CN107294810B - Network delay measurement optimization method under software defined network environment - Google Patents
Network delay measurement optimization method under software defined network environment Download PDFInfo
- Publication number
- CN107294810B CN107294810B CN201710575567.5A CN201710575567A CN107294810B CN 107294810 B CN107294810 B CN 107294810B CN 201710575567 A CN201710575567 A CN 201710575567A CN 107294810 B CN107294810 B CN 107294810B
- Authority
- CN
- China
- Prior art keywords
- delay
- time
- network
- measurement
- controller
- 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
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0852—Delays
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
Abstract
The invention relates to a network delay measurement optimization method, which comprises the steps of firstly forming a delay set X by using network delays measured for the last n times, then subtracting the average value of each element in the delay set X to form a difference sequence set Y, obtaining a fitting trigonometric function by using the elements in the set Y, and finally obtaining a time interval which is in direct proportion to the last measurement period and in inverse proportion to the variance of the elements in the set Y, wherein the time interval of the next measurement can be longer when the change of the network delay in the set X is smaller, and the time interval of the next measurement is shorter when the change of the network delay in the set X is larger, so that the accuracy of the network delay measurement can be ensured and improved, and the network resource consumption occupied by the measurement delay can be reduced.
Description
Technical Field
The present invention relates to a method for optimizing Network delay measurement, and more particularly, to a method for optimizing Network delay measurement in a Software Defined Network (SDN) environment.
Background
The internet has grown rapidly over the last twenty years and has become a vital part of the social infrastructure. The current internet runs on a core architecture designed over 20 years ago, which presents more and more problems with the rapid expansion of network size and the increasing abundance of application types. The continuous complication of the internet structure and functions causes the network management and control capability to be continuously weakened. As the functionality carried continues to expand (e.g., packet filtering, differentiated services, multicasting, quality of service, traffic engineering, etc.), routers that are the core of the network have become unwieldy. Due to the consideration of market share and the like, a switch or a router in the traditional network can only open a small number of service and management interfaces through modes such as a command line interface and the like, researchers are difficult to experiment and deploy a novel network system in a real network, and the method is helpful to the innovation of internet technology. In order to solve a plurality of problems faced by the existing internet system structure, the world countries have developed the future internet research on a large scale.
In 2008, McKeown, stanford university, usa, issued research results on OpenFlow on ACM sigcomp, an international top meeting. The OpenFlow allows a user to control network processing behaviors through an open flow table of the user, and is a novel network model supporting network innovation research. Based on the programmable nature brought by OpenFlow, the McKeown professor further proposes a Software Defined Network (SDN) concept. The main advantages of SDN technology include: forwarding control is separated, and hardware equipment is generalized; network resources are optimized in a centralized manner, and the utilization rate of the network resources is improved; network management is greatly simplified; the network innovation speed is accelerated, the line period of the new function is obviously shortened, and the like. In 2011, the international organization open Network foundation (onf) (open Network foundation) was established under the initiative of several companies such as yahoo, google, and german telecommunications, and is dedicated to standardization and commercialization of the SDN and its core technology, OpenFlow. In 2012, google announced that its backbone network has been fully operated on OpenFlow, and linked 12 data centers distributed around the world through a 10G network, so that the utilization rate of wide area lines is increased from 30% to near saturation. The success case of google proves the feasibility of the SDN in practice, and advantageously promotes the development of the SDN. In 2012, the MIT Technology Review magazine of international shared reputation selects OpenFlow as ten major future technologies; the international research institute Gartner predicts that the SDN technology will become one of ten key trends and technical impacts in the IT field in five years to come. In short years, SDN technology has recently raised new IT heat trends worldwide, causing extensive attention and research in academia and industry.
Under the software-defined network environment, network measurement is a basic condition for effective resource scheduling, and at present, related research is weak, and existing research cannot effectively meet low-load and high-accuracy measurement of a software-defined network, particularly a large-scale network. Especially, many applications benefit from the knowledge of network end-to-end delay, network delay is an important index of network state, and many software-defined network applications sensitive to quality of service require that network end-to-end delay meet certain constraint conditions, so that it has important practical significance to research network measurement and optimization methods for delay.
Currently, many software-defined network applications sensitive to the quality of service require that the end-to-end delay of the network meets a certain constraint condition, the existing network measurement method can only perform measurement regularly and uniformly, and if the existing network measurement method is more accurate, the measurement can only be achieved by increasing the measurement times and frequency, because the measurement itself also occupies network resources, the original network resources are greatly consumed. And thus actual network delay variation cannot be obtained more accurately. Aiming at the problem, the invention provides a method for automatically optimizing and adjusting the frequency and the period of network delay measurement according to the network delay change in a software defined network environment, which can ensure and improve the accuracy of the network delay measurement and reduce the network resource consumption occupied by the measurement delay.
