongs to.
Introduce local path information, be used for the situation of the flow distribution of dynamically recording upstream link, and propose to utilize the method for local path information calculations virtual link utilance at every downstream links.This method can be given upstream link to the dynamical feedback that influences of downstream links with the flow of upstream link, feedback method is adding up and comparing the link utilization of the link utilization of upstream link and its all downstream links, if add up with surpass the link utilization of upstream link, just think that the flow of upstream link is excessive for the influence of downstream links, will be increased to the link utilization of upstream link this moment adds up and is worth; Otherwise, can keep the link utilization of this upstream link constant.
Introducing utilizes the virtual link utilance dynamically to generate and be provided with the method for virtual link weights, this method adjusts link metric according to current link flow situation and service priority, corresponding Route Selection also can be upgraded thereupon, so just can avoid causing route oscillation, can guarantee the service quality of polymorphic type business because of the virtual link weights change.Simultaneously, this method dynamically is embodied in link utilization on the link metric, when link utilization increases, improves link metric, thereby reduces this link selected probability next time, and flow is directed in the lower link of load.
Description of drawings
Fig. 1 is a schematic diagram of adjusting the weight feedback loop.
Fig. 2 is the network system composition structural representation that the present invention ensures the polymorphic type QoS.
Fig. 3 is that each service agent of network system of the present invention constitutes the virtual topological structure schematic diagram.
Fig. 4 is the method operating procedure flow chart that network system of the present invention ensures the polymorphic type QoS,
Fig. 5 (A), (B) are respectively two schematic diagrames of local path information of the present invention;
Fig. 6 is that each service agent constitutes the virtual topological structure schematic diagram in the embodiment of the invention.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, the present invention is described in further detail below in conjunction with drawings and Examples.
Referring to Fig. 2, introduce the structure composition that the present invention ensures the network system of polymorphic type QoS, this system is provided with: send and receive the source end main frame and the end host of business data packet, and edge router and the core router of bearing transfer of data and forwarding; The business datum of being sent by service source end main frame finally arrives the service terminal main frame through the forwarding of a plurality of edge routers and core router.In order to ensure the service quality of polymorphic type business, wherein source end main frame and two kinds of routers are all transformed.
Source end main frame, set up professional sensing module and preliminary making module, the former is used for the business data packet that sends is carried out perception,, sends after according to perception information business data packet being labeled as different semantic priority in advance when the network layer encapsulation business data packet for the latter.
Edge router, set up and admit controller and a class or multiclass service agent, the semantic priority mapping that the former is used for the business data packet that will receive is a network priority, and with the allocation of packets of heterogeneous networks priority to the service agent of handling the respective priority business; Latter's kind depends on the number of degrees of the network priority that this system supports, every class service agent is responsible for handling a class business, and the same line of business in each edge router and the core router agency interconnect, thereby form virtual topological structure (referring to shown in Figure 3) separately, each virtual topological structure is provided with separately independently routing table, is carrying out route in the corresponding virtual topological structure separately respectively for the business data packet of heterogeneous networks priority.
Core router is responsible for transmitting business data packet, and it sets up a class or multiclass service agent, and its function is identical with edge router.
Referring to Fig. 3, the A among the figure, B, C are three routers, and wherein A is an edge router, and B, C are core router.Be convenient narration, suppose that each router has the professional three kinds of service agents of AF business, BE business and EF, all kinds of service agents are responsible for handling a class business datum respectively.Same line of business in each router agency has constituted a virtual topology, and each virtual topology is safeguarded self routing table, and such business is route on this virtual topology.For example, the BE-Agent in each router forms a virtual network, is responsible for treatments B E business.
The present invention is based on above-mentioned network system, also proposed a kind of method for routing that ensures the polymorphic type QoS: source end main frame is according to the semantic priority of application layer semantic differentiation business data packet, and carries out corresponding preliminary making when network layer encapsulates; When business data packet arrives the DiffServ network, the ingress edge router is network priority according to setting strategy with the semantic priority mapping of the preliminary making of each packet grouping, select corresponding different routing policy according to its network priority again: comprise traditional Best-Effort service packet is carried out forwarded hop-by-hop and the high-priority service packet calculating optimum route to there being QoS to ensure: edge router and core router are also collected the business data packet contextual information respectively, select corresponding business agency and routing policy to send it to end host according to the network priority and the current network resources of business data packet.
