CN107438026A - The failure recovery method and apparatus of inter-domain routing system - Google Patents
The failure recovery method and apparatus of inter-domain routing system Download PDFInfo
- Publication number
- CN107438026A CN107438026A CN201610363674.7A CN201610363674A CN107438026A CN 107438026 A CN107438026 A CN 107438026A CN 201610363674 A CN201610363674 A CN 201610363674A CN 107438026 A CN107438026 A CN 107438026A
- Authority
- CN
- China
- Prior art keywords
- node
- inter
- adjacent
- routing system
- domain routing
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/28—Routing or path finding of packets in data switching networks using route fault recovery
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/02—Topology update or discovery
- H04L45/04—Interdomain routing, e.g. hierarchical routing
Abstract
The invention discloses a kind of failure recovery method and apparatus of inter-domain routing system.Methods described includes：According to default cascading failure model and default Rule of judgment, in inter-domain routing system, the overall connective key node of statistics influence system；In inter-domain routing system, the node in addition to key node and leaf node is chosen, as node to be selected；Count the adjacent node of node to be selected, and according to corresponding to preset rules calculate node to be selected all adjacent sides of adjacent node weighted value；Count the remaining adjacent node of non-key node in all of its neighbor node of node to be selected；Retain in remaining adjacent node, the maximum adjacent side of weighted value in adjacent side corresponding to each node, and delete other adjacent sides of remaining adjacent node, generate corresponding backup subgraph.As long as the bottom link of the method inter-domain routing system of the present invention is not turned off, the data forwarding function of fast quick-recovery key node, the fast quick-recovery that inter-domain routing system fails can be realized by backing up subgraph.
Description
Technical field
The present invention relates to inter-domain routing system security technology area, more particularly to a kind of failure recovery of inter-domain routing system
Method and apparatus.
Background technology
Inter-domain routing system based on Border Gateway Protocol (Border Gateway Protocol, referred to as " BGP ") is mutual
The critical infrastructures of networking, although with the fast development of internet, inter-domain routing system scale, which has, largely to expand
, but the degree of its node is distributed still great heterogeneity and defers to power-law distribution, and whole inter-domain routing system has scaleless property
Matter.Compared to random network, scales-free network has stronger robustness, and (or resistance is transsexual, English：Robustness), but
When facing targeted attack, scales-free network is then very fragile, and only a small number of " core nodes " are attacked, the service of overall network
Ability will drastically decline, or even a wide range of paralysis occur.
Survivability is theoretical to be established on the basis such as conventional security theory, trust computing, and its basic assumption is appointing for system
What part may break down or be invaded, and system of paying close attention to remains to continue to provide pass after under attack, failure, destruction
The ability of key service.Survivability theory is while network security intension is further enriched, there is provided exploring from different perspectives should
To the resolving ideas of complex network attack.So carry out the theoretical research of inter-domain routing system survivability, especially between exploration domain
The correlation technique and method of route system failure recovery, it is significant.
The content of the invention
In order to solve to need a kind of a small number of key nodes in inter-domain routing system to cause system by attack in the prior art
During paralysis, can quick recovery system operation method the problem of, the embodiments of the invention provide a kind of failure of inter-domain routing system
Restoration methods and device.The technical scheme is as follows：
On the one hand, the embodiments of the invention provide a kind of failure recovery method of inter-domain routing system, methods described to include：
According to default cascading failure model and default Rule of judgment, in the inter-domain routing system, statistics influences
The overall connective key node of system；
In the inter-domain routing system, the node in addition to the key node and leaf node is chosen, as section to be selected
Point；
Count the adjacent node of the node to be selected, and the adjacent section according to corresponding to preset rules calculate the node to be selected
The weighted value of all adjacent sides of point；
Count the remaining adjacent node of non-key node in all of its neighbor node of the node to be selected；
Retain in the remaining adjacent node, the maximum adjacent side of weighted value in adjacent side corresponding to each node, and delete institute
Other adjacent sides of remaining adjacent node to be stated, generate corresponding backup subgraph, the backup subgraph is used for when key node fails,
Data forwarding paths are provided for the key node of failure.
In the failure recovery method of the above-mentioned inter-domain routing system of the embodiment of the present invention, the default cascading failure mould
Type, including：
When in the inter-domain routing system, after node i failure, the adjacent node j of node i is loaded by equation below
Reassign：
lj(t+1)=(lj(t)+Δlj i)；
Wherein,
ΓiRepresent the set of node i all of its neighbor node；Node k is any of set of node i all of its neighbor node
Node；li(0)、lj(0)、lk(0) the original load of node i, node j, node k is represented respectively；D represents the degree of node, the i.e. section
The quantity of point all of its neighbor node；α, β are initial load parameter, and are all higher than zero；Δlj iNode is distributed to after being failed for node i
J newly-increased load；lj(t) at the time of before being reassigned for load during t, the load on node j；lj(t+1) it is to redistribute load
Afterwards, loads of the node j when the moment, t+1 was restarted.
In the failure recovery method of the above-mentioned inter-domain routing system of the embodiment of the present invention, the default Rule of judgment,
Including：
Choosing influences the node of preset ratio before ranking on the overall connectedness of the inter-domain routing system, is saved as key
Point.
It is described to be calculated according to preset rules in the failure recovery method of the above-mentioned inter-domain routing system of the embodiment of the present invention
The weighted value of all adjacent sides of adjacent node corresponding to the node to be selected, including：
Range intervals according to belonging to node by transmitting data flow amount in the adjacent side unit interval, assign adjacent side and weigh accordingly
Weight values.
In the failure recovery method of the above-mentioned inter-domain routing system of the embodiment of the present invention, methods described also includes：
Judge that whether all leaf node set are with owning in the inter-domain routing system in multiple backup subgraphs of generation
The set of node is identical,
If it is not the same, then continuing to choose node to be selected, and generate corresponding backup subgraph；
If identical, stop generation backup subgraph.
On the other hand, the embodiments of the invention provide a kind of failure recovery device of inter-domain routing system, described device bag
Include：
Statistical module, for according to default cascading failure model and default Rule of judgment, in the inter-domain routing system
In system, the overall connective key node of statistics influence system；
Module is chosen, in the inter-domain routing system, choosing the section in addition to the key node and leaf node
Point, as node to be selected；
Processing module, the section to be selected is calculated for counting the adjacent node of the node to be selected, and according to preset rules
The weighted value of all adjacent sides of adjacent node corresponding to point；
Statistical module, it is additionally operable to count the adjacent section of residue of non-key node in all of its neighbor node of the node to be selected
Point；
Generation module, for retaining in the remaining adjacent node, weighted value maximum in adjacent side corresponding to each node
Adjacent side, and other adjacent sides of the remaining adjacent node are deleted, corresponding backup subgraph is generated, the backup subgraph is used to close
During key node failure, data forwarding paths are provided for the key node of failure.
In the failure recovery device of the above-mentioned inter-domain routing system of the embodiment of the present invention, the default cascading failure mould
Type, including：
When in the inter-domain routing system, after node i failure, the adjacent node j of node i is loaded by equation below
Reassign：
lj(t+1)=(lj(t)+Δlj i)；
Wherein,
ΓiRepresent the set of node i all of its neighbor node；Node k is any of set of node i all of its neighbor node
Node；li(0)、lj(0)、lk(0) the original load of node i, node j, node k is represented respectively；D represents the degree of node, the i.e. section
The quantity of point all of its neighbor node；α, β are initial load parameter, and are all higher than zero；Δlj iNode is distributed to after being failed for node i
J newly-increased load；lj(t) at the time of before being reassigned for load during t, the load on node j；lj(t+1) it is to redistribute load
Afterwards, loads of the node j when the moment, t+1 was restarted.
In the failure recovery device of the above-mentioned inter-domain routing system of the embodiment of the present invention, the default Rule of judgment,
Including：
Choosing influences the node of preset ratio before ranking on the overall connectedness of the inter-domain routing system, is saved as key
Point.
In the failure recovery device of the above-mentioned inter-domain routing system of the embodiment of the present invention, the processing module, it is additionally operable to
Range intervals according to belonging to node by transmitting data flow amount in the adjacent side unit interval, assign the corresponding weighted value of adjacent side.
In the failure recovery device of the above-mentioned inter-domain routing system of the embodiment of the present invention, described device also includes：
Judge module, for judge generation multiple backup subgraphs in all leaf node set whether with the Yu Jian roads
Set by all nodes in system is identical,
The processing module, it is additionally operable to, when the judged result of the judge module is differs, continue to choose section to be selected
Point, and generate corresponding backup subgraph；When the judged result of the judge module is identical, stop generation backup subgraph.
The beneficial effect that technical scheme provided in an embodiment of the present invention is brought is：
By the way that according to default cascading failure model and default Rule of judgment, in inter-domain routing system, statistics influences
The overall connective key node of system；In inter-domain routing system, the node in addition to key node and leaf node is chosen, is made
For node to be selected；Count the adjacent node of node to be selected, and the adjacent node according to corresponding to preset rules calculate node to be selected
The weighted value of all adjacent sides；Count the remaining adjacent node of non-key node in all of its neighbor node of node to be selected；Retain surplus
In remaining adjacent node, the maximum adjacent side of weighted value in adjacent side corresponding to each node, and delete other neighbours of remaining adjacent node
Side, corresponding backup subgraph is generated, can be by backing up son as long as the bottom link of so inter-domain routing system is not turned off
Figure, the data forwarding function of fast quick-recovery key node, realize the fast quick-recovery that inter-domain routing system fails.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, make required in being described below to embodiment
Accompanying drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for
For those of ordinary skill in the art, on the premise of not paying creative work, other can also be obtained according to these accompanying drawings
Accompanying drawing.
Fig. 1 is a kind of failure recovery method flow diagram for inter-domain routing system that the embodiment of the present invention one provides；
Fig. 2 is a kind of backup subgraph generating process exemplary plot that the embodiment of the present invention one provides；
Fig. 3 is a kind of failure recovery method flow diagram for inter-domain routing system that the embodiment of the present invention one provides；
Fig. 4 is a kind of failure recovery apparatus structure schematic diagram for inter-domain routing system that the embodiment of the present invention two provides；
Fig. 5 is a kind of failure recovery apparatus structure schematic diagram for inter-domain routing system that the embodiment of the present invention two provides.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing to embodiment party of the present invention
Formula is described in further detail.
Embodiment one
The embodiments of the invention provide a kind of failure recovery method of inter-domain routing system, referring to Fig. 1, methods described bag
Include：
Step S11, according to default cascading failure model and default Rule of judgment, in inter-domain routing system, statistics
The overall connective key node of influence system.
It should be noted that in actual applications, when is there is large area paralysis by network attack in inter-domain routing system,
Caused by being failed typically mainly due to key node by attack, in order to be carried out fast and effectively to the inter-domain routing system of failure
Recover, it is necessary to succinct effective link path be provided for the key node of these failures, so that the key node of failure recovers number
According to forwarding.In addition, in inter-domain routing system, single independent autonomy field system (Autonomous System, referred to as " AS ")
Corresponding to a node.
In the present embodiment, inter-domain routing system is considered as the link structure by being connected between multiple load nodes and node
Into its operation relies on Border Gateway Protocol, i.e. bgp protocol.In inter-domain routing system, if a node because of malfunction and failure,
Then the load on the node can be reallocated on other nodes for giving the node adjacency, if having in these adjacent nodes because
Redistribute load and overload, and then trigger failure, then the load of these nodes can also be reassigned to the adjoining of these nodes
Node, the like, cascading failure model reflects this load reassignment process.
Specifically, in above-mentioned steps S11, default cascading failure model includes：
When again in inter-domain routing system, after node i failure, the adjacent node j of node i is loaded by equation below to be divided
Match somebody with somebody：
lj(t+1)=(lj(t)+Δlj i)；
Wherein,
ΓiRepresent the set of node i all of its neighbor node；Node k is any of set of node i all of its neighbor node
Node；li(0)、lj(0)、lk(0) the original load of node i, node j, node k is represented respectively；D represents the degree of node, the i.e. section
The quantity of point all of its neighbor node；α, β are initial load parameter, and be all higher than zero (such as：1.2,6)；Δlj iFailed for node i
Node j newly-increased load is distributed to afterwards；lj(t) at the time of before being reassigned for load during t, the load on node j；lj(t+1) attach most importance to
After new distribution load, loads of the node j when the moment, t+1 was restarted.
In the present embodiment, above-mentioned formula interior joint i, node j can be any to adjacent node in inter-domain routing system,
After node i failure, the original load in node i can be assigned on the adjacent node of node i according to above-mentioned formula.
In addition, in order to coordinate above-mentioned cascading failure model, when meeting equation below, decision node m is failure node：
Wherein,
lm(0) node m original load is represented；Loaded for node m tolerable；γ is tolerance coefficient, and is more than or waits
In zero (such as：3)；lm(t+1) after to redistribute load, loads of the node j when the moment, t+1 was restarted；fΓmFor node to be measured
In failing and causing the node of failure, the set with node m adjacent nodes；Δlm nDistribute to node m's after being failed for node n
Newly-increased load, can be calculated according to the formula in step S11, repeat no more here.
In the present embodiment, node m is any node in inter-domain routing system, after node failure to be measured, inter-domain routing
The load on other nodes in system can be redistributed, if after redistributing, the load on the node can beyond it
Tolerance load, then the node can fail.
Further, in above-mentioned steps S11, default Rule of judgment can be：
Choose it is overall connective on inter-domain routing system influence preset ratio before ranking (such as：Preceding node 15%), make
For key node.
In the present embodiment, in above-mentioned cascading failure model, the failure effect degree of egress can also be calculated, to quantify
Node connective influence overall on inter-domain routing system, specifically, by equation below come the failure effect degree of calculate node：
Wherein, Ω is the failure effect degree of node；A is the different types of sum of failed link；X represents failed link
Type；NXRepresent the failed link sum that type is X；RXThe weight for the failed link that type is X is represented, if the link is used for
By the information transfer on upper level node to low-level nodes, then the weight of the link is 1.2；If the link is used on node
Node on information transfer to same level, then the weight of the link is 1；If the link is used for the information in bottom layer node
Transmit to upper level node, then the weight of the link is 0.8；| E | it is the sum of inter-domain routing system link.
After the failure effect degree of egress is calculated, it can be sorted from high to low according to failure effect degree, failure effect degree
Ranking is higher, and its connective influence overall on inter-domain routing system is bigger, in actual applications, can be according to the failure shadow of node
The node of preset ratio before ranking is chosen in the sequence of loudness.
Step S12, in inter-domain routing system, the node in addition to key node and leaf node is chosen, as section to be selected
Point.
In the present embodiment, leaf node is the node being only connected with another node, i.e., is 0 in discrete mathematics moderate
Node.
Step S13, all of its neighbor node of node to be selected is counted, and it is adjacent according to corresponding to preset rules calculate node to be selected
The weighted value of all adjacent sides of node is connect, the weighted value reflects significance level of the adjacent side in terms of transmitting data stream amount.
Specifically, can be realized in the following way in above-mentioned steps S13：
Range intervals according to belonging to node by transmitting data flow amount in the adjacent side unit interval, assign adjacent side and weigh accordingly
Weight values.
In the present embodiment, preset rules can be the different range section belonging to the data traffic transmitted on adjacent side, right
Should in corresponding weighted value, such as：When the data traffic transmitted on adjacent side is less than or equal to 500M, its weighted value is 1；It is more than
500M and when being less than or equal to 1G, its weighted value is 2；When more than 1G and being less than or equal to 10G, its weighted value is 3 etc., above number
According to being only for example, specific preset rules can be set according to actual conditions, be not limited here.
Step S14, count the remaining adjacent node of non-key node in all of its neighbor node of node to be selected.
In the present embodiment, all of its neighbor relation of inter-domain routing system, can be obtained in several ways, first, be
It is about to respective syntople certainly by each autonomy field system (i.e. each node) to be uploaded in the server of inter-domain routing system；
Meanwhile the server of inter-domain routing system can also be extracted by network from the public route data such as Route Views, RIPE
Adjacent set.Wherein, can be by verifying the neighbour when server is received from autonomy field system voluntarily above-mentioned syntople
Whether the autonomy field system that the relation of connecing is related to, which has, uploads the syntople, if not all uploaded, gives up adjoining pass
System；Identical syntople is uploaded if had, stores the syntople.
Step S15, retain in remaining adjacent node, the maximum adjacent side of weighted value in adjacent side corresponding to each node, and delete
Except other adjacent sides of remaining adjacent node, corresponding backup subgraph is generated, the backup subgraph is used for when key node fails, and is
The key node of failure provides data forwarding paths.
It should be noted that having a variety of link connections in inter-domain routing system between multiple nodes, i.e., key node can
Data are forwarded in a manner of there are a variety of link connections.When the attack of key node hand is failed, it is only necessary to for its quick offer one
Effective data transfer path, you can recover the forwarding datamation of failure node, and then fast quick-recovery inter-domain routing system
The purpose of system work, and key node one effective data forwarding paths of offer that subgraph is just used to failure are provided, with
For the fast quick-recovery of key node of failure.
In the present embodiment, if the key node of inter-domain routing system is failed by attack, then as long as Yu Jian roads
It is not turned off by the bottom link (physical connection i.e. between node) of system, may be referred to backup subgraph, the crucial section of fast quick-recovery
The data forwarding function of point, realize the fast quick-recovery of inter-domain routing system failure.
In addition, referring to Fig. 2, Fig. 2 simply illustrates the generating process of backup subgraph, and wherein AS is autonomy field system, and its is right
Should be in abstract concept node, i.e., each AS is considered as a node.
Alternatively, can also include referring to Fig. 3, this method：
Step S16, judge generation multiple backup subgraphs in all leaf node set whether with inter-domain routing system
The set of all nodes is identical.If it is not the same, then perform step S17；If identical, step S18 is performed.
Step S17, continue to choose node to be selected, and generate corresponding backup subgraph.
In the present embodiment, if execution of step S17, it can also continue to perform step S16, so repeat, until holding
Row step S18.
Step S18, stop generation backup subgraph.
In the present embodiment, when in multiple backup subgraphs of generation all leaf node set whether with inter-domain routing system
In all nodes set it is identical when, show the backup subgraph quantity of generation each key node in enough inter-domain routing systems
Data forwarding work recovery, and need not additionally generate backup subgraph again, and then can effectively reduce generation backup subgraph
Workload, improve backup subgraph formation efficiency.
It should be noted that in the method, central controlled method can be used, i.e., by the service of inter-domain routing system
Device concentrated controling management, including back up the recovery of sub- map generalization, the collection of nodal information, and failure node etc..Wherein, drum
Encourage autonomy field system and be actively added global route recovery control network, and actively upload all of this Autonomous Domain to central server
Adjacency information, the route based on architectural characteristic is utilized then using global adjacency information as input controlling the central server of network
The backup subgraph of backup subgraph generating algorithm structural texture, participating in node can be from central server when needing route recovery
Download route backup subgraph and implement route recovery.
The embodiment of the present invention is by according to default cascading failure model and default Rule of judgment, in inter-domain routing system
In, the overall connective key node of statistics influence system；In inter-domain routing system, selection removes key node and leaf node
Outer node, as node to be selected；The adjacent node of node to be selected is counted, and according to corresponding to preset rules calculate node to be selected
The weighted value of all adjacent sides of adjacent node；Count the adjacent section of residue of non-key node in all of its neighbor node of node to be selected
Point；Retain in remaining adjacent node, the maximum adjacent side of weighted value in adjacent side corresponding to each node, and delete remaining adjacent node
Other adjacent sides, generate corresponding backup subgraph, as long as so the bottom of inter-domain routing system is linked and is not turned off, can pass through
Backup subgraph, the data forwarding function of fast quick-recovery key node, realize the fast quick-recovery that inter-domain routing system fails.
Embodiment two
The embodiments of the invention provide a kind of failure recovery device of inter-domain routing system, referring to Fig. 4, including：Count mould
Block 100, choose module 200, processing module 300, statistical module 400, generation module 500.
Statistical module 100, for according to default cascading failure model and default Rule of judgment, in inter-domain routing system
In, the overall connective key node of statistics influence system.
It should be noted that in actual applications, when is there is large area paralysis by network attack in inter-domain routing system,
Caused by being failed typically mainly due to key node by attack, in order to be carried out fast and effectively to the inter-domain routing system of failure
Recover, it is necessary to succinct effective link path be provided for the key node of these failures, so that the key node of failure recovers number
According to forwarding.
In the present embodiment, inter-domain routing system is considered as the link structure by being connected between multiple load nodes and node
Into its operation relies on Border Gateway Protocol, i.e. bgp protocol.In inter-domain routing system, if a node because of malfunction and failure,
Then the load on the node can be reallocated on other nodes for giving the node adjacency, if having in these adjacent nodes because
Redistribute load and overload, and then trigger failure, then the load of these nodes can also be reassigned to the adjoining of these nodes
Node, the like, cascading failure model reflects this load reassignment process.
Module 200 is chosen, in inter-domain routing system, choosing the node in addition to key node and leaf node, is made
For node to be selected.
In the present embodiment, leaf node is the node being only connected with another node, i.e., is 0 in discrete mathematics moderate
Node.
Processing module 300, node to be selected is calculated for counting all of its neighbor node of node to be selected, and according to preset rules
The weighted value of all adjacent sides of corresponding adjacent node, it is important in terms of transmitting data stream amount that the weighted value reflects the adjacent side
Degree.
Statistical module 400, the remaining adjacent node of non-key node in all of its neighbor node for counting node to be selected.
In the present embodiment, all of its neighbor relation of inter-domain routing system, can be obtained in several ways, first, be
It is about to respective syntople certainly by each autonomy field system (i.e. each node) to be uploaded in the server of inter-domain routing system；
Meanwhile the server of inter-domain routing system can also be extracted by network from the public route data such as Route Views, RIPE
Adjacent set.Wherein, can be by verifying the neighbour when server is received from autonomy field system voluntarily above-mentioned syntople
Whether the autonomy field system that the relation of connecing is related to, which has, uploads the syntople, if not all uploaded, gives up adjoining pass
System；Identical syntople is uploaded if had, stores the syntople.
Generation module 500, for retaining in remaining adjacent node, the maximum neighbour of weighted value in adjacent side corresponding to each node
Side, and other adjacent sides of remaining adjacent node are deleted, corresponding backup subgraph is generated, the backup subgraph is used to lose in key node
During effect, data forwarding paths are provided for the key node of failure.
It should be noted that having a variety of link connections in inter-domain routing system between multiple nodes, i.e., key node can
Data are forwarded in a manner of there are a variety of link connections.When the attack of key node hand is failed, it is only necessary to for its quick offer one
Effective data transfer path, you can recover the forwarding datamation of failure node, and then fast quick-recovery inter-domain routing system
The purpose of system work, and key node one effective data forwarding paths of offer that subgraph is just used to failure are provided, with
For the fast quick-recovery of key node of failure.
In the present embodiment, if the key node of inter-domain routing system is failed by attack, then as long as Yu Jian roads
It is not turned off by the bottom link (physical connection i.e. between node) of system, may be referred to backup subgraph, the crucial section of fast quick-recovery
The data forwarding function of point, realize the fast quick-recovery of inter-domain routing system failure.
Specifically, default cascading failure model includes：
When again in inter-domain routing system, after node i failure, the adjacent node j of node i is loaded by equation below to be divided
Match somebody with somebody：
lj(t+1)=(lj(t)+Δlj i)；
Wherein,
ΓiRepresent the set of node i all of its neighbor node；Node k is any of set of node i all of its neighbor node
Node；li(0)、lj(0)、lk(0) the original load of node i, node j, node k is represented respectively；D represents the degree of node, the i.e. section
The quantity of point all of its neighbor node；α, β are initial load parameter, and be all higher than zero (such as：1.2,6)；Δlj iFailed for node i
Node j newly-increased load is distributed to afterwards；lj(t) at the time of before being reassigned for load during t, the load on node j；lj(t+1) attach most importance to
After new distribution load, loads of the node j when the moment, t+1 was restarted.
In the present embodiment, above-mentioned formula interior joint i, node j can be any to adjacent node in inter-domain routing system,
After node i failure, the original load in node i can be assigned on the adjacent node of node i according to above-mentioned formula.
In addition, in order to coordinate above-mentioned cascading failure model, when meeting equation below, decision node m is failure node：
Wherein,
lm(0) node m original load is represented；Loaded for node m tolerable；γ is tolerance coefficient, and is more than or waits
In zero (such as：3)；lm(t+1) after to redistribute load, loads of the node j when the moment, t+1 was restarted；fΓmFor node to be measured
In failing and causing the node of failure, the set with node m adjacent nodes；Δlm nDistribute to node m's after being failed for node n
Newly-increased load, can be calculated according to the formula in step S11, repeat no more here.
In the present embodiment, node m is any node in inter-domain routing system, after node failure to be measured, inter-domain routing
The load on other nodes in system can be redistributed, if after redistributing, the load on the node can beyond it
Tolerance load, then the node can fail.
Further, default Rule of judgment can be：
Choose it is overall connective on inter-domain routing system influence preset ratio before ranking (such as：Preceding node 15%), make
For key node.
In the present embodiment, in above-mentioned cascading failure model, the failure effect degree of egress can also be calculated, to quantify
Node connective influence overall on inter-domain routing system, specifically, by equation below come the failure effect degree of calculate node：
Wherein, Ω is the failure effect degree of node；A is the different types of sum of failed link；X represents failed link
Type；NXRepresent the failed link sum that type is X；RXThe weight for the failed link that type is X is represented, if the link is used for
By the information transfer on upper level node to low-level nodes, then the weight of the link is 1.2；If the link is used on node
Node on information transfer to same level, then the weight of the link is 1；If the link is used for the information in bottom layer node
Transmit to upper level node, then the weight of the link is 0.8；| E | it is the sum of inter-domain routing system link.
After the failure effect degree of egress is calculated, it can be sorted from high to low according to failure effect degree, failure effect degree
Ranking is higher, and its connective influence overall on inter-domain routing system is bigger, in actual applications, can be according to the failure shadow of node
The node of preset ratio before ranking is chosen in the sequence of loudness.
Alternatively, processing module 300, it is additionally operable to according to belonging to node by transmitting data flow amount in the adjacent side unit interval
Range intervals, assign the corresponding weighted value of adjacent side.
In the present embodiment, preset rules can be the different range section belonging to the data traffic transmitted on adjacent side, right
Should in corresponding weighted value, such as：When the data traffic transmitted on adjacent side is less than or equal to 500M, its weighted value is 1；It is more than
500M and when being less than or equal to 1G, its weighted value is 2；When more than 1G and being less than or equal to 10G, its weighted value is 3 etc., above number
According to being only for example, specific preset rules can be set according to actual conditions, be not limited here.
Alternatively, also include referring to Fig. 5, the device：Judge module 600.
Judge module 600, for judge generation multiple backup subgraphs in all leaf node set whether with Yu Jian roads
Set by all nodes in system is identical.
Processing module 300, it is additionally operable to, when the judged result of judge module 600 is differs, continue to choose node to be selected,
And generate corresponding backup subgraph；When the judged result of judge module 600 is identical, stop generation backup subgraph.
In the present embodiment, when in multiple backup subgraphs of generation all leaf node set whether with inter-domain routing system
In all nodes set it is identical when, show the backup subgraph quantity of generation each key node in enough inter-domain routing systems
Data forwarding work recovery, and need not additionally generate backup subgraph again, and then can effectively reduce generation backup subgraph
Workload, improve backup subgraph formation efficiency.
The embodiment of the present invention is by according to default cascading failure model and default Rule of judgment, in inter-domain routing system
In, the overall connective key node of statistics influence system；In inter-domain routing system, selection removes key node and leaf node
Outer node, as node to be selected；The adjacent node of node to be selected is counted, and according to corresponding to preset rules calculate node to be selected
The weighted value of all adjacent sides of adjacent node；Count the adjacent section of residue of non-key node in all of its neighbor node of node to be selected
Point；Retain in remaining adjacent node, the maximum adjacent side of weighted value in adjacent side corresponding to each node, and delete remaining adjacent node
Other adjacent sides, generate corresponding backup subgraph, as long as so the bottom of inter-domain routing system is linked and is not turned off, can pass through
Backup subgraph, the data forwarding function of fast quick-recovery key node, realize the fast quick-recovery that inter-domain routing system fails.
The embodiments of the present invention are for illustration only, do not represent the quality of embodiment.
It should be noted that：The failure recovery device for the inter-domain routing system that above-described embodiment provides is realizing inter-domain routing
, can basis only with the division progress of above-mentioned each functional module for example, in practical application during the failure recovery method of system
Need and complete above-mentioned function distribution by different functional modules, i.e., the internal structure of equipment is divided into different function moulds
Block, to complete all or part of function described above.In addition, the failure for the inter-domain routing system that above-described embodiment provides is extensive
The failure recovery embodiment of the method for apparatus for coating and inter-domain routing system belongs to same design, and its specific implementation process refers to method reality
Example is applied, is repeated no more here.
One of ordinary skill in the art will appreciate that hardware can be passed through by realizing all or part of step of above-described embodiment
To complete, by program the hardware of correlation can also be instructed to complete, described program can be stored in a kind of computer-readable
In storage medium, storage medium mentioned above can be read-only storage, disk or CD etc..
The foregoing is only presently preferred embodiments of the present invention, be not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent substitution and improvements made etc., it should be included in the scope of the protection.
Claims (10)
1. a kind of failure recovery method of inter-domain routing system, it is characterised in that including：
According to default cascading failure model and default Rule of judgment, in the inter-domain routing system, influence system is counted
Overall connective key node；
In the inter-domain routing system, the node in addition to the key node and leaf node is chosen, as node to be selected；
Count the adjacent node of the node to be selected, and the adjacent node according to corresponding to preset rules calculate the node to be selected
The weighted value of all adjacent sides；
Count the remaining adjacent node of non-key node in all of its neighbor node of the node to be selected；
Retain in the remaining adjacent node, the maximum adjacent side of weighted value in adjacent side corresponding to each node, and delete described surplus
Other adjacent sides of remaining adjacent node, corresponding backup subgraph is generated, the backup subgraph is used for when key node fails, to lose
The key node of effect provides data forwarding paths.
2. according to the method for claim 1, it is characterised in that the default cascading failure model, including：
When again in the inter-domain routing system, after node i failure, the adjacent node j of node i is loaded by equation below to be divided
Match somebody with somebody：
lj(t+1)=(lj(t)+Δlj i)；
Wherein,
ΓiRepresent the set of node i all of its neighbor node；Node k is any of the set of node i all of its neighbor node node；
li(0)、lj(0)、lk(0) the original load of node i, node j, node k is represented respectively；D represents the degree of node, i.e. the node institute
There is the quantity of adjacent node；α, β are initial load parameter, and are all higher than zero；Δlj iDistribute to node j's after being failed for node i
Newly-increased load；lj(t) at the time of before being reassigned for load during t, the load on node j；lj(t+1) after to redistribute load, section
Loads of the point j when the moment, t+1 was restarted.
3. according to the method for claim 2, it is characterised in that the default Rule of judgment, including：
Choosing influences the node of preset ratio before ranking on the overall connectedness of the inter-domain routing system, as key node.
4. according to the method for claim 1, it is characterised in that described corresponding according to the preset rules calculating node to be selected
Adjacent node all adjacent sides weighted value, including：
Range intervals according to belonging to node by transmitting data flow amount in the adjacent side unit interval, assign the corresponding weight of adjacent side
Value.
5. according to the method described in claim any one of 1-4, it is characterised in that methods described also includes：
Judge generation multiple backup subgraphs in all leaf node set whether with all nodes in the inter-domain routing system
Set it is identical,
If it is not the same, then continuing to choose node to be selected, and generate corresponding backup subgraph；
If identical, stop generation backup subgraph.
A kind of 6. failure recovery device of inter-domain routing system, it is characterised in that including：
Statistical module, for according to default cascading failure model and default Rule of judgment, in the inter-domain routing system,
Count the overall connective key node of influence system；
Module is chosen, in the inter-domain routing system, choosing the node in addition to the key node and leaf node, is made
For node to be selected；
Processing module, the node pair to be selected is calculated for counting the adjacent node of the node to be selected, and according to preset rules
The weighted value of all adjacent sides for the adjacent node answered；
Statistical module, it is additionally operable to count the remaining adjacent node of non-key node in all of its neighbor node of the node to be selected；
Generation module, for retaining in the remaining adjacent node, the maximum adjacent side of weighted value in adjacent side corresponding to each node,
And other adjacent sides of the remaining adjacent node are deleted, corresponding backup subgraph is generated, the backup subgraph is used to save in key
During point failure, data forwarding paths are provided for the key node of failure.
7. device according to claim 6, it is characterised in that the default cascading failure model, including：
When again in the inter-domain routing system, after node i failure, the adjacent node j of node i is loaded by equation below to be divided
Match somebody with somebody：
lj(t+1)=(lj(t)+Δlj i)；
Wherein,
ΓiRepresent the set of node i all of its neighbor node；Node k is any of the set of node i all of its neighbor node node；
li(0)、lj(0)、lk(0) the original load of node i, node j, node k is represented respectively；D represents the degree of node, i.e. the node institute
There is the quantity of adjacent node；α, β are initial load parameter, and are all higher than zero；Δlj iDistribute to node j's after being failed for node i
Newly-increased load；lj(t) at the time of before being reassigned for load during t, the load on node j；lj(t+1) after to redistribute load, section
Loads of the point j when the moment, t+1 was restarted.
8. device according to claim 7, it is characterised in that the default Rule of judgment, including：
Choosing influences the node of preset ratio before ranking on the overall connectedness of the inter-domain routing system, as key node.
9. device according to claim 6, it is characterised in that the processing module, be additionally operable to pass through adjacent side according to node
Range intervals in unit interval belonging to transmitting data flow amount, assign the corresponding weighted value of adjacent side.
10. device according to claim 6, it is characterised in that described device also includes：
Judge module, for judge generation multiple backup subgraphs in all leaf node set whether with the inter-domain routing system
The set of all nodes is identical in system,
The processing module, it is additionally operable to, when the judged result of the judge module is differs, continue to choose node to be selected, and
The corresponding backup subgraph of generation；When the judged result of the judge module is identical, stop generation backup subgraph.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610363674.7A CN107438026A (en) | 2016-05-27 | 2016-05-27 | The failure recovery method and apparatus of inter-domain routing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610363674.7A CN107438026A (en) | 2016-05-27 | 2016-05-27 | The failure recovery method and apparatus of inter-domain routing system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107438026A true CN107438026A (en) | 2017-12-05 |
Family
ID=60454591
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610363674.7A Pending CN107438026A (en) | 2016-05-27 | 2016-05-27 | The failure recovery method and apparatus of inter-domain routing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107438026A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111131028A (en) * | 2019-10-16 | 2020-05-08 | 河南工程学院 | Inter-domain route recovery method based on minimum spanning tree of degree constraint |
| CN111698152A (en) * | 2019-03-15 | 2020-09-22 | 华为技术有限公司 | Fault protection method, node and storage medium |
| CN111862594A (en) * | 2020-06-04 | 2020-10-30 | 中国兵器科学研究院 | Method, device and storage medium for identifying weak unit in road traffic network |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102158397A (en) * | 2011-02-17 | 2011-08-17 | 中兴通讯股份有限公司 | Re-route method and point local repair (PLR) node |
| CN103200096A (en) * | 2013-03-13 | 2013-07-10 | 南京理工大学 | Heuristic routing method avoiding key nodes in complex network |
| CN103326899A (en) * | 2013-06-16 | 2013-09-25 | 周健 | Weighting network node importance assessment method based on network heterogeneity |
-
2016
- 2016-05-27 CN CN201610363674.7A patent/CN107438026A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102158397A (en) * | 2011-02-17 | 2011-08-17 | 中兴通讯股份有限公司 | Re-route method and point local repair (PLR) node |
| CN103200096A (en) * | 2013-03-13 | 2013-07-10 | 南京理工大学 | Heuristic routing method avoiding key nodes in complex network |
| CN103326899A (en) * | 2013-06-16 | 2013-09-25 | 周健 | Weighting network node importance assessment method based on network heterogeneity |
Non-Patent Citations (2)
| Title |
|---|
| 王振兴等: "域间路由系统相继故障问题分析", 《信息工程大学学报》 * |
| 王禹等: "一种基于结构化备份子图的路由系统失效恢复方法", 《电子与信息学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111698152A (en) * | 2019-03-15 | 2020-09-22 | 华为技术有限公司 | Fault protection method, node and storage medium |
| CN111698152B (en) * | 2019-03-15 | 2021-09-14 | 华为技术有限公司 | Fault protection method, node and storage medium |
| CN111131028A (en) * | 2019-10-16 | 2020-05-08 | 河南工程学院 | Inter-domain route recovery method based on minimum spanning tree of degree constraint |
| CN111131028B (en) * | 2019-10-16 | 2021-09-21 | 河南工程学院 | Inter-domain route recovery method based on minimum spanning tree of degree constraint |
| CN111862594A (en) * | 2020-06-04 | 2020-10-30 | 中国兵器科学研究院 | Method, device and storage medium for identifying weak unit in road traffic network |
| CN111862594B (en) * | 2020-06-04 | 2021-09-03 | 中国兵器科学研究院 | Method, device and storage medium for identifying weak unit in road traffic network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kuhn et al. | A self-repairing peer-to-peer system resilient to dynamic adversarial churn | |
| Chawathe et al. | Making gnutella-like p2p systems scalable | |
| Wang et al. | BCDC: a high-performance, server-centric data center network | |
| CN108027805A (en) | Load distribution in data network | |
| Zhang et al. | ExCCC-DCN: A highly scalable, cost-effective and energy-efficient data center structure | |
| Ma et al. | Improved efficient routing strategy on two-layer complex networks | |
| Wang et al. | Reliability-oriented and resource-efficient service function chain construction and backup | |
| Wang et al. | Designing efficient high performance server-centric data center network architecture | |
| Lam et al. | Failure recovery for structured p2p networks: Protocol design and performance evaluation | |
| CN107438026A (en) | The failure recovery method and apparatus of inter-domain routing system | |
| CN106209615A (en) | A kind of Dynamic Route Control method calculating forward-path based on SPFA algorithm | |
| CN102724064B (en) | Method for building network application simulation system | |
| Ding et al. | A MapReduce‐supported network structure for data centers | |
| Rui et al. | Analysis of cascade fault optimization based on regional fault and traffic reallocation in complex networks | |
| Al-Rumaih et al. | Spare capacity planning for survivable mesh networks | |
| Montana et al. | Adaptive reconfiguration of data networks using genetic algorithms | |
| CN104219163B (en) | The load-balancing method that a kind of node based on dynamic copies method and dummy node method dynamically moves forward | |
| CN102647424B (en) | Data transmission method and data transmission device | |
| Yi et al. | SQHCP: Secure-aware and QoS-guaranteed heterogeneous controller placement for software-defined networking | |
| CN107612980B (en) | Adjustable and reliable consistency maintenance method in structured P2P network | |
| Zhu et al. | A hybrid reliable heuristic mapping method based on survivable virtual networks for network virtualization | |
| Zhao et al. | On peer-assisted data dissemination in data center networks: Analysis and implementation | |
| CN107438027A (en) | Inter-domain routing node importance appraisal procedure and device | |
| Nam et al. | Synchronous load balancing in hypercube multicomputers with faulty nodes | |
| Fu et al. | Random choices for churn resilient load balancing in peer-to-peer networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171205 |
|
| RJ01 | Rejection of invention patent application after publication |