CN104618159B - A kind of Internet resources reassignment method based on non-linear capacity load module - Google Patents
A kind of Internet resources reassignment method based on non-linear capacity load module Download PDFInfo
- Publication number
- CN104618159B CN104618159B CN201510044272.6A CN201510044272A CN104618159B CN 104618159 B CN104618159 B CN 104618159B CN 201510044272 A CN201510044272 A CN 201510044272A CN 104618159 B CN104618159 B CN 104618159B
- Authority
- CN
- China
- Prior art keywords
- network
- capacity
- node
- newly
- resource
- 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.)
- Active
Links
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
- Computer And Data Communications (AREA)
Abstract
The invention discloses a kind of Internet resources reassignment method based on non-linear capacity load module, belong to communication network and reliability engineering field.Comprise the following steps：Step one：Non-linear relation based on network normal operating conditions lower node capacity with load, proposes non-linear capacity load module；Step 2：The node capacity load characteristic after cascading failure occurs according to non-linear capacity load module and network, it is proposed that the reassignment policy step 3 for increasing network capacity resource newly：Newly-increased Internet resources reassignment policy is emulated and contrasted.This method by with diplomatic Experimental comparison, newly-increased network capacity resource can more reasonably be distributed after by calculated attack hydraulic performance decline occurs for network by demonstrating proposed resource reassignment policy, network performance is set significantly to be recovered, so as to the destruction that effectively reduction cascading failure is caused to network, cascade failure of removal is preferably resisted.
Description
Technical field
The invention belongs to communication network and reliability engineering field, and in particular to one kind loads mould based on non-linear capacity
The Internet resources reassignment method of type.
Background technology
In actual environment, people cause after network performance decline, just newly increased certain often in network failure
Internet resources carry out sub-distribution again to some network nodes, for example, performance upgrade is carried out to these nodes, so that performance of the entire network is obtained
Recovery on to a certain extent, meets requirement of the people to some network service capabilities.Such issues that in Practical Project very
It is common, in the case of input certain cost, network system is obtained robustness as well as possible similar.
Internet resources are optimized with configuration can improve network transmission performance and control network failure propagation, all the time
Have been a great concern.Internet resources allocation strategy mainly includes in the application of network reliability research direction at present：Network
Control and prevention of the resources configuration optimization to network failure propagation phenomenon, have been achieved for certain progress, specifically at present
Research angle be also not quite similar.
But, existing most of researchs still suffer from certain limitation in terms of the analysis, evaluation of Internet resources allocation strategy
Property with it is not enough, for example, the validity and feasibility of Network resource allocation model can not be weighed comprehensively, and network section can not be combined
The actual characteristic of point, such as the influence of node capacity and load relationship, for the purpose of the prevention and control of network system cascading failure
Carry out the research of network capacity optimization distribution.
After cascading failure occurs for network, the network capacity resource newly increased is reallocated, network performance can be made
Necessarily recovered.Current network resource optimization allocation strategy is that the initial resource for being directed to network is allocated mostly, is ignored
For distributing rationally for follow-up newly-increased Internet resources.
For example：Ghamry W K, Elsayed K M F document " Network design methods for
Mitigation of intentional attacks in scale-free networks ", simply carry out simple resource
Configure without considering the real time information in network, while also not accounting for occurring cascading failure under network traffics DYNAMIC DISTRIBUTION
Situation；
Prior art：Paolo Crucitti, Vito Latora document " A model for cascading
The situation for occurring cascading failure is considered in failures in complex networks ", is proposed according to linear ratio relation
Linear capacity load module, does not account for the resource allocation policy of network capacity and load nonlinear characteristic.
The content of the invention
The purpose of the present invention is：On the premise of network capacity and load nonlinear characteristic is considered, attacked for network
A kind of situation of the raw cascading failure of percussion, it is proposed that Internet resources reassignment method based on non-linear capacity load module.
Comprise the following steps：
Step one：Non-linear relation based on network normal operating conditions lower node capacity with load, proposes non-linear hold
Measure load module；
Ci=α × Li(0)+α×Li(0)1-β
The ＜ β ＜ 1 of i=1,2...N, α >=1,0
In formula：Arbitrary node i capacity is C in networki, node i is L in the load of initial timei(0), α joins for tolerance
Number, β is nonlinear factor.