Referring to Fig. 4, introduce the following operating procedure of the inventive method:
Step 1, source end main frame utilization cross-layer method is distinguished its session priority to the business data packet that sends, and when network layer encapsulated, preliminary making was corresponding different semantic priority respectively.
This step comprises following two content of operation:
(11) source end host analysis business data packet, distinguish the importance and the type of each grouping in this business data packet, according to the type of each grouping, importance, to the contribution of QoS of survice and the dependence between each grouping, each grouping of packet is divided into the different semantic priority that a plurality of sequence of importance are successively decreased;
(12) when network layer encapsulation business data packet, source end main frame carries out preliminary making to the class of service and the semantic priority thereof of each grouping: adopt and keep for classification to select the mark value identification service classification of the per hop behavior PHB (Per Hop Behavior) of CS (Class Selector) in the IP packet header in Differentiated Services Code Point DSCP (the DifferentiatedServices Code Point) field, promptly use the front three identification service classification of DSCP; The semantic priority of each grouping of semantic priority index PI (Priority Index) value sign are adopted in three of backs, and the grade sequence number of sequence number that should semanteme priority index PI value and semantic priority is consistent.
Step 2, when business data packet arrived the DiffServ network edge, the ingress edge router was remapped to different network priorities according to semantic priority and its labelling strategies of preliminary making.
In this step, the ingress edge router of DiffServ network is reset for the semantic priority preliminary making value of each grouping to source end main frame according to mapping relations and its labelling strategies; Mapping relations and its labelling strategies are: each semantic priority all is mapped to a unique network priority, or other semantic priority of a plurality of adjacent level all is mapped as same network priority; And, the rank of the network priority of its mapping, be that the numerical value of network priority index PI is non-decreasing along with the rank of semantic priority increases progressively.
Step 3, the admittance controller of edge router is according to current Internet resources, the business data packet of heterogeneous networks priority is distributed in the virtual topology of the service agent formation of handling the respective priority business, and calculate the virtual link weights of every virtual link in each virtual topology by each router, be used for routing and calculate.
This step comprises following content of operation:
(31) the admittance controller of edge router carries out perception to the business data packet that receives, and collect the contextual information that comprises packet, contextual information comprises network priority, physical link bandwidth, virtual link available bandwidth, virtual link utilance and the local routing information of session.When the local path information definition was certain business data packet sequential delivery of setting priority through a plurality of node, wherein the local path information of certain node was the flow on each bar downstream virtual link after business datum wraps in this node, and unit is Mbps.For example, A, the B among Fig. 4 (A), C, D node, for Node B, its upstream node is A, and downstream node is C and D, and the local path information that B is ordered is professional s is arrived C and D through B by A flow size.The local path information bank of multibusiness network is that the three-dimensional array that is made of the starting point of each local path, terminal point and data packets for transmission class of service thereof is (referring to Fig. 4 (B), x axle among the figure and y axle are respectively the starting point and the terminal point of local path, the z axle is a class of service), wherein the local path information of each nodes records is used to calculate the virtual link utilance of the upstream virtual link and the downstream virtual link of this node.
(32) router calculates the virtual link utilance of each virtual link earlier, calculates the virtual link weights of each virtual link that is used for routing then according to the result.So this step comprises following content of operation:
(321) grouped data of certain class business wraps in route on the virtual topology of being responsible for such professional agent node.In each virtual topology, the present invention is quantitatively to calculate with its link utilization for the flow status of certain bar link.When considering the link utilization of certain bar link L, consider himself link utilization and it influence simultaneously to downstream node, analyze between the downstream that how flow is distributed in and they are all one hop node of this link on the link.Therefore, the utilance of the virtual link in the virtual topology network that each router calculates the virtual link utilance of each virtual link, promptly be made up of each service agent, its computational methods are as follows:
Earlier according to the following equation calculating priority level be k service traffics two routers (b, e) between available bandwidth on the virtual link
Be that its numerical value is that the total bandwidth of this physical link deducts on this link by priority and is higher than the poor of bandwidth that all business of k have taken: in the formula, natural number k is the priority sequence number, and wherein 1 is the BE stream of the lowest priority of no service quality guarantee; C (b, e) be two routers (b, e) between the total bandwidth of physical link, F
i(b, e) be two routers (b, e) between on the physical link priority be the professional shared bandwidth of i.
Because of the virtual link utilance is the ratio of professional actual bandwidth that has taken of the type and the professional available bandwidth of the type:
Therefore, for the minimum BE of priority stream, its available bandwidth be physical link actual bandwidth C (b, e), its virtual link utilance is the ratio of service traffics summation and this link actual physical bandwidth of all priority:
All one jump off trip virtual link and all and jump onto the influence that the trip virtual link produces to it in order to embody changes in flow rate on every virtual link, determine the virtual link utilance of every virtual link with following method:
When the grouped data of certain type service wraps on the virtual topology that the service agent of being in charge of the type business constitutes route, represent flow size on this virtual link with the virtual link utilance of every virtual link; And, to consider also how the flow on the virtual link of upstream is distributed in its downstream virtual link in order to find out flow on the virtual link of upstream to the influence of downstream virtual link; Calculating when the virtual link utilance of certain bar virtual link is set, with the flow distribution on this virtual link at it all one jump off the shared virtual link utilance of trip virtual link add up and
Current virtual link utilance u with this virtual link
LCompare, select wherein higher value as the virtual link utilance of this virtual link:
If
Then the virtual link utilance of this virtual link is
If
Then the virtual link utilance of this virtual link still is its currency u
L, in the formula, p
iJump shared virtual link utilance on the virtual link for the flow distribution on this virtual link in its each downstream one, i is the sequence number that virtual link is jumped in downstream one, and its maximum is n;
(322) router calculates the virtual link weights of virtual link: carry out in the virtual network topology of route in the business that to network priority is k, adopt following have linear approximation function formula that M/M/1 postpones calculate two routers (b, e) between the virtual link weights Φ of virtual link
k(b, e):
So that utilize the virtual link weights Φ that calculates
kBeing used for route calculates.
Step 4, all kinds of service agents of edge router and core router are taked corresponding manner to handle according to its virtual topological structure and are transmitted business data packet.Just for the business of setting priority, the weights that the edge router end obtains according to step (3) carry out route and calculate, after calculating one or more shortest path, when transmitting, on this one or more shortest path, press the scheduling mode of poll and select wherein paths transmission.
The present invention has repeatedly implemented test, and the result of test is successful, has realized goal of the invention.
The above only is preferred embodiment of the present invention, and is in order to restriction the present invention, within the spirit and principles in the present invention not all, any modification of being made, is equal to replacement, improvement etc., all should be included within the scope of protection of the invention.
Claims (8)
1. a network system that ensures the polymorphic type QoS comprises: the source end main frame and the end host of transmission and reception business data packet, and edge router and the core router of bearing transfer of data and forwarding; It is characterized in that: in the described system,
Source end main frame, set up professional sensing module and preliminary making module, the former is used for the business data packet that sends is carried out perception,, sends after according to perception information business data packet being labeled as different semantic priority in advance when the network layer encapsulation business data packet for the latter;
Edge router, set up and admit controller and a class or multiclass service agent, the semantic priority mapping that the former is used for the business data packet that will receive is a network priority, and with the allocation of packets of heterogeneous networks priority to the service agent of handling the respective priority business; Latter's kind depends on the number of degrees of the network priority that this system supports, every class service agent is responsible for handling a class priority service, and the same line of business in each edge router and core router agency interconnects and forms separately virtual topological structure, each virtual topological structure is provided with separately independently routing table, is carrying out route in the corresponding virtual topological structure separately respectively for the business data packet of heterogeneous networks priority; Transmit the core router of business data packet and only set up a class or multiclass service agent, its function is identical with edge router.
2. method that adopts the described network system of claim 1 to ensure the polymorphic type QoS is characterized in that: source end main frame is according to the semantic priority of application layer semantic differentiation business data packet, and carries out corresponding preliminary making when network layer encapsulates; When business data packet arrives Differentiated Services DiffServ network, the ingress edge router is network priority according to setting strategy with the semantic priority mapping of the preliminary making of each packet grouping, again according to its network priority select to comprise to traditional Best-Effort service packet carry out forwarded hop-by-hop with to the different routing policies of the high-priority service packet calculating optimum route that QoS ensures are arranged: edge router and core router are also collected the business data packet contextual information respectively, select corresponding business agency and routing policy to send it to end host according to the network priority and the current network resources of business data packet.
3. method according to claim 2 is characterized in that: described method comprises following operating procedure:
(1) source end main frame utilization cross-layer method is distinguished its session priority to the business data packet that sends, and when network layer encapsulated, preliminary making was corresponding different semantic priority respectively;
When (2) business data packet arrived the DiffServ network edge, the ingress edge router was remapped to different network priorities according to semantic priority and its labelling strategies of preliminary making;
(3) the admittance controller of edge router is according to current Internet resources, the business data packet of heterogeneous networks priority is distributed in the virtual topology of the service agent formation of handling the respective priority business, and calculate the virtual link weights of every virtual link in each virtual topology by each router, be used for routing and calculate;
(4) all kinds of service agents of edge router and core router are taked corresponding manner to handle according to its virtual topological structure and are transmitted business data packet.
4. method according to claim 3 is characterized in that: described step (1) further comprises following content of operation:
(11) source end host analysis business data packet, distinguish the importance and the type of each grouping in this business data packet, according to the type of each grouping, importance, to the contribution of QoS of survice and the dependence between each grouping, each grouping of packet is divided into the different semantic priority that a plurality of sequence of importance are successively decreased;
(12) when network layer encapsulation business data packet, source end main frame carries out preliminary making to the class of service and the semantic priority thereof of each grouping: adopt and keep for classification to select the mark value identification service classification of the per hop behavior PHB of CS in the IP packet header in the Differentiated Services Code Point dscp field, promptly use the front three identification service classification of DSCP; Three of backs adopt semantic priority index PI values to identify the semantic priority of each grouping, and the grade sequence number of sequence number that should semanteme priority index PI value and semantic priority is consistent.
5. network route method according to claim 3, it is characterized in that: in the described step (2), the ingress edge router of DiffServ network is reset to the semantic priority preliminary making value that source end main frame is done for each grouping according to mapping relations and its labelling strategies; Described mapping relations and its labelling strategies are: each semantic priority all is mapped to a unique network priority, or other semantic priority of a plurality of adjacent level all is mapped as same network priority; And, the rank of the network priority of its mapping, be that the numerical value of network priority index PI is non-decreasing along with the rank of semantic priority increases progressively.
6. method for routing according to claim 3 is characterized in that: described step (3) comprises following content of operation:
(31) the admittance controller of edge router carries out perception to the business data packet that receives, and collection comprises the contextual information of network priority, physical link bandwidth, virtual link available bandwidth, virtual link utilance and the local routing information of packet, when described local path information is certain business data packet sequential delivery of setting priority through a plurality of node, wherein the local path information of certain node is the flow on each bar downstream virtual link after business datum wraps in this node, and unit is Mbps; The local path information bank of multibusiness network is the three-dimensional array that is made of the starting point of each local path, terminal point and data packets for transmission class of service thereof, and wherein the local path information of each nodes records is used to calculate the virtual link utilance of the upstream virtual link and the downstream virtual link of this node;
(32) router calculates the virtual link utilance of each virtual link earlier, calculates the virtual link weights of each virtual link that is used for routing then according to the result.
7. method for routing according to claim 5 is characterized in that: described step (32) comprises following content of operation:
(321) each router calculate the virtual link utilance of each virtual link, the utilance of the virtual link in the virtual topology network promptly formed by each service agent, its computational methods are as follows:
Earlier according to the following equation calculating priority level be k service traffics two routers (b, e) between available bandwidth on the virtual link
Be that its numerical value is that the total bandwidth of this physical link deducts on this link by priority and is higher than the poor of bandwidth that all business of k have taken: in the formula, natural number k is the priority sequence number, and wherein 1 is the BE stream of the lowest priority of no service quality guarantee; C (b, e) be two routers (b, e) between the total bandwidth of physical link, F
i(b, e) be two routers (b, e) between on the physical link priority be the professional shared bandwidth of i;
Because of the virtual link utilance is the ratio of professional actual bandwidth that has taken of the type and the professional available bandwidth of the type:
Therefore, for the minimum BE of priority stream, its available bandwidth be physical link actual bandwidth C (b, e), its virtual link utilance is the ratio of service traffics summation and this link actual physical bandwidth of all priority:
All one jump off trip virtual link and all and jump onto the influence that the trip virtual link produces to it in order to embody changes in flow rate on every virtual link, determine the virtual link utilance of every virtual link with following method:
When the grouped data of certain type service wraps on the virtual topology that the service agent of being in charge of the type business constitutes route, represent flow size on this virtual link with the virtual link utilance of every virtual link; And, to consider also how the flow on the virtual link of upstream is distributed in its downstream virtual link in order to find out flow on the virtual link of upstream to the influence of downstream virtual link; Calculating when the virtual link utilance of certain bar virtual link is set, with the flow distribution on this virtual link at it all one jump off the shared virtual link utilance of trip virtual link add up and
Current virtual link utilance u with this virtual link
LCompare, select wherein higher value as the virtual link utilance of this virtual link:
If
Then the virtual link utilance of this virtual link is
If
Then the virtual link utilance of this virtual link still is its currency u
L, in the formula, p
iJump shared virtual link utilance on the virtual link for the flow distribution on this virtual link in its each downstream one, i is the sequence number that virtual link is jumped in downstream one, and its maximum is n;
(322) router calculates the virtual link weights of virtual link: carry out in the virtual network topology of route in the business that to network priority is k, adopt following have linear approximation function formula that M/M/1 postpones calculate two routers (b, e) between the virtual link weights Φ of virtual link
k(b, e):
So that utilize the virtual link weights Φ that calculates
kBeing used for route calculates.
8. method for routing according to claim 3, it is characterized in that: described step (4) comprises following content of operation: for the business of setting priority, the weights that the edge router end obtains according to step (3) carry out route and calculate, after calculating one or more shortest path, when transmitting, on this one or more shortest path, press the scheduling mode of poll and select wherein paths transmission.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110047722.9A CN102075444B (en) | 2011-02-28 | 2011-02-28 | Network system and method for guaranteeing multi-type service quality |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110047722.9A CN102075444B (en) | 2011-02-28 | 2011-02-28 | Network system and method for guaranteeing multi-type service quality |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102075444A true CN102075444A (en) | 2011-05-25 |
| CN102075444B CN102075444B (en) | 2013-10-02 |
Family
ID=44033796
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110047722.9A Expired - Fee Related CN102075444B (en) | 2011-02-28 | 2011-02-28 | Network system and method for guaranteeing multi-type service quality |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102075444B (en) |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102571583A (en) * | 2012-01-11 | 2012-07-11 | 浙江工业大学 | Quality of service (QoS) aware self-adaptive bandwidth distribution system for wireless-optical broadband access networks (WOBAN) and self-adaptive bandwidth distribution method |
| CN103067982A (en) * | 2012-12-14 | 2013-04-24 | 华为技术有限公司 | Method and equipment of data transmission |
| CN103188149A (en) * | 2011-12-28 | 2013-07-03 | 中国移动通信集团广东有限公司 | Routing computation method and router |
| CN103685054A (en) * | 2013-12-18 | 2014-03-26 | 武汉烽火网络有限责任公司 | Multipath load balancing method based on service awareness |
| WO2014205660A1 (en) * | 2013-06-26 | 2014-12-31 | 华为技术有限公司 | Method, apparatus and routing device for forwarding data packet |
| CN104753751A (en) * | 2013-12-27 | 2015-07-01 | 中兴通讯股份有限公司 | Method and system for dynamically determining virtual network |
| CN105072050A (en) * | 2015-08-26 | 2015-11-18 | 联想(北京)有限公司 | Data transmission method and data transmission device |
| CN105099927A (en) * | 2015-08-31 | 2015-11-25 | 上海斐讯数据通信技术有限公司 | Method for realizing multi-service channel and router |
| CN105827471A (en) * | 2015-01-04 | 2016-08-03 | 中国移动通信集团江苏有限公司 | Port uplink and downlink traffic array monitoring methods, devices and systems |
| CN105959218A (en) * | 2016-05-09 | 2016-09-21 | 中国联合网络通信集团有限公司 | Network route selection method and network route selection system |
| CN106453121A (en) * | 2016-09-14 | 2017-02-22 | 上海烟草集团有限责任公司 | Dynamic link load configuration method and system, and server |
| CN106936645A (en) * | 2017-04-19 | 2017-07-07 | 西安电子科技大学 | The optimization method of the tree network topology structure based on queueing theory |
| CN107241272A (en) * | 2016-03-29 | 2017-10-10 | 瞻博网络公司 | Mthods, systems and devices for improving transfer capability during route convergence |
| CN107896192A (en) * | 2017-11-20 | 2018-04-10 | 电子科技大学 | The QoS control method of differentiated service priority in a kind of SDN |
| CN108124002A (en) * | 2017-12-01 | 2018-06-05 | 北京明朝万达科技股份有限公司 | A kind of data transmission method for uplink and device |
| CN108463805A (en) * | 2016-01-14 | 2018-08-28 | 索尼移动通讯有限公司 | User equipment for moving edge calculations selects |
| CN108718236A (en) * | 2018-05-21 | 2018-10-30 | 西安交通大学 | The data transmission method of internet self-organizing confluence |
| CN109088822A (en) * | 2018-09-21 | 2018-12-25 | 平安科技（深圳）有限公司 | Data traffic retransmission method, device, system, computer equipment and storage medium |
| WO2020001308A1 (en) * | 2018-06-29 | 2020-01-02 | 华为技术有限公司 | Network optimization method and system, and network device |
| CN111711961A (en) * | 2020-04-30 | 2020-09-25 | 南京邮电大学 | Service end-to-end performance analysis method introducing random probability parameter |
| CN112152935A (en) * | 2020-09-01 | 2020-12-29 | 中国联合网络通信集团有限公司 | Method and device for determining transmission path |
| CN113067627A (en) * | 2021-03-18 | 2021-07-02 | 西安电子科技大学 | Self-adaptive survivable satellite routing method based on virtual nodes |
| WO2022057638A1 (en) * | 2020-09-18 | 2022-03-24 | 华为技术有限公司 | Method and system for guaranteeing service application experience |
| CN114401227A (en) * | 2021-12-22 | 2022-04-26 | 网络通信与安全紫金山实验室 | Data forwarding method and device |
| CN114448816A (en) * | 2021-12-30 | 2022-05-06 | 中国航空研究院 | Integrated IP networking method based on heterogeneous data chain |
| CN114567907A (en) * | 2022-03-09 | 2022-05-31 | 广东电网有限责任公司 | Resource management method, device and system for cross-domain network |
| CN115051950A (en) * | 2022-06-23 | 2022-09-13 | 南京邮电大学 | Adaptive data forwarding system and method for different service types |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108632163B (en) * | 2017-03-24 | 2021-11-16 | 中国联合网络通信集团有限公司 | Data packet transmission method, boundary router and core router |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1731757A (en) * | 2004-08-05 | 2006-02-08 | 华为技术有限公司 | Method and system for providing QoS assurance in edge router |
| US7215637B1 (en) * | 2000-04-17 | 2007-05-08 | Juniper Networks, Inc. | Systems and methods for processing packets |
| CN101552726A (en) * | 2009-05-14 | 2009-10-07 | 北京交通大学 | A grading services edge router |
| CN101692657A (en) * | 2009-10-22 | 2010-04-07 | 北京交通大学 | Differentiated service core router and data forwarding method thereof |
| CN101860475A (en) * | 2010-04-02 | 2010-10-13 | 北京邮电大学 | Autonomous queue management method based on context awareness |
-
2011
- 2011-02-28 CN CN201110047722.9A patent/CN102075444B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7215637B1 (en) * | 2000-04-17 | 2007-05-08 | Juniper Networks, Inc. | Systems and methods for processing packets |
| CN1731757A (en) * | 2004-08-05 | 2006-02-08 | 华为技术有限公司 | Method and system for providing QoS assurance in edge router |
| CN101552726A (en) * | 2009-05-14 | 2009-10-07 | 北京交通大学 | A grading services edge router |
| CN101692657A (en) * | 2009-10-22 | 2010-04-07 | 北京交通大学 | Differentiated service core router and data forwarding method thereof |
| CN101860475A (en) * | 2010-04-02 | 2010-10-13 | 北京邮电大学 | Autonomous queue management method based on context awareness |
Non-Patent Citations (2)
| Title |
|---|
| 《Advanced Intelligence and Awareness Internet(AIAI 2010)》 20101025 Mingqiao等 A Routing Mechanism to Support QoS of Multi-Class Traffic , * |
| 《International Conference On Emerging Networking Experiments And Technologies(CoNEXT"07)》 20071213 Kin-Wah Kwong等 Improving Service Differentiation in IP Networks through Dual Topology Routing , * |
Cited By (43)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103188149A (en) * | 2011-12-28 | 2013-07-03 | 中国移动通信集团广东有限公司 | Routing computation method and router |
| CN103188149B (en) * | 2011-12-28 | 2016-05-25 | 中国移动通信集团广东有限公司 | A kind of route computing method and router |
| CN102571583A (en) * | 2012-01-11 | 2012-07-11 | 浙江工业大学 | Quality of service (QoS) aware self-adaptive bandwidth distribution system for wireless-optical broadband access networks (WOBAN) and self-adaptive bandwidth distribution method |
| CN103067982B (en) * | 2012-12-14 | 2015-09-09 | 华为技术有限公司 | A kind of data transmission method and equipment |
| CN103067982A (en) * | 2012-12-14 | 2013-04-24 | 华为技术有限公司 | Method and equipment of data transmission |
| WO2014205660A1 (en) * | 2013-06-26 | 2014-12-31 | 华为技术有限公司 | Method, apparatus and routing device for forwarding data packet |
| CN103685054B (en) * | 2013-12-18 | 2017-02-01 | 武汉烽火网络有限责任公司 | Multipath load balancing method based on service awareness |
| CN103685054A (en) * | 2013-12-18 | 2014-03-26 | 武汉烽火网络有限责任公司 | Multipath load balancing method based on service awareness |
| CN104753751A (en) * | 2013-12-27 | 2015-07-01 | 中兴通讯股份有限公司 | Method and system for dynamically determining virtual network |
| CN105827471A (en) * | 2015-01-04 | 2016-08-03 | 中国移动通信集团江苏有限公司 | Port uplink and downlink traffic array monitoring methods, devices and systems |
| CN105827471B (en) * | 2015-01-04 | 2019-04-19 | 中国移动通信集团江苏有限公司 | Port uplink and downlink flow array monitoring method, apparatus and system |
| CN105072050A (en) * | 2015-08-26 | 2015-11-18 | 联想(北京)有限公司 | Data transmission method and data transmission device |
| CN105099927B (en) * | 2015-08-31 | 2018-04-06 | 上海斐讯数据通信技术有限公司 | A kind of method and router for realizing multi-service passage |
| CN105099927A (en) * | 2015-08-31 | 2015-11-25 | 上海斐讯数据通信技术有限公司 | Method for realizing multi-service channel and router |
| CN108463805B (en) * | 2016-01-14 | 2022-05-31 | 索尼移动通讯有限公司 | User equipment selection for mobile edge computing |
| CN108463805A (en) * | 2016-01-14 | 2018-08-28 | 索尼移动通讯有限公司 | User equipment for moving edge calculations selects |
| CN107241272A (en) * | 2016-03-29 | 2017-10-10 | 瞻博网络公司 | Mthods, systems and devices for improving transfer capability during route convergence |
| CN107241272B (en) * | 2016-03-29 | 2020-04-21 | 瞻博网络公司 | Method, system and apparatus for improving forwarding capability during route convergence |
| CN105959218A (en) * | 2016-05-09 | 2016-09-21 | 中国联合网络通信集团有限公司 | Network route selection method and network route selection system |
| CN106453121A (en) * | 2016-09-14 | 2017-02-22 | 上海烟草集团有限责任公司 | Dynamic link load configuration method and system, and server |
| CN106453121B (en) * | 2016-09-14 | 2019-05-21 | 上海烟草集团有限责任公司 | A kind of link dynamic load configuration method, system and server |
| CN106936645A (en) * | 2017-04-19 | 2017-07-07 | 西安电子科技大学 | The optimization method of the tree network topology structure based on queueing theory |
| CN106936645B (en) * | 2017-04-19 | 2019-10-11 | 西安电子科技大学 | The optimization method of tree network topology structure based on queueing theory |
| CN107896192A (en) * | 2017-11-20 | 2018-04-10 | 电子科技大学 | The QoS control method of differentiated service priority in a kind of SDN |
| CN107896192B (en) * | 2017-11-20 | 2020-09-25 | 电子科技大学 | QoS control method for differentiating service priority in SDN network |
| CN108124002A (en) * | 2017-12-01 | 2018-06-05 | 北京明朝万达科技股份有限公司 | A kind of data transmission method for uplink and device |
| CN108718236A (en) * | 2018-05-21 | 2018-10-30 | 西安交通大学 | The data transmission method of internet self-organizing confluence |
| WO2020001308A1 (en) * | 2018-06-29 | 2020-01-02 | 华为技术有限公司 | Network optimization method and system, and network device |
| US11546215B2 (en) | 2018-06-29 | 2023-01-03 | Huawei Technologies Co., Ltd. | Method, system, and device for data flow metric adjustment based on communication link state |
| CN109088822A (en) * | 2018-09-21 | 2018-12-25 | 平安科技（深圳）有限公司 | Data traffic retransmission method, device, system, computer equipment and storage medium |
| CN111711961A (en) * | 2020-04-30 | 2020-09-25 | 南京邮电大学 | Service end-to-end performance analysis method introducing random probability parameter |
| CN112152935A (en) * | 2020-09-01 | 2020-12-29 | 中国联合网络通信集团有限公司 | Method and device for determining transmission path |
| CN112152935B (en) * | 2020-09-01 | 2022-05-17 | 中国联合网络通信集团有限公司 | Method and device for determining transmission path |
| WO2022057638A1 (en) * | 2020-09-18 | 2022-03-24 | 华为技术有限公司 | Method and system for guaranteeing service application experience |
| CN113067627A (en) * | 2021-03-18 | 2021-07-02 | 西安电子科技大学 | Self-adaptive survivable satellite routing method based on virtual nodes |
| CN114401227A (en) * | 2021-12-22 | 2022-04-26 | 网络通信与安全紫金山实验室 | Data forwarding method and device |
| CN114401227B (en) * | 2021-12-22 | 2024-03-26 | 网络通信与安全紫金山实验室 | Data forwarding method and device |
| CN114448816B (en) * | 2021-12-30 | 2023-10-10 | 中国航空研究院 | Integrated IP networking method based on heterogeneous data chain |
| CN114448816A (en) * | 2021-12-30 | 2022-05-06 | 中国航空研究院 | Integrated IP networking method based on heterogeneous data chain |
| CN114567907A (en) * | 2022-03-09 | 2022-05-31 | 广东电网有限责任公司 | Resource management method, device and system for cross-domain network |
| CN114567907B (en) * | 2022-03-09 | 2024-01-30 | 广东电网有限责任公司 | Resource management method, device and system of cross-domain network |
| CN115051950B (en) * | 2022-06-23 | 2023-08-18 | 南京邮电大学 | Self-adaptive data forwarding system and method for different service types |
| CN115051950A (en) * | 2022-06-23 | 2022-09-13 | 南京邮电大学 | Adaptive data forwarding system and method for different service types |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102075444B (en) | 2013-10-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102075444B (en) | Network system and method for guaranteeing multi-type service quality | |
| Yaghmaee et al. | Priority-based rate control for service differentiation and congestion control in wireless multimedia sensor networks | |
| CN106027416B (en) | A kind of data center network traffic scheduling method and system based on space-time combination | |
| US8477627B2 (en) | Content routing in digital communications networks | |
| CN103716208B (en) | Support network management, system, interchanger and the network of elephant stream | |
| US7917648B2 (en) | Self-adaptive scheduling method and network element | |
| CN106533960A (en) | Data center network routing method based on Fat-Tree structure | |
| CN101340374B (en) | Method, system, apparatus for control transmission priority and user network edge equipment | |
| CN104272680A (en) | Signalling congestion | |
| EP2263354B1 (en) | Admission control in a packet network | |
| CN107948067B (en) | Link load balancing method for QoS guarantee of multiple service flows in software defined network | |
| CN103222248A (en) | Multi-class data transport | |
| CN108400936A (en) | Information Network method for routing based on MPLS | |
| Dolama et al. | Modified smallest available report first: new dynamic bandwidth allocation schemes in QoS-capable EPONs | |
| CN110266606A (en) | Active queue management optimization method and device in a kind of edge network | |
| Kato et al. | A congestion control method for named data networking with hop-by-hop window-based approach | |
| Hertiana et al. | A joint approach to multipath routing and rate adaptation for congestion control in openflow software defined network | |
| Kim et al. | Providing absolute differentiated services for optical burst switching networks: loss differentiation | |
| US8126004B2 (en) | Method for optimising the sharing of a plurality of network resources between a plurality of application flows | |
| Shi | Token-based congestion control: Achieving fair resource allocations in P2P networks | |
| JP5613261B2 (en) | Method and allocation unit for allocating communication pipes in a communication network | |
| Lee et al. | Traffic engineering with constrained multipath routing in MPLS networks | |
| Domżał | Flow-aware networking as an architecture for the IPv6 QoS Parallel Internet | |
| Bai et al. | Dynamic end-to-end QoS support for video over the Internet | |
| Jiang et al. | Name Weighted Round Robin (NWRR) Algorithm for Named Data Networking |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20131002 Termination date: 20150228 |
|
| EXPY | Termination of patent right or utility model |