Disclosure of Invention
In order to overcome the defects of the technical problems, the invention provides a network delay measurement optimization method under a software-defined network environment, which can ensure the network delay measurement accuracy and avoid the excessive consumption of network resources.
The invention discloses a network delay measurement optimization method under a software defined network environment, which is characterized by comprising the following steps of:
a) acquiring network delay data, measuring the network delay X of a link between two adjacent switches once at fixed time t by a controller under the environment of a software defined network, and measuring n times to obtain a group of delay sets X (X)1,x2,...,xn),xiI is more than or equal to 1 and less than or equal to n for the network delay obtained by the ith measurement;
b) the fitting of a trigonometric function,will delay set X (X)1,x2,...,xn) The delay data in (2) is subjected to trigonometric function fitting, which is realized by the following steps:
b-1) calculating an average value, and calculating an average value E of network delays in the delay set X through a formula (1):
in the formula: n is the number of network delays in delay set X;
b-2) calculating a difference sequence, calculating the difference between each network delay in the delay set X and the average value E to obtain a delay difference sequence Y (Y)1,y2,...,yn),y1=x1-E,y2=x2-E,…,yn=xn-E;
b-3) solving the variance of network delay, and calculating the variance sigma of the delay set X by formula (2)2:
Wherein i is more than or equal to 1 and less than or equal to n;
b-4), obtaining a fitting trigonometric function, substituting the value in the solved delay difference value sequence Y into a formula (2) to obtain the fitting trigonometric function:
wherein tau is the latest measuring period and the initial value is t; z is time;
c) calculating a measurement interval, and obtaining a time interval of next network delay measurement by fitting a trigonometric function, wherein the method is realized by the following steps:
c-1) determining the position of the peak or trough, taking the derivative of the fitting trigonometric function F (z), and lettingCalculating the nearest derivative asTime t of 00Then the time position of the most recent peak or trough of the trigonometric function F (z) is t0;
c-2), obtaining a measuring time point, and calculating the time point of the next measurement through a formula (5):
wherein, tn+1Time point for next measurement, t0Fitting a trigonometric function F (z) at the moment of the last peak or trough, wherein tau is the last measurement period and k is the integer of the multiple of tau contained in the total time of the X sequence; n is the number of network delays contained in the X sequence;
c-3) obtaining a measurement interval, and solving a time interval tau' of the next measurement through a formula (6):
tnthe measurement time of the nth time in the X sequence is shown;
d) measuring the delay and updating the sequence, according to the time interval τ', t found in step c)nAfter the time period of tau' is timed at the beginning, the method in the step a) is adopted to obtain the network delays of the two exchangers, then the measured network delays are added into a delay set X, and the first network delay in the X is deleted to obtain a new delay set X; delay values of the latest n measurements in X;
e) if the delay of the corresponding link is not wanted to be measured, stopping the measurement operation; and if the network delay of the corresponding link is continuously measured, returning to execute the step b) by using the new delay set X obtained in the step d), and repeating the steps b), c) and d).
The invention relates to a network delay measurement optimization method under a software defined network environment, wherein the network delay data acquisition in the step a) is realized by the following steps:
a-1) obtaining time of a switch, Software Defined Network (SDN) controllerSending a measurement data packet to one of the switches si, and recording the sending time; the switch si returns to the controller immediately after receiving the data packet, the receiving time is recorded after the switch si controls the data packet to be received, and the SDN controller records the difference between the receiving time and the sending time as TController to si time;
a-2), acquiring the time of another switch, and sending a measurement data packet to another switch sj by the SDN controller and recording the sending time; the switch sj returns to the controller immediately after receiving the data packet, the receiving time is recorded after the switch sj controls the data packet to be received, and the SDN controller records the difference between the receiving time and the sending time as TController to sj time;
a-3), acquiring total time, sending a measurement data packet to a switch si by an SDN controller, and recording sending time; meanwhile, the controller issues a routing flow table, so that the switch si forwards the data packet to sj when receiving the data packet, and the switch sj returns the message to the controller immediately after receiving the message; after receiving the message, the controller records the receiving time; the controller calculates the time difference between the sending and receiving of the data packet, which is recorded as TTotal time of day;
a-4) calculating delay time, and enabling the SDN controller to pass through a formulaAnd calculating the network delay.
According to the method for optimizing the network delay measurement in the software-defined network environment, the number n of delay time contained in the delay set X is between 10 and 30.
The invention has the beneficial effects that: the invention relates to a network delay measurement optimization method, which comprises the steps of firstly forming a delay set X by using network delays measured for the last n times, then subtracting the average value of each element in the delay set X to form a difference sequence set Y, obtaining a fitting trigonometric function by using the elements in the set Y, and finally obtaining a time interval which is in direct proportion to the last measurement period and in inverse proportion to the variance of the elements in the set Y, wherein the time interval of the next measurement can be longer when the change of the network delay in the set X is smaller, and the time interval of the next measurement is shorter when the change of the network delay in the set X is larger, so that the accuracy of the network delay measurement can be ensured and improved, and the network resource consumption occupied by the measurement delay can be reduced.
Drawings
FIG. 1 is a flow chart of a network delay measurement optimization method in a software defined networking environment according to the present invention;
fig. 2 is a flow chart of acquiring network delay data according to the present invention.
Detailed Description
The invention is further described with reference to the following figures and examples.
As shown in fig. 1, a flowchart of a network delay measurement optimization method in a software-defined network environment according to the present invention is provided, which is implemented by the following steps:
a) acquiring network delay data, measuring the network delay X of a link between two adjacent switches once at fixed time t by a controller under the environment of a software defined network, and measuring n times to obtain a group of delay sets X (X)1,x2,...,xn),xiI is more than or equal to 1 and less than or equal to n for the network delay obtained by the ith measurement;
as shown in fig. 2, a flow chart of the present invention for obtaining network delay data is provided, which is implemented by the following steps:
a-1), acquiring the time of one switch, and sending a measurement data packet to one switch si by a Software Defined Network (SDN) controller and recording the sending time; the switch si returns to the controller immediately after receiving the data packet, the receiving time is recorded after the switch si controls the data packet to be received, and the SDN controller records the difference between the receiving time and the sending time as TController to si time;
a-2), acquiring the time of another switch, and sending a measurement data packet to another switch sj by the SDN controller and recording the sending time; the switch sj returns to the controller immediately after receiving the data packet, the receiving time is recorded after the switch sj controls the data packet to be received, and the SDN controller records the difference between the receiving time and the sending time as TController to sj time;
a-3) obtaining the total timeThe SDN controller sends a measurement data packet to the switch si and records sending time; meanwhile, the controller issues a routing flow table, so that the switch si forwards the data packet to sj when receiving the data packet, and the switch sj returns the message to the controller immediately after receiving the message; after receiving the message, the controller records the receiving time; the controller calculates the time difference between the sending and receiving of the data packet, which is recorded as TTotal time of day;
a-4) calculating delay time, and enabling the SDN controller to pass through a formulaAnd calculating the network delay.
b) Trigonometric function fitting, set of delays X (X)1,x2,...,xn) The delay data in (2) is subjected to trigonometric function fitting, which is realized by the following steps:
b-1) calculating an average value, and calculating an average value E of network delays in the delay set X through a formula (1):
in the formula: n is the number of network delays in delay set X;
b-2) calculating a difference sequence, calculating the difference between each network delay in the delay set X and the average value E to obtain a delay difference sequence Y (Y)1,y2,...,yn),y1=x1-E,y2=x2-E,…,yn=xn-E;
b-3) solving the variance of network delay, and calculating the variance sigma of the delay set X by formula (2)2:
Wherein i is more than or equal to 1 and less than or equal to n;
b-4), obtaining a fitting trigonometric function, substituting the value in the solved delay difference value sequence Y into a formula (2) to obtain the fitting trigonometric function:
wherein tau is the latest measuring period and the initial value is t; z is time;
c) calculating a measurement interval, and obtaining a time interval of next network delay measurement by fitting a trigonometric function, wherein the method is realized by the following steps:
c-1) determining the position of the peak or trough, taking the derivative of the fitting trigonometric function F (z), and lettingFinding the time t at which the last derivative is 00Then the time position of the most recent peak or trough of the trigonometric function F (z) is t0;
c-2), obtaining a measuring time point, and calculating the time point of the next measurement through a formula (5):
wherein, tn+1Time point for next measurement, t0Fitting a trigonometric function F (z) at the moment of the last peak or trough, wherein tau is the last measurement period and k is the integer of the multiple of tau contained in the total time of the X sequence; n is the number of network delays contained in the X sequence;
c-3) obtaining a measurement interval, and solving a time interval tau' of the next measurement through a formula (6):
tnthe measurement time of the nth time in the X sequence is shown;
d) measuring the delay and updating the sequence, according to the time interval τ', t found in step c)nAfter the time period of tau' is timed from the beginning of the time, the method in step a) is adopted to obtain the network delay of the two exchangers, and then the measured network delay is usedAdding the network delay into a delay set X, and deleting the first network delay in the X to obtain a new delay set X; delay values of the latest n measurements in X;
e) if the delay of the corresponding link is not wanted to be measured, stopping the measurement operation; and if the network delay of the corresponding link is continuously measured, returning to execute the step b) by using the new delay set X obtained in the step d), and repeating the steps b), c) and d).
Claims (3)
1. A network delay measurement optimization method in a software-defined network environment is characterized by being realized by the following steps:
a) acquiring network delay data, measuring the network delay X of a link between two adjacent switches once at fixed time t by a controller under the environment of a software defined network, and measuring n times to obtain a group of delay sets X (X)1,x2,...,xn),xiI is more than or equal to 1 and less than or equal to n for the network delay obtained by the ith measurement;
b) trigonometric function fitting, set of delays X (X)1,x2,...,xn) The delay data in (2) is subjected to trigonometric function fitting, which is realized by the following steps:
b-1) calculating an average value, and calculating an average value E of network delays in the delay set X through a formula (1):
in the formula: n is the number of network delays in delay set X;
b-2) calculating a difference sequence, calculating the difference between each network delay in the delay set X and the average value E to obtain a delay difference sequence Y (Y)1,y2,...,yn),y1=x1-E,y2=x2-E,…,yn=xn-E;
b-3) solving the variance of network delay, and calculating the variance sigma of the delay set X by formula (2)2:
Wherein i is more than or equal to 1 and less than or equal to n;
b-4), obtaining a fitting trigonometric function, substituting the value in the solved delay difference value sequence Y into a formula (2) to obtain the fitting trigonometric function:
wherein tau is the latest measuring period and the initial value is t; z is time;
c) calculating a measurement interval, and obtaining a time interval of next network delay measurement by fitting a trigonometric function, wherein the method is realized by the following steps:
c-1) determining the position of the peak or trough, taking the derivative of the fitting trigonometric function F (z), and lettingFinding the time t at which the last derivative is 00Then the time position of the most recent peak or trough of the trigonometric function F (z) is t0;
c-2), obtaining a measuring time point, and calculating the time point of the next measurement through a formula (5):
wherein, tn+1Time point for next measurement, t0Fitting a trigonometric function F (z) at the moment of the last peak or trough, wherein tau is the last measurement period and k is the integer of the multiple of tau contained in the total time of the X sequence; n is the number of network delays contained in the X sequence;
c-3) obtaining a measurement interval, and solving a time interval tau' of the next measurement through a formula (6):
tnthe measurement time of the nth time in the X sequence is shown;
d) measuring the delay and updating the sequence, according to the time interval τ', t found in step c)nAfter the time period of tau' is timed at the beginning, the method in the step a) is adopted to obtain the network delays of the two exchangers, then the measured network delays are added into a delay set X, and the first network delay in the X is deleted to obtain a new delay set X; delay values of the latest n measurements in X;
e) if the delay of the corresponding link is not wanted to be measured, stopping the measurement operation; and if the network delay of the corresponding link is continuously measured, returning to execute the step b) by using the new delay set X obtained in the step d), and repeating the steps b), c) and d).
2. The method for optimizing network delay measurement in a software-defined networking environment according to claim 1, wherein the acquiring of the network delay data in step a) is specifically implemented by the following steps:
a-1), acquiring the time of one switch, and sending a measurement data packet to one switch si by a Software Defined Network (SDN) controller and recording the sending time; the switch si returns to the controller immediately after receiving the data packet, the receiving time is recorded after the switch si controls the data packet to be received, and the SDN controller records the difference between the receiving time and the sending time as TController to si time;
a-2), acquiring the time of another switch, and sending a measurement data packet to another switch sj by the SDN controller and recording the sending time; the switch sj returns to the controller immediately after receiving the data packet, the receiving time is recorded after the switch sj controls the data packet to be received, and the SDN controller records the difference between the receiving time and the sending time as TController to sj time;
a-3), acquiring total time, sending a measurement data packet to a switch si by an SDN controller, and recording sending time; meanwhile, the controller issues a routing flow table, so that when the switch si receives the data packet, the data packet is forwarded to sj, and the switch sj receives the data packetReturning to the controller immediately after the rest; after receiving the message, the controller records the receiving time; the controller calculates the time difference between the sending and receiving of the data packet, which is recorded as TTotal time of day;
a-4) calculating delay time, and enabling the SDN controller to pass through a formulaAnd calculating the network delay.
3. The method for optimizing network delay measurement in a software-defined networking environment according to claim 1 or 2, wherein: the number n of delay times included in the delay set X is between 10 and 30.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710575567.5A CN107294810B (en) | 2017-07-14 | 2017-07-14 | Network delay measurement optimization method under software defined network environment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710575567.5A CN107294810B (en) | 2017-07-14 | 2017-07-14 | Network delay measurement optimization method under software defined network environment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107294810A CN107294810A (en) | 2017-10-24 |
CN107294810B true CN107294810B (en) | 2020-01-17 |
Family
ID=60101821
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710575567.5A Active CN107294810B (en) | 2017-07-14 | 2017-07-14 | Network delay measurement optimization method under software defined network environment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107294810B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101335975A (en) * | 2007-06-28 | 2008-12-31 | 华为技术有限公司 | Method, apparatus and system for measurement interval control |
CN104486159A (en) * | 2014-12-18 | 2015-04-01 | 东南大学 | Inter-switch link time delay measuring method based on software defined network |
CN105357046A (en) * | 2015-11-23 | 2016-02-24 | 北京邮电大学 | Network information detection method for software defined networking (SDN) |
CN106230722A (en) * | 2016-08-05 | 2016-12-14 | 山东省计算中心(国家超级计算济南中心) | SDN congestion link method of adjustment based on transfer value |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9503344B2 (en) * | 2014-07-25 | 2016-11-22 | Telefonaktiebolaget L M Ericsson (Publ) | Data path performance measurement using network traffic in a software defined network |
US9667518B2 (en) * | 2015-09-11 | 2017-05-30 | Telefonaktiebolaget L M Ericsson (Publ) | Method and system for delay measurement of a traffic flow in a software-defined networking (SDN) system |
-
2017
- 2017-07-14 CN CN201710575567.5A patent/CN107294810B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101335975A (en) * | 2007-06-28 | 2008-12-31 | 华为技术有限公司 | Method, apparatus and system for measurement interval control |
CN104486159A (en) * | 2014-12-18 | 2015-04-01 | 东南大学 | Inter-switch link time delay measuring method based on software defined network |
CN105357046A (en) * | 2015-11-23 | 2016-02-24 | 北京邮电大学 | Network information detection method for software defined networking (SDN) |
CN106230722A (en) * | 2016-08-05 | 2016-12-14 | 山东省计算中心(国家超级计算济南中心) | SDN congestion link method of adjustment based on transfer value |
Also Published As
Publication number | Publication date |
---|---|
CN107294810A (en) | 2017-10-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108600102B (en) | Flexible data transmission system based on intelligent cooperative network | |
CN111585683B (en) | High-reliability clock synchronization system and method for time-sensitive network | |
CN111130928B (en) | Network measurement method based on in-band detection in wide area network | |
CN107682191B (en) | Power dispatching network information centralized management and control method and system | |
CN115632939B (en) | Automatic network selection and routing method for achieving multi-target achievement of computational power network | |
CN106027288A (en) | Communication traffic prediction method for distribution line information monitoring service | |
Mohammed et al. | Modified re-iterated Kalman filter for handling delayed and lost measurements in power system state estimation | |
CN103873364A (en) | Inter-domain multi-path rooting implementation method | |
CN109768904A (en) | A kind of time trigger business vanning dispatching method based on time trigger Ethernet | |
CN115314355A (en) | Electric power communication network architecture system and method based on deterministic network | |
CN107634813B (en) | The complete trails method for synchronizing time of software definition in information centre's smart grid | |
CN114465945A (en) | Identification analysis network construction method based on SDN | |
Huang et al. | Towards network-wide scheduling for cyclic traffic in ip-based deterministic networks | |
CN114827195A (en) | Electric power 5G low-delay jitter implementation method for distribution network stability protection | |
CN112153153A (en) | Coordinated distributed intra-network resource scheduling method and system and storage medium | |
CN107294810B (en) | Network delay measurement optimization method under software defined network environment | |
CN114070448A (en) | Master clock selection method based on multi-attribute decision | |
CN115022182A (en) | QSILP algorithm-based train communication network real-time flow scheduling optimization method | |
CN104954477B (en) | One kind is based on concurrent improved large-scale graph data streaming division methods and system | |
Hu et al. | Optimal status update in IoT systems: An age of information violation probability perspective | |
Bilen et al. | Handover-aware content replication for mobile-cdn | |
Quaglia et al. | Grain sensitive event scheduling in time warp parallel discrete event simulation | |
CN110324837A (en) | A kind of network management and device | |
Huang et al. | Time-sensitive network technology and its application in energy internet | |
CN114205309A (en) | Frame sending time accurate calculation method facing TSN output interface |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |