CN103780445B - A kind of network flow monitoring system and method for threshold adaptive amendment - Google Patents
A kind of network flow monitoring system and method for threshold adaptive amendment Download PDFInfo
- Publication number
- CN103780445B CN103780445B CN201210401916.9A CN201210401916A CN103780445B CN 103780445 B CN103780445 B CN 103780445B CN 201210401916 A CN201210401916 A CN 201210401916A CN 103780445 B CN103780445 B CN 103780445B
- Authority
- CN
- China
- Prior art keywords
- flow
- data
- module
- threshold
- value
- 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
Abstract
The invention provides a kind of network flow monitoring system and method for threshold adaptive amendment, including flow collection module, adaptive threshold correcting module, flow monitoring module and alarm display module；Flow collection module obtains flow information data from interchanger, is sent to adaptive threshold correcting module and flow monitoring module；Adaptive threshold correcting module, which connects, to be confirmed to threshold parameter initial value or threshold parameter is updated, and sends the threshold parameter after renewal to flow monitoring module；The more current flow of flow monitoring module and historical traffic situation, judge whether network occurs integrated flux mutation, save as flow histories data using present flow rate as historical traffic after the completion of judgement, and will determine that result is sent to alarm display module；Alarm display module is shown according to the result of calculation of flow detection module transfer by human-computer interaction interface.
Description
Technical field
The invention belongs to computer communication technology field, and in particular to a kind of network flow monitoring of threshold adaptive amendment
System and method.
Background technology
With the development of network technology, the bandwidth of network is constantly becoming big, and the speed that can be transmitted also becomes higher and higher, and one
As for situation, due to the presence of the load-balancing mechanism in procotol, the load of network generally each node is
Relative equilibrium.And flow is mutated when typically occurring in network failure, so abrupt climatic change is for ensureing internet security
Play an important roll with reliability.
Currently, the algorithm of the data flow abrupt climatic change with self-adaptive features has had many correlative studys.It is overall
For upper, in order to meet adaptive flow abrupt climatic change, abrupt climatic change is mainly carried out by the aggregation of logarithm value.Lead to
The comparison of current network flow and web-based history flow is crossed, comparative result and threshold value are compared judgement, draws whether occur
It is abnormal, but existing technical scheme put into specific Network Traffic Monitoring, there is problems with：
It is well known that the outlier threshold selection of conventional solution is all a fixed value, or perhaps empirical value, for
The data transmission system in the fields such as Aerospace test, change of network environment is very big, and threshold value, which is chosen, can not possibly lean on empirical value, institute every time
Techniques described above can not be applied to the situation of dynamic network traffic real-time change.Accordingly, it is desirable to provide a side being more suitable for
Case, it is not only adaptive that efficient exception monitoring is carried out to network traffics, while the network environment at the place to current system
Have one it is adaptive well, i.e., the threshold value to current exception monitoring also carries out adaptive correction, is waited with meeting Aerospace test
The demand that web database technology is continually changing in journey.
The content of the invention
The purpose of the present invention is the defect for overcoming prior art, by introducing the mechanism of adaptive correction come to mutation threshold value
Carry out adaptive updates, it is possible to resolve threshold value, which is fixed, in conventional solution causes the disadvantage that it is not applied for dynamic network environment
End, improves the adaptability and maintainability of system.
In order to achieve the above object, the technical scheme is that, a kind of network flow monitoring of threshold adaptive amendment
System, including flow collection module, adaptive threshold correcting module, flow monitoring module and alarm display module；Wherein, flow
Acquisition module obtains flow information data from interchanger, is sent to adaptive threshold correcting module and flow monitoring module；It is adaptive
Answer threshold values correcting module to receive the flow information data that flow acquisition module is transmitted, threshold parameter initial value is confirmed or right
Threshold parameter is updated, and sends the threshold parameter after renewal to flow monitoring module；Flow monitoring module is more current
Flow and historical traffic situation, judge whether network occurs integrated flux mutation, using present flow rate as going through after the completion of judgement
History flow saves as flow histories data, and will determine that result is sent to alarm display module；Alarm display module is according to flow
The result of calculation of monitoring modular transmission, is shown by human-computer interaction interface, is alarmed if occurring integrated flux mutation
Prompting, does not occur, and points out normal.
A kind of network flow monitoring method of threshold adaptive amendment, comprises the following steps：
Step 1, flow collection module from interchanger obtain flow information data, be sent to adaptive threshold correcting module and
Flow monitoring module；
Step 2, adaptive threshold correcting module receives the flow information data that flow acquisition module is transmitted, to threshold parameter
Initial value is confirmed or threshold parameter is updated, and sends the threshold parameter after renewal to flow monitoring module；
Step 3, the more current flow of flow monitoring module and historical traffic situation, judge whether network occurs cumulative flow
Amount mutation, saves as flow histories data, and will determine that result is sent to after the completion of judgement using present flow rate as historical traffic
Alarm display module；
Step 4, alarm display module is carried out according to the result of calculation of flow monitoring module transfer by human-computer interaction interface
It has been shown that, carries out alarm if occurring integrated flux mutation, does not occur, and points out normal.
It is described to save as flow histories data, completed by assembling historical data storage method, the process of implementing is：It is right
Data flows of traffic x1,x2,...,xnIn element i (1≤i≤n) ask prefix and, be designated as F (i), i.e.,If si
=F (i), calculating obtains prefix and data flow s1,s2,...,sn, it is designated as first group of aggregation historical data；Then, by data flow
s1,s2,...,snIt is stored in the sliding window that length is n；The length of data flow is limited as n, i.e., the size of sliding window is not
Become, when new data arrives, existing s is attached to successively1,s2,...,snOn, it can obtain new data stream s1+xn+1, s2+
xn+1,...,sn+xn+1；Ensure that the size of window is constant, by xn+1As first data, while by data sn+xn+1Reject, obtain
To the data flow x of n bit lengthsn+1,s1+xn+1,...,sn-1+xn+1, it is designated as s '1,s′2,...,s′n, similarly, as addition xn+2When
Obtain data flow xn+2,xn+1+xn+2,s1+xn+1+xn+2,...,sn-2+xn+1+xn+2, it is designated as s "1,s″2,...,s″n；The like,
Update aggregation historical data successively by this method.
The threshold parameter is β, then its initial value isIn formula, Cr(j)
=(F (2j)-F (j))/F (j)；Its updated value is
Wherein, when updating first time, β is initial value, and when updating next time, β takes the β that last round of formula is drawnnewValue, α joins for weighting
Count, generally α can use a value between 0.5-0.99；α value depends on the size of window, if window is larger,
So α is also larger, if window is smaller, then α is also smaller, and when length of window is 1, it is 0.5 to take α.
It is described to judge whether network occurs integrated flux mutation, completed by abrupt climatic change ergodic algorithm, only on flow
The situation of liter is monitored, and does not detect that the process of implementing is to the unexpected decline situation of flow：
Step 1, for the sliding window that length is n, the algorithm collects data n in advance, i.e., sliding window data is filled up
Afterwards, step 2 is transferred to, the algorithm that brings into operation carries out abrupt climatic change；
Step 2, the algorithm travels through i from 1 to n/2, judges s2i≥(β+1)(s2i-si) whether (β ＞ 1) set up, such as set up,
Outputting alarm, shows have mutation to produce；Otherwise, do not point out, continue to run with algorithm until traversal is completed；After the completion of, such as data flow
New element is not reached, then is continued waiting for, and is otherwise transferred in step 3, formula, s2iFor F (2i), siFor F (i)；
Step 3, when new element is reached in data flow, after being updated the data according to aggregation historical data storage method, again
The algorithm is run to carry out abrupt climatic change；
Beneficial effects of the present invention are：For the space flight network test process that change of network environment is larger, the present invention is proposed
Using integrated flux mutation monitoring method detection network mutation, and threshold parameter is set using the method for adaptive correction
Put, frequently test and adjustment threshold parameter can be needed by threshold values in the past by overcoming before setting, and the parameter set does not adapt to dynamic
The drawbacks of network traffics change, is that experimentation saves substantial amounts of manpower, while also ensure that the correctness of parameter and adaptive
Ying Xing, effective foundation is provided for the abrupt climatic change of network traffics, it is ensured that test process is smoothed out.
Brief description of the drawings
Fig. 1 is a kind of network flow monitoring system schematic of threshold adaptive amendment.
Fig. 2 is a kind of network flow monitoring method flow chart of threshold adaptive amendment.
Fig. 3 is the sliding window schematic diagram that length is n.
Fig. 4 is aggregation historical data schematic diagram.
Fig. 5 is threshold value update process schematic.
Embodiment
The present invention is described further with reference to the accompanying drawings and examples.
A kind of network flow monitoring system of threshold adaptive amendment, as shown in figure 1, including flow collection module, adaptive
Answer threshold values correcting module, flow monitoring module and alarm display module；Wherein, flow collection module obtains flow letter from interchanger
Data are ceased, adaptive threshold correcting module and flow monitoring module is sent to；Adaptive threshold correcting module receives flow collection
The flow information data that module is transmitted, are confirmed to threshold parameter initial value or threshold parameter are updated, and will be updated
Threshold parameter afterwards sends flow monitoring module to；The more current flow of flow monitoring module and historical traffic situation, judge
Whether network occurs integrated flux mutation, and flow histories data are saved as using present flow rate as historical traffic after the completion of judgement,
And will determine that result is sent to alarm display module；Alarm display module leads to according to the result of calculation of flow monitoring module transfer
Cross human-computer interaction interface to be shown, carry out alarm if occurring integrated flux mutation, do not occur, point out normal.
The method that network monitoring is carried out using a kind of network flow monitoring system of threshold adaptive amendment, as shown in Fig. 2
Comprise the following steps：
Step 1, flow collection module from interchanger obtain flow information data, be sent to adaptive threshold correcting module and
Flow monitoring module；
Step 2, adaptive threshold correcting module receives the flow information data that flow acquisition module is transmitted, to threshold parameter
Initial value is confirmed or threshold parameter is updated, and sends the threshold parameter after renewal to flow monitoring module；
Step 3, the more current flow of flow monitoring module and historical traffic situation, judge whether network occurs cumulative flow
Amount mutation, saves as flow histories data, and will determine that result is sent to after the completion of judgement using present flow rate as historical traffic
Alarm display module；
Step 4, alarm display module is carried out according to the result of calculation of flow monitoring module transfer by human-computer interaction interface
It has been shown that, carries out alarm if occurring integrated flux mutation, does not occur, and points out normal.Operating personnel are according to alarm condition
Alert data is checked, the situation for producing flow mutation is analyzed.
It is described to save as flow histories data, completed by assembling historical data storage method, as shown in figure 4, specific real
Now process is：To data flows of traffic x1,x2,...,xnIn element i (1≤i≤n) ask prefix and, be designated as F (i), i.e.,
If si=F (i), calculating obtains prefix and data flow s1,s2,...,sn, it is designated as first group of aggregation historical data；So
Afterwards, by data flow s1,s2,...,snIt is stored in the sliding window that length is n, as shown in Figure 3；Limit the length of data flow as
N, i.e. sliding window size are constant, as shown in figure 4, when new data arrives, existing s is attached to successively1,s2,...,sn
On, it can obtain new data stream s1+xn+1, s2+xn+1,...,sn+xn+1；Ensure that the size of window is constant, by xn+1It is used as the first number
According to while by data sn+xn+1Reject, obtain the data flow x of n bit lengthsn+1,s1+xn+1,...,sn-1+xn+1, it is designated as s '1,s
′2,...,s′n, similarly, as addition xn+2When obtain data flow xn+2,xn+1+xn+2,s1+xn+1+xn+2,...,sn-2+xn+1+
xn+2, it is designated as s "1,s″2,...,s″n；The like, update aggregation historical data successively by this method.
The threshold parameter is β, and the method to set up of its initial value is：J is traveled through from 1 to n/2, n/2 groups rate of change row can be arrived
Table, rate of change Cr(j) calculation formula is：
Cr(j)=(F (2j)-F (j))/F (j) (4)
To calculating the rate of change C completedr(j) average, obtain average rate of change Crm, using two times of this rate of change as
Initial threshold parameter β, i.e.,
Adaptive threshold correcting module is updated to threshold parameter to be completed by adaptive correction mode, due to adaptive
The threshold parameter that validation period is obtained is the adaptive threshold of current n data, with the continuous renewal of sliding window, threshold value ginseng
Number updates in the following ways：As data xn+1Enter after sliding window, βnewUpdate mode it is as shown in Figure 5：
, it is necessary to according to current data window, recalculate once after thering are new data to enter in data window
Cr(i) ' (i=1,2 ... n/2) value, according to the C calculatedr(i) ', average operation is once taken again to it, according to
(5) formula obtains new average value β ',
For the new threshold parameter weight α different with the imparting of history threshold parameter and 1- α, the threshold parameter after updating is drawn
βnew, it is as follows：
βnew=α * β+(1- α) * β ' (6)
In formula, α is weighting parameters, and generally α can use a value between 0.5-0.99；α value is dependent on cunning
The length n of dynamic window size, if window is larger, then α is also larger, if window is smaller, then α is also smaller, α is with sliding
The length n of window change is linear, and when window is 1, it is 0.5 to take α, and table 1 is α and n corresponding tables；
Table 1
| n | α |
| 1 | 0.5 |
| 4 | 0.58 |
| 8 | 0.67 |
| 16 | 0.75 |
| 32 | 0.82 |
| 64 | 0.90 |
| 128 | 0.99 |
| 1 | 0.5 |
| 4 | 0.58 |
β calculation formula is deployed, following threshold value more new formula is obtained：
It is described to judge whether network occurs integrated flux mutation, completed by abrupt climatic change ergodic algorithm, only on flow
The situation of liter is monitored, and does not detect that the process of implementing is to the unexpected decline situation of flow：
Using formula (1), judgement ascendant trend mutation is used as using inequality F (i) >=β (F (2i)-F (i)) (β ＞ 1)
Criterion；Formula deformation is obtained into following formula：
F (2i) >=(β+1) (F (2i)-F (i)) (β ＞ 1) (2)
I.e.
s2i≥(β+1)(s2i-si) (β ＞ 1) (3)
Utilize s2i≥(β+1)(s2i-si) (β ＞ 1) judged, judge to comprise the steps of
Step 1, for the sliding window that length is n, the algorithm collects data n in advance, i.e., sliding window data is filled up
Afterwards, step 2 is transferred to, the algorithm that brings into operation carries out abrupt climatic change；
Step 2, the algorithm travels through i from 1 to n/2, judges s2i≥(β+1)(s2i-si) whether (β ＞ 1) set up, such as set up,
Outputting alarm, shows have mutation to produce；Otherwise, do not point out, continue to run with algorithm until traversal is completed；After the completion of, such as data flow
New element is not reached, then is continued waiting for, and is otherwise transferred to step 3；
Step 3, when new element is reached in data flow, after being updated the data according to aggregation historical data storage method, again
The algorithm is run to carry out abrupt climatic change.
Above to embodiments of the invention to being explained in detail, above-mentioned embodiment is only the optimal implementation of the present invention
Example, but the present invention is not limited to above-described embodiment, in the knowledge that those of ordinary skill in the art possess, can be with
Various changes can be made on the premise of present inventive concept is not departed from.
Claims (4)
1. a kind of network flow monitoring system of threshold adaptive amendment, it is characterised in that including flow collection module, adaptively
Threshold values correcting module, flow monitoring module and alarm display module；Wherein, flow collection module obtains flow information from interchanger
Data, are sent to adaptive threshold correcting module and flow monitoring module；Adaptive threshold correcting module receives flow collection mould
The flow information data that block is transmitted, are confirmed to threshold parameter initial value or threshold parameter are updated, and by after renewal
Threshold parameter send flow monitoring module to；The more current flow of flow monitoring module and historical traffic situation, judge net
Whether network occurs integrated flux mutation, and flow histories data are saved as using present flow rate as historical traffic after the completion of judgement, and
It will determine that result is sent to alarm display module；Alarm display module passes through according to the result of calculation of flow monitoring module transfer
Human-computer interaction interface is shown, carries out alarm if occurring integrated flux mutation, does not occur, and is pointed out normal；
Flow monitoring module preserves flow histories data and is accomplished by the following way：To data flows of traffic x1,x2,...,xnIn
Element i, 1≤i≤n, seek prefix and are designated as F (i), i.e.,If si=F (i), calculating obtains prefix sum
According to stream s1,s2,...,sn, it is designated as first group of aggregation historical data；Then, by data flow s1,s2,...,snLength is stored in for n
Sliding window in；The length of data flow is limited as n, i.e. the size of sliding window is constant, when new data arrives, added successively
To existing s1,s2,...,snOn, it can obtain new data stream s1+xn+1, s2+xn+1,...,sn+xn+1；Ensure the size of window not
Become, by xn+1As first data, while by data sn+xn+1Reject, obtain the data flow x of n bit lengthsn+1,s1+xn+1,...,
sn-1+xn+1, it is designated as s '1,s′2,...,s′n, similarly, as addition xn+2When obtain data flow xn+2,xn+1+xn+2,s1+xn+1+
xn+2,...,sn-2+xn+1+xn+2, it is designated as s "1,s″2,...,s″n；The like, update aggregation history number successively by this way
According to；
Threshold parameter initial value is confirmed or threshold parameter is updated, is accomplished by the following way：If threshold parameter is β, then
Its initial value isIn formula, rate of change Cr(j)=(F (2j)-F (j))/F (j), F (j) are
Prefix and CrmFor the average rate of change；Its updated value isIn formula,
C'r (j) is the rate of change of updated value, and when updating first time, β is initial value, and when updating next time, β takes last round of formula to obtain
The β gone outnewValue, α is weighting parameters, and it works as window according to length of window n certain certain value of different sizes taken between 0.5-0.99
When mouth length is 1, it is 0.5 to take α.
2. a kind of network flow monitoring system of threshold adaptive amendment as claimed in claim 1, it is characterised in that judge net
Whether network occurs integrated flux mutation, is completed by abrupt climatic change ergodic algorithm, and the process of implementing is：
Step 1, for the sliding window that length is n, the algorithm collects data n in advance, i.e., after sliding window data is filled up, and turns
Enter step 2, the algorithm that brings into operation carries out abrupt climatic change；
Step 2, the algorithm travels through i from 1 to n/2, judges s2i≥(β+1)(s2i-si), β ＞ 1, if set up such as are set up, output
Alarm, shows have mutation to produce；Otherwise, do not point out, continue to run with algorithm until traversal is completed；After the completion of, such as data flow Singapore dollar
Element is not reached, then continues waiting for, be otherwise transferred in step 3, formula, s2iFor F (2i), siFor F (i)；
Step 3, when new element is reached in data flow, after being updated the data according to aggregation historical data storage method, rerun
The algorithm carries out abrupt climatic change.
3. a kind of network flow monitoring method of threshold adaptive amendment, it is characterised in that comprise the following steps：
Step 1, flow collection module obtains flow information data from interchanger, is sent to adaptive threshold correcting module and flow
Monitoring modular；
Step 2, adaptive threshold correcting module receives the flow information data that flow acquisition module is transmitted, initial to threshold parameter
Value is confirmed or threshold parameter is updated, and sends the threshold parameter after renewal to flow monitoring module；
Step 3, the more current flow of flow monitoring module and historical traffic situation, judge whether network occurs integrated flux and dash forward
Become, save as flow histories data using present flow rate as historical traffic after the completion of judgement, and will determine that result is sent to alarm
Display module；
Step 4, alarm display module is shown according to the result of calculation of flow monitoring module transfer by human-computer interaction interface
Show, carry out alarm if occurring integrated flux mutation, do not occur, point out normal；
The preservation of flow histories data is completed in step 3 by assembling historical data storage method, the process of implementing is：Convection current
Measure data flow x1,x2,...,xnIn element i, 1≤i≤n, ask prefix and, be designated as F (i), i.e.,If si=
F (i), calculating obtains prefix and data flow s1,s2,...,sn, it is designated as first group of aggregation historical data；Then, by data flow s1,
s2,...,snIt is stored in the sliding window that length is n；The length of data flow is limited as n, i.e. the size of sliding window is constant, when
When new data arrives, existing s is attached to successively1,s2,...,snOn, it can obtain new data stream s1+xn+1, s2+xn+1,...,sn+
xn+1；Ensure that the size of window is constant, by xn+1As first data, while by data sn+xn+1Reject, obtain the number of n bit lengths
According to stream xn+1,s1+xn+1,...,sn-1+xn+1, it is designated as s '1,s′2,...,s′n, similarly, as addition xn+2When obtain data flow
xn+2,xn+1+xn+2,s1+xn+1+xn+2,...,sn-2+xn+1+xn+2, it is designated as s "1,s″2,...,s″n；The like, update successively poly-
Collect historical data；
Threshold parameter initial value is confirmed or threshold parameter is updated in step 2, is accomplished by the following way：If threshold parameter
For β, then its initial value isIn formula, rate of change Cr(j)=(F (2j)-F (j))/F (j),
F (j) is prefix and CrmFor the average rate of change；Its updated value is
In formula, C'r (j) is the rate of change of updated value, and when updating first time, β is initial value, and when updating next time, β takes last round of public affairs
The β that formula is drawnnewValue, α is weighting parameters, its according to length of window n certain certain value of different sizes taken between 0.5-0.99,
When length of window is 1, it is 0.5 to take α.
4. a kind of network flow monitoring method of threshold adaptive amendment as claimed in claim 3, it is characterised in that in step 2
The threshold parameter is β, and its initial value is In formula, Cr(j)=(F (2j)-
F(j))/F(j)；Its updated value is Wherein, first
β is initial value during secondary renewal, and when updating next time, β takes the β that last round of formula is drawnnewValue, α is weighting parameters, and α can use
A value between 0.5-0.99；α value depends on the size of window, if window is larger, then α is also larger, if window
Mouth is smaller, then α is also smaller.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210401916.9A CN103780445B (en) | 2012-10-22 | 2012-10-22 | A kind of network flow monitoring system and method for threshold adaptive amendment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210401916.9A CN103780445B (en) | 2012-10-22 | 2012-10-22 | A kind of network flow monitoring system and method for threshold adaptive amendment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103780445A CN103780445A (en) | 2014-05-07 |
| CN103780445B true CN103780445B (en) | 2017-10-27 |
Family
ID=50572312
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210401916.9A Active CN103780445B (en) | 2012-10-22 | 2012-10-22 | A kind of network flow monitoring system and method for threshold adaptive amendment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103780445B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103986515B (en) * | 2014-05-09 | 2016-06-08 | 浙江中烟工业有限责任公司 | The performance indications monitoring method of fibre channel media |
| CN104967540B (en) * | 2014-06-10 | 2019-02-26 | 腾讯科技（深圳）有限公司 | Server state detection method and device |
| CN105681063A (en) * | 2014-11-18 | 2016-06-15 | 中国移动通信集团北京有限公司 | Method and apparatus for monitoring network index |
| CN105991362B (en) * | 2015-02-12 | 2019-10-29 | 腾讯科技（深圳）有限公司 | The fluctuation threshold range setting method and device of data traffic |
| CN104811828A (en) * | 2015-04-27 | 2015-07-29 | 无锡天脉聚源传媒科技有限公司 | Data processing method and device |
| CN107181601B (en) * | 2016-03-09 | 2019-12-06 | 中国移动通信集团湖南有限公司 | Flow reminding method and device |
| CN106656583A (en) * | 2016-12-02 | 2017-05-10 | 郑州云海信息技术有限公司 | Dynamic threshold alarming method and device |
| CN110351163B (en) * | 2019-06-05 | 2022-11-18 | 华南理工大学 | OpenStack cloud host traffic peak detection method |
| CN112203320B (en) * | 2019-07-08 | 2023-04-28 | 中国移动通信集团贵州有限公司 | Method and device for predicting target network parameters based on gray model |
| CN110381456B (en) * | 2019-07-19 | 2020-10-02 | 珠海格力电器股份有限公司 | Flow management system, flow threshold calculation method and air conditioning system |
| CN111817923B (en) * | 2020-07-28 | 2021-09-14 | 城云科技（中国）有限公司 | Early warning analysis method and device for sudden change of flow of switch port |
| CN113157505B (en) * | 2021-04-07 | 2022-10-18 | 苏州瑞立思科技有限公司 | Bandwidth self-adaptive abnormal flow detection method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1617512A (en) * | 2004-11-25 | 2005-05-18 | 中国科学院计算技术研究所 | Adaptive network flow forecasting and abnormal alarming method |
| CN101729301A (en) * | 2008-11-03 | 2010-06-09 | 中国移动通信集团湖北有限公司 | Monitor method and monitor system of network anomaly traffic |
| US8121024B1 (en) * | 1999-06-29 | 2012-02-21 | Cisco Technology, Inc. | Technique for providing dynamic modification of application specific policies in a feedback-based, adaptive data network |
-
2012
- 2012-10-22 CN CN201210401916.9A patent/CN103780445B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8121024B1 (en) * | 1999-06-29 | 2012-02-21 | Cisco Technology, Inc. | Technique for providing dynamic modification of application specific policies in a feedback-based, adaptive data network |
| CN1617512A (en) * | 2004-11-25 | 2005-05-18 | 中国科学院计算技术研究所 | Adaptive network flow forecasting and abnormal alarming method |
| CN101729301A (en) * | 2008-11-03 | 2010-06-09 | 中国移动通信集团湖北有限公司 | Monitor method and monitor system of network anomaly traffic |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103780445A (en) | 2014-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103780445B (en) | A kind of network flow monitoring system and method for threshold adaptive amendment | |
| WO2022228049A1 (en) | Method for diagnosing malfunction in aero-engine on basis of 5g edge computing and deep learning | |
| CN107634911B (en) | Adaptive congestion control method based on deep learning in information center network | |
| WO2020077672A1 (en) | Method and device for training service quality evaluation model | |
| CN108197845A (en) | A kind of monitoring method of the transaction Indexes Abnormality based on deep learning model LSTM | |
| CN109409567B (en) | Complex equipment residual life prediction method based on double-layer long-short term memory network | |
| CN109787846A (en) | A kind of 5G network service quality exception monitoring and prediction technique and system | |
| CN103702360B (en) | A kind of method and device of the data rate for determining service access port | |
| CN112073473B (en) | Internet of things equipment heartbeat packet data acquisition method | |
| WO2014183782A1 (en) | Method and network device for cell anomaly detection | |
| CN106856508A (en) | The cloud monitoring method and cloud platform of data center | |
| CN110535723A (en) | The message method for detecting abnormality of deep learning is used in a kind of SDN | |
| CN103888315A (en) | Self-adaptation burst flow detection device and detection method thereof | |
| CN116192888A (en) | Network state monitoring and management method and system based on Internet of things | |
| CN113762604B (en) | Industrial Internet big data service system | |
| CN105721194A (en) | Intelligent positioning system of faults and hidden dangers of mobile network | |
| CN105141446A (en) | Network equipment health degree assessment method determined based on objective weight | |
| CN107454009B (en) | Data center-oriented offline scene low-bandwidth overhead traffic scheduling scheme | |
| CN112787878B (en) | Network index prediction method and electronic equipment | |
| WO2023273837A1 (en) | Model training method and apparatus, traffic prediction method and apparatus, traffic load balancing method and apparatus, and storage medium | |
| CN113904948A (en) | 5G network bandwidth prediction system and method based on cross-layer multi-dimensional parameters | |
| CN107765617B (en) | Train axle temperature data processing method and device | |
| CN111614520B (en) | IDC flow data prediction method and device based on machine learning algorithm | |
| CN117092953A (en) | Production data acquisition management and control system based on industrial Internet of things | |
| CN116361377A (en) | Load prediction system, method and medium based on industrial Internet of things service platform |
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 |