Step 2：The node capacity after cascading failure occurs according to non-linear capacity load module and network and loads spy
Property, it is proposed that the reassignment policy for increasing network capacity resource newly；
Reassignment policy for increasing network capacity resource newly, is represented with Cnlinr：
In formula：CI increasesIt is the newly-increased capacity of node i after resource is reallocated；θ is scale parameter, 0 ＜ θ≤1；Network is sent out
When hydraulic performance decline is to the stable state T moment after raw cascading failure, the load of node i is Li(T).η, γ are nonlinear factor.
Step 3：Newly-increased Internet resources reassignment policy is emulated and contrasted；
By emulation experiment, the resource reassignment policy proposed is compared with existing three kinds of resource reassignment policies
Compared with.
The specific steps of emulation：
Step 3.2.1：Network topology structure figure is set up, while removing the node that maximum is loaded in network；
Step 3.2.2：After cascade failure occurs, calculating network reaches the network efficiency E (T) during the new stable state T moment.
eijRepresent the efficiency of transmission of the shortest path between node i and node j.
Step 3.2.3：Judge network efficiency E (T) enters step 3.2.4 after whether reaching stabilization, stabilization, otherwise, return
Step 3.2.2 is recalculated；
Step 3.2.4：Newly-increased network capacity resource is allocated according to four kinds of resource reassignment policies respectively；
Step 3.2.5：After newly-increased resource reallocation, calculating network reaches the network efficiency E (f) during stable state again；
Step 3.2.6：Judge network efficiency E (f) terminates after whether reaching stabilization, stabilization, otherwise, return to step 3.2.5.
Advantages of the present invention is with good effect：
(1) a kind of Internet resources reassignment method based on non-linear capacity load module of the present invention, according to live network
Capacity load feature, it is proposed that more meet the non-linear capacity load module of real network.
(2) a kind of Internet resources reassignment method based on non-linear capacity load module of the present invention, it is newly-increased for network
The reassignment policy of capacity resource, by with diplomatic Experimental comparison, it was demonstrated that the resource reassignment policy proposed exists
Network can more reasonably distribute newly-increased network capacity resource after occurring hydraulic performance decline by calculated attack, obtain network performance
Significantly recover, so that the destruction that effectively reduction cascading failure is caused to network, preferably resists cascade failure of removal.
(3) a kind of Internet resources reassignment method based on non-linear capacity load module of the present invention, what is proposed is newly-increased
Internet resources reassignment policy for infrastructure networks systems reliability design and cascade failure of removal prevention,
Control is respectively provided with important references value.
Brief description of the drawings
Fig. 1 is the Internet resources reassignment method flow chart of the invention based on non-linear capacity load module；
Fig. 2 is four kinds of live networks of the invention node capacity and relation of load in normal state；
Fig. 3 is the simulation contact surface of the newly-increased network capacity resource reallocation of the present invention；
Fig. 4 is the variation diagram of network efficiency after BA networks of the present invention are reallocated through capacity；
Fig. 5 is the variation diagram of network efficiency after ER networks of the present invention are reallocated through capacity；
Fig. 6 is the variation diagram of network efficiency after WS networks of the present invention are reallocated through capacity.
Embodiment
Below in conjunction with drawings and examples, the present invention is described in further detail.
The present invention proposes a kind of Internet resources reassignment method based on non-linear capacity load module, passes through emulation experiment
Compared with existing three kinds of resource reassignment policies, illustrate the feasibility and validity of proposed resource reallocation.
As shown in figure 1, comprising the following steps：
Step one：Non-linear relation based on network normal operating conditions lower node capacity with load, proposes non-linear hold
Measure load module.
According to the capacity load feature of live network, as shown in Fig. 2 the data used for：(a) 2005 it is international in the U.S.
Corresponding relation between the seating capacity and the seating capacity of offer that are used in the aircraft of airport landing；(b) U.S. in 2005
Corresponding relation between vehicle flowrate/hour of the state of Colorado section expressway design and the terrain vehicle capacity of estimation；(c)
The actual loading of Texas electrical grid transmission line and the relation between corresponding capacity during summer 2000 peak of power consumption；(d)2006
Year June MIT and Princeton University between campus network backbone routing interface moon average flow and the corresponding relation of bandwidth；
Understand in real network, be not simple linear ratio relation between the capacity of node and load, in inhomogeneity
A kind of similar non-linear capacity load characteristic is shown in the live network of type, between the capacity of node and load.
It can be seen from the hypothesis described in prior art, arbitrary node i capacity C in networkiNode is proportional to initial
The load L at momenti(0) the linear capacity load module, and according to linear ratio relation obtained is as follows：
Ci=α × Li(0), i=1,2...N
Wherein：α is tolerance parameter, α >=1.
On the basis of existing linear capacity load module, the node capacity load relationship built using network True Data
Curve, passes through curve matching, it is proposed that a kind of non-linear capacity load module for more meeting real network normal condition, as follows：
Ci=α × Li(0)+α×Li(0)1-β
The ＜ β ＜ 1 of i=1,2...N, α >=1,0
In formula：β is nonlinear factor, and the corresponding relation between node capacity load can be adjusted flexibly.
Step 2：The node capacity after cascading failure occurs according to non-linear capacity load module and network and loads spy
Property, it is proposed that the reassignment policy for increasing network capacity resource newly.
Node capacity load characteristic before and after described cascade failure, refers to that network occurs before cascading failure, network
Capacity with load be in non-linear relation, and when network occur cascading failure after, network capacity with load corresponding relation dynamic
Situation of change.
After by calculated attack cascading failure occurs for network, the non-linear capacity load module obtained according to step one, with
And network reaches the stable state time after the non-linear relation in network node capacity load, network node initial load and cascading failure
On the basis of the load of point is known, it is proposed that increase the reassignment policy of network capacity resource after being failed for cascade newly, use
Cnlinr is represented：
In formula：CI increasesIt is the newly-increased capacity of node i after resource is reallocated, the former capacity of node i is Ci；θ joins for ratio
Number；Node i is L in the load of initial timei(0), hydraulic performance decline occurs after cascading failure for network to node i during the stable state T moment
Load be Li(T).η, γ are nonlinear factor, can adjust the reallocation relation that each node increases capacity newly.
Scale parameter θ is used for controlling newly-increased capacity CI increasesSize, 0 ＜ θ≤1, θ size is according to actual available newly-increased
Capacity is adjusted, and has reacted the constraint of cost in practice.
It is load and the ratio of former capacity of the node i in initial time, node i when reflecting network normal work
State.During normal work the non-linear capacity load characteristic of node i will influence network overload failure propagation, so make η >=
1, newly-increased capacity is distributed accordingly；Capacity CiBigger, the ratio is smaller, and the reliability of network is higher, initial capacity degree of redundancy
It is bigger, represent that the robustness of initial network is higher.
During for network after cascading failure up to the stable state T moment, the ratio between the load of node i and the former capacity of node i, reflection
Load redistribution posterior nodal point i new steady s tate.The distribution of newly-increased capacity is now carried out, so needing to consider
Value, make 0 ＜ γ≤1 carry out control ratioWeighing factor to increasing capacity resource newly.
|Li(T)-Ci| the difference between the former capacity of load and node i for cascading failure posterior nodal point i, reflect cascading failure
The intensity of variation of posterior nodal point load, will directly affect the size of newly-increased capacity.
Step 3：Newly-increased Internet resources reassignment policy is emulated and contrasted；
The newly-increased Internet resources reassignment policy proposed in step 2 is emulated, and divided again with existing three kinds of resources
The contrast of simulation result is carried out with strategy；
Step 3.1, three kinds of existing resource reassignment policies are as follows：
(1) node i increases the degree that capacity is proportional to node newly, is represented with CDlinr：
CI increases=b × Di, i=1,2...N (1)
DiThe number of degrees of initial network node i are represented, b represents proportionality coefficient.
This method, which does not account for network, to be occurred after cascading failure, and the topological structure of network has occurred that change, causes section
The number of degrees of point are changed, and also do not account for the situation of network traffics DYNAMIC DISTRIBUTION.
(2) node i increases the initial load that capacity is proportional to node newly, is represented with CLlinr：
CI increases=b × Li(0), i=1,2...N (2)
This method only considered network it is initial when load distribution situation, do not account for network occur cascading failure after, when
Network reaches that the load of steady timed node is changed again.
(3) the newly-increased capacity of each node is equal, is represented with Ceql：
CI increases=c, i=1,2...N (3)
C is constant, represents that the newly-increased capacity of each network node i is equal.
This method is that newly-increased capacity is simply given into each node, has not both accounted for topology after cascade failure
The change of structure, does not account for the dynamic change situation of node load yet.
Step 3.2, newly-increased Internet resources reassignment policy is emulated.
The specific steps of emulation are as shown in Figure 3：
Step 3.2.1：Network topology structure figure is set up, while removing the node that maximum is loaded in network；
According to real network, network topology structure figure G (N, K) is set up, wherein N represents network node i number, and K is represented
The bar number on side in network；The initial load L of N number of node is determined according to network topology structure figurei(0) value.
The non-linear capacity load module proposed according to step one, with reference to given tolerance parameter and nonlinear factor, enters
One step determines N number of network node i capacity Ci。
Simulation causes the node failure that maximum is loaded in network, causes network for the calculated attack in real network
Cascading failure, so to remove the node that maximum is loaded in network.
Step 3.2.2：After cascade failure occurs, calculating network reaches the network efficiency E (T) during the new stable state T moment.
eijRepresent the efficiency of transmission of the shortest path between node i and node j.
Step 3.2.3：Judge network efficiency E (T) enters step 3.2.4 after whether reaching stabilization, stabilization, otherwise, return
Step 3.2.2 is recalculated；
Step 3.2.4：Newly-increased network capacity resource is allocated according to four kinds of resource reassignment policies respectively；
When network is after cascading failure reaches stable state again, increases certain network capacity resource newly to network, utilize step
The resource reassignment policy and existing three kinds of resource reassignment policies proposed in rapid two is reallocated respectively to newly-increased capacity.
Step 3.2.5：After newly-increased resource reallocation, calculating network reaches the network efficiency E (f) during stable state again；
Step 3.2.6：Judge network efficiency E (f) terminates after whether reaching stabilization, stabilization, otherwise, return to step 3.2.5；
Step 3.3, newly-increased Internet resources reassignment policy is contrasted；
Scales-free network, ER random networks and the class network of WS small-world networks three are contrasted according to simulation flow：
On the premise of available new total increase-volume amount is the same, first, stored for scales-free network, also known as BA networks
The situation of meaning attack, the specific data of amplification Δ R of network efficiency are shown in Table 1.Setting node N takes 100, K to take 500, non-linear capacity
In load module, parameter alpha takes α=1.01 and α=1.05, β=0.5 respectively；Set resource reassignment policy in parameter γ=
0.3, θ=0.8, η=2, the amplification Δ R of network efficiency is：
Δ R=(E (f)-E (T))/E (T) (6)
Node is reallocated by four kinds of allocation strategies and increased newly after capacity, the variation tendency of network efficiency is as shown in figure 4, work as net
After network cascading failure occurs, network reaches the minimum point that the network efficiency E (T) at stable state T moment is each network efficiency value in figure, together
When, after newly-increased resource reallocation, network reaches that the network efficiency E (f) at stable state f moment is each network efficiency value in figure again
Peak.
As α=1.01, E (T) the value average under 4 kinds of strategies is：0.384；E (f) value：It is under Ceql strategies
0.4562；It is 0.4629 under CLlinr tactics；It is 0.4634 under CDlinr strategies；It is under Cnlinr strategies
0.4795；
So the amplification Δ R of network efficiency is calculated as：
Under Ceql strategies, Δ R is：(0.4562-0.384)/0.384=18.79%；
Under CLlinr strategies, Δ R is：(0.4629-0.384)/0.384=20.55%；
Under CDlinr strategies, Δ R is：(0.4634-0.384)/0.384=20.68%；
Under Cnlinr strategies, Δ R is：(0.4795-0.384)/0.384=24.87%；
Similarly, as α=1.05, E (T) the value average under 4 kinds of strategies is：0.3903；E (f) value：In Ceql strategies
It is 0.4572 down；It is 0.4617 under CLlinr tactics；It is 0.4622 under CDlinr strategies；It is under Cnlinr strategies
0.4805；
The amplification Δ R of network efficiency is calculated as：
Under Ceql strategies, Δ R is：(0.4572-0.3903)/0.3903=17.13%；
Under CLlinr strategies, Δ R is：(0.4617-0.3903)/0.3903=18.28%；
Under CDlinr strategies, Δ R is：(0.4622-0.3903)/0.3903=18.41%；
Under Cnlinr strategies, Δ R is：(0.4805-0.3903)/0.3903=23.1%；
Newly-increased network capacity accounts for the ratio of the former capacity of network：Δ C is the ratio that newly-increased network capacity accounts for the former capacity of network：
According to the parameter setting in table 1, the initial load L determined during network topology structure is builti(0), network node i
Capacity Ci, and it is L that the load of hydraulic performance decline to node i during the stable state T moment after cascading failure, which occurs, for networki(T), according to
Cnlinr strategy formula calculate the newly-increased capacity C for obtaining each nodeI increases, in α=1.01, CI increasesSummation is chosen for 5012.1；CiAlways
With for 11393；The ratio Δ C that its newly-increased network capacity accounts for the former capacity of network is：5012.1/11393=43.99%；
In α=1.05, CI increasesSummation chooses 4391.2；CiSummation is 11845；Its newly-increased network total capacity accounts for network original appearance
The ratio Δ C of amount is：
The network efficiency amplification Δ R that the BA networks of table 1 are obtained under different allocation strategies
Suffer from table 1 and Fig. 4, BA network after calculated attack, under conditions of newly-increased total capacity resource is equal, use
Cnlinr resource reassignment policies, the efficiency amplification of network is maximum, network efficiency can be made to obtain maximum lift.For other rule
The simulation result of mould BA networks has also obtained identical conclusion.
Then, for ER random networks by after calculated attack, the specific data of amplification Δ R of network efficiency are shown in Table 2；By
Capacity is reallocated, the variation tendency of network efficiency as shown in figure 5, setting node N take α in 100, non-linear capacity load module=
1.05, β=0.5；Parameter γ=0.3, θ=0.6, η=2.3 in resource reassignment policy.
CI increasesSummation chooses 2740.5；CiSummation is 13464；The ratio Δ C that its newly-increased network capacity accounts for the former capacity of network is：
The amplification Δ R of network efficiency is calculated as：
Under Ceql strategies, Δ R is：(0.468-0.4273)/0.4273=9.52%；
Under CLlinr strategies, Δ R is：(0.4708-0.4273)/0.4273=10.19%；
Under CDlinr strategies, Δ R is：(0.4746-0.4273)/0.4273=11.06%；
Under Cnlinr strategies, Δ R is：(0.4821-0.4273)/0.4273=12.82%；
Network efficiency amplification Δ R of the ER networks of table 2 under different allocation strategies
From simulation result Fig. 5 and table 2, in addition to network reaches that the transit time of stable needs is longer, it is compared
His three kinds of reassignment policies, the network efficiency amplification of the ER networks by calculated attack can be made using Cnlinr reassignment policies
It is bigger.It is obvious that on the problem of infrastructure network is protected, in the case of available resources identical, making network performance obtain
It is a greater degree of to recover to be very attractive.Emulation experiment for other scales ER networks has also drawn similar knot
By this has further showed that Cnlinr increases the applicability and validity of Internet resources reassignment policy newly.
Finally, emulated for WS small-world networks, set nodes N=100, network is by after calculated attack, net
The specific data of amplification Δ R of network efficiency are shown in Table 3；After being reallocated through overcapacity, the variation tendency of network efficiency is as shown in Figure 6.Its
α=1.04, β=0.3 in middle non-linear capacity load module；Parameter γ=0.33, θ=0.66, η in resource reassignment policy=
1.9。
CI increasesSummation chooses 3251.5；CiSummation is 13022；The ratio Δ C that its newly-increased network capacity accounts for the former capacity of network is：
The amplification Δ R of network efficiency is calculated as：
Under Ceql strategies, Δ R is：(0.4643-0.4061)/0.4061=14.34%；
Under CLlinr strategies, Δ R is：(0.4588-0.4061)/0.4061=12.98%；
Under CDlinr strategies, Δ R is：(0.464-0.4061)/0.4061=14.26%；
Under Cnlinr strategies, Δ R is：(0.4782-0.4061)/0.4061=17.77%；
Network efficiency amplification Δ R of the WS networks of table 3 under different allocation strategies
It was found from Fig. 6 and table 3, network efficiency has than larger shake and reached after being just allocated except newly-increased capacity
Beyond transit time needed for stable state is longer, compared to the simulation result of other three kinds of reassignment policies, reallocated using Cnlinr
Strategy can be such that the network performance of the WS networks by calculated attack is greatly recovered, the stabilization of network efficiency finally
Value is bigger, and similarly supports above-mentioned conclusion for the simulation result of other similar scale networks.
Claims (1)
1. a kind of Internet resources reassignment method based on non-linear capacity load module, it is characterised in that including following
Step：
Step one：Non-linear relation based on network normal operating conditions lower node capacity with load, proposes that non-linear capacity is born
Carry model；
Ci=α × Li(0)+α×Li(0)1-β
The ＜ β ＜ 1 of i=1,2...N, α >=1,0
Wherein, CiFor the capacity of arbitrary node i in network, Li(0) it is load of the node i in initial time, α is tolerance parameter, β
For nonlinear factor；
Step 2：Node capacity load characteristic after cascading failure is occurred according to non-linear capacity load module and network, obtained
The reassignment policy for newly-increased network capacity resource is arrived；
It is as follows for the reassignment policy of newly-increased network capacity resource：
Wherein, CI increasesIt is the newly-increased capacity by resource reallocation posterior nodal point i；θ is scale parameter, 0 ＜ θ≤1；Network is cascaded
When hydraulic performance decline is to the stable state T moment after failure, the load of node i is Li(T)；η, γ are nonlinear factor；
Step 3：Newly-increased Internet resources reassignment policy is emulated and contrasted；
By emulation experiment, the resource reassignment policy proposed is compared with existing three kinds of resource reassignment policies；
Existing three kinds of resource reassignment policies are as follows:
(1) node i increases the degree that capacity is proportional to node newly, is represented with CDlinr：
CI increases=b × Di, i=1,2...N
DiThe number of degrees of initial network node i are represented, b represents proportionality coefficient；
This method, which does not account for network, to be occurred after cascading failure, and the topological structure of network has occurred that change, causes node
The number of degrees are changed, and also do not account for the situation of network traffics DYNAMIC DISTRIBUTION；
(2) node i increases the initial load that capacity is proportional to node newly, is represented with CLlinr：
CI increases=b × Li(0), i=1,2...N
This method only considered network it is initial when load distribution situation, do not account for network occur cascading failure after, work as network
Reach that the load of steady timed node is changed again；
(3) the newly-increased capacity of each node is equal, is represented with Ceql：
CI increases=c, i=1,2...N
C is constant, represents that the newly-increased capacity of each network node i is equal；
This method is that newly-increased capacity is simply given into each node, has not both accounted for topological structure after cascade failure
Change, do not account for the dynamic change situation of node load yet；
What is emulated comprises the following steps that：
Step 3.2.1：Network topology structure figure is set up, while removing the node that maximum is loaded in network；
Step 3.2.2：After cascade failure occurs, calculating network reaches the network efficiency E (T) during the new stable state T moment；
eijRepresent the efficiency of transmission of the shortest path between node i and node j；N represents network node i number；
Step 3.2.3：Judge network efficiency E (T) enters step 3.2.4, otherwise, return to step after whether reaching stabilization, stabilization
3.2.2 recalculate；
Step 3.2.4：Newly-increased network capacity resource is allocated according to four kinds of resource reassignment policies respectively；
Step 3.2.5：After newly-increased resource reallocation, calculating network reaches the network efficiency E (f) during stable state again；
Step 3.2.6：Judge network efficiency E (f) terminates after whether reaching stabilization, stabilization, otherwise, return to step 3.2.5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510044272.6A CN104618159B (en) | 2015-01-28 | 2015-01-28 | A kind of Internet resources reassignment method based on non-linear capacity load module |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510044272.6A CN104618159B (en) | 2015-01-28 | 2015-01-28 | A kind of Internet resources reassignment method based on non-linear capacity load module |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104618159A CN104618159A (en) | 2015-05-13 |
| CN104618159B true CN104618159B (en) | 2017-09-05 |
Family
ID=53152450
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510044272.6A Active CN104618159B (en) | 2015-01-28 | 2015-01-28 | A kind of Internet resources reassignment method based on non-linear capacity load module |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104618159B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107872344B (en) * | 2016-09-28 | 2022-01-18 | 华为技术有限公司 | Network data acquisition method and related device |
| CN106789376B (en) * | 2017-03-24 | 2020-09-22 | 大连大学 | Construction method of cascade failure model of command control network with hierarchical structure |
| CN110162912B (en) * | 2019-05-30 | 2020-09-01 | 北京航空航天大学 | Network resource optimization method considering resource competition coupling |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1377562A (en) * | 1999-09-30 | 2002-10-30 | 高通股份有限公司 | Closed loop resource allocation in high speed wireless communications network |
| CN101399708A (en) * | 2007-09-28 | 2009-04-01 | 华为技术有限公司 | Method and device for establishing network performance model |
-
2015
- 2015-01-28 CN CN201510044272.6A patent/CN104618159B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1377562A (en) * | 1999-09-30 | 2002-10-30 | 高通股份有限公司 | Closed loop resource allocation in high speed wireless communications network |
| CN101399708A (en) * | 2007-09-28 | 2009-04-01 | 华为技术有限公司 | Method and device for establishing network performance model |
| US8478570B2 (en) * | 2007-09-28 | 2013-07-02 | Huawei Technologies Co., Ltd. | Method and apparatus for establishing network performance model |
Non-Patent Citations (2)
| Title |
|---|
| "an improved model for cascading failures in complex networks";Jian Zhou.etc;《cloud computing and intelligent system(CCIS).2012 IEEE 2nd international conference on》;20121114;全文 * |
| "复杂网络上级联失效的负载容量模型";窦炳琳，张世永;《系统仿真学报》;20110731;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104618159A (en) | 2015-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sharma et al. | Risk-energy aware service level agreement assessment for computing quickest path in computer networks | |
| Ji et al. | Improving interdependent networks robustness by adding connectivity links | |
| Zhong et al. | Restoration of interdependent network against cascading overload failure | |
| Shakkottai et al. | Network optimization and control | |
| CN111193629B (en) | Fault propagation method for dynamic load cascade failure of multilayer information network | |
| CN101841482B (en) | Energy-saving routing method and device for network of data center | |
| CN103812949B (en) | A kind of task scheduling towards real-time cloud platform and resource allocation methods and system | |
| Zhao et al. | Cascading failures in interconnected networks with dynamical redistribution of loads | |
| CN105554102A (en) | Elastic expansion method based on container cluster and application system thereof | |
| CN104618159B (en) | A kind of Internet resources reassignment method based on non-linear capacity load module | |
| CN103780467B (en) | communication connection method, communication device and communication system | |
| Borylo et al. | Latency and energy-aware provisioning of network slices in cloud networks | |
| CN113645146B (en) | New stream density-based software defined network controller load balancing method and system | |
| Li et al. | Cost-minimizing bandwidth guarantee for inter-datacenter traffic | |
| CN101267315B (en) | An irregular topology structure generation method for chip network | |
| CN104639437A (en) | Forwarding method and apparatus of broadcast messages in stack system | |
| CN102082734B (en) | Service message transmission method and equipment | |
| Sandoval-Arechiga et al. | Shifting the network-on-chip paradigm towards a software defined network architecture | |
| CN104464382A (en) | Aviation network cascade effect regulation and control method and device | |
| CN106230737B (en) | A kind of software definition network-building method of state aware | |
| Liu et al. | Robustness of networks formed from interdependent correlated networks under intentional attacks | |
| CN107734475A (en) | Note transmission method and business platform based on short message link | |
| CN109739513B (en) | Dynamic scheduling method and device for service requests under multi-edge cloud | |
| Wang et al. | Robustness of spontaneous cascading dynamics driven by reachable area | |
| CN103986744B (en) | Throughput-based file parallel transmission method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |