CN109088795A - A kind of controller performance analysis method based on systematic sampling - Google Patents
A kind of controller performance analysis method based on systematic sampling Download PDFInfo
- Publication number
- CN109088795A CN109088795A CN201811079839.3A CN201811079839A CN109088795A CN 109088795 A CN109088795 A CN 109088795A CN 201811079839 A CN201811079839 A CN 201811079839A CN 109088795 A CN109088795 A CN 109088795A
- Authority
- CN
- China
- Prior art keywords
- interchanger
- network
- controller
- time
- data flow
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
- H04L43/02—Capturing of monitoring data
- H04L43/022—Capturing of monitoring data by sampling
-
- 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
- H04L43/02—Capturing of monitoring data
- H04L43/026—Capturing of monitoring data using flow identification
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Environmental & Geological Engineering (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The controller performance analysis method based on systematic sampling that the invention discloses a kind of, belongs to future network technical field.The present invention can comprehensively analyze the performance of more different controllers with system.Specific step is as follows: step 1: netinit, inputs relevant parameter, selects source node and destination node;Step 2: utilizing interchanger selection algorithm, source interchanger and purpose interchanger is selected, to measure its traffic flow information to calculate the performance indicator of network;Step 3: whithin a period of time, controller regularly sends data stream statistics and requests to source interchanger and purpose interchanger;Step 4: the information of the corresponding data flow of data stream statistics request in measuring process three;Step 5: calculating network performance index, such as calculate network packet loss rate, network delay, handling capacity, delay variation and measurement error, measures performance characteristic of the controller when data packet occurs for the first time comprehensively by calculating the above network performance index.
Description
Technical field
The invention belongs to future network technical fields, more particularly to a kind of controller performance based on systematic sampling point
Analysis method.
Background technique
Currently, with the continuous expansion of network size, traditional network architecture is difficult to meet current operator, enterprise and use
The needs at family.In order to effectively be managed complex network, possessing data plane and control the software of plane phase separation characteristic
It defines network (Software Defined Networks, SDN) framework to be gradually accepted, this feature can be effectively
Simplify network management.In SDN normal form, controlling plane has obtained data plane greatly with the feature that data plane mutually separates
Simplify, complicated treatment process and calculating process all transfers to control plane.As the core component of control plane, controller can be with
Regard an important foundation stone as because it can according to the strategy of network operator and software developer, to network flow into
The control management of row centralization.As can be seen that the performance of controller decides the scalability of software defined network, this will be to software
The framework for defining network plays a very important role.
Currently, the realization of controller becomes varied, and industry and academia develop different types of control in succession
Device, the realization of these controllers have used different programming language and function collection, also, from initial prototype to controller
There is sizable difference.It is current so how accurately, efficiently assessing the controller performance in software defined network
The significant challenge that the field SDN is faced.
SDN is a kind of novel network architecture system.The framework is characterized in data plane and control planar separation, and
It can be directly controlled by program.In existing network, the control of flow and forwarding fully rely on the network equipment and (such as exchange
Machine, router) it carries out, and SDN separates control function from the network equipment, and the network-control concentrated.Number
It include forwarded unit (interchanger and router) according to plane, control plane includes controller.Controller is turned using control-
It sends out communication interface and centerized fusion is carried out to the forwarded unit on data plane, and flexible programmability is provided, and
In existing network, this is not present.Therefore, SDN changes the existing network architecture completely, and controller is as SDN framework
An element, it will be helpful to the development of SDN and it is necessary to existing SDN controller carry out assessment with than
Compared with.But for so numerous SDN controller, there is presently no systematically, comprehensively analysis method.It is enough due to lacking
Information, researcher simply by virtue of experience or secondary source carrys out selection control to complete the research of oneself, however this lacks
The method of weary strong foundation, prevent SDN controller resource is from making full use of.
Summary of the invention
For disadvantage of the existing technology, provide more fully, the measurement method of system carry out control to current mainstream
Device carries out systematically analysis and compares, and the present invention provides a kind of controller performance analysis method based on systematic sampling, this method
It can be with system, comprehensively quantitative analysis controller performance.
Technical solution of the present invention is a kind of controller performance analysis method based on systematic sampling, this method comprises:
Step 1: netinit
Netinit, and source node and destination node are randomly choosed from network topology;
Step 2: selection interchanger
Using interchanger selection algorithm, source interchanger and purpose interchanger are selected, the source interchanger and mesh selected are measured
Exchange data stream information and computing controller performance indicator;
Step 3: sending data stream statistics request
Whithin a period of time, controller regularly sends stream statistics and requests to source interchanger and purpose interchanger;
Step 4: measurement data stream information
The corresponding stream information of the data stream statistics request that measuring process 3 is sent;
Step 5: calculating network performance index
Network packet loss rate, network delay, handling capacity, delay variation and measurement error are calculated, by calculating the above internetworking
It can the next performance characteristic for measuring controller comprehensively when data packet occurs for the first time of index.
Further, the netinit in the step 1, particular content include: controller, the non-side for connecting controller
Edge interchanger, the edge switch for connecting controller connect the currently transmitted active flow of host, the host of edge switch, source
Node, destination node, the measurement period of interchanger, the time of measuring upper limit of source interchanger and purpose inter-exchange.
Further, the selection interchanger process of the step 2, detailed process is as follows:
Entire supervised path is traversed, if edge switch is source node and is not destination node, selects this exchange
Machine is as source interchanger;If edge switch is destination node and is not source node, this interchanger is selected as a purpose
Interchanger.
Further, data stream statistics request is sent in the step 3, detailed process is as follows: at the appointed time, from working as
Preceding measurement moment, at regular intervals, controller give source interchanger and purpose interchanger to send stream statistics request respectively.
Further, measurement data stream information in the step 4, detailed process is as follows:
It by stream matching rule and matching algorithm, is requested according to the stream statistics that controller in step 3 is sent, interchanger crawl
The data that data flow corresponding to the stream statistics request of the interchanger sent with controller in step 3 matches are flowed through in network
Stream, obtains the statistical information of the data flow and the metrical information is returned to controller.
Further, calculating network performance index described in step 5, detailed process is as follows:
The traffic flow information surveyed according to step 4 calculates packet loss, time delay, link utilization, delay variation, and passes through
It calculates the above network performance index and to measure performance characteristic of the controller when data packet occurs for the first time comprehensively.
Specific calculation formula is as follows:
(1) packet loss li:
Cfirst=Cfirst_j-Cfirst_(j-i) (1)
Cfinal=Cfinal_j-Cfinal_(j-i) (2)
Li=Cfirst-Cfinal (3)
Wherein, Cfirst_jIndicate the packet number of the data flow of counter in moment j record First interchanger;Cfirst_(j-i)
Indicate the packet number of the data flow of counter in moment j-i record First interchanger;Cfinal_jIndicate that moment j records last
The packet number of the data flow of counter in platform interchanger;Cfinal_(j-i)It indicates that moment j-i is recorded to count in last interchanger
The packet number of the data flow of device;LiIt indicates in measurement period i, counter in First interchanger and last interchanger
The packet number difference of incremental data stream;
(2) network delay tdelay:
T1=Tstart_1-Tleave_2 (5)
T2=Tstart_2-Tleave_1 (6)
Wherein, T1Indicate the time difference of two switch processes data flows;Tstart_1Indicate data flow by handing in network
It changes planes at the beginning of 1;Tleave_2Indicate that data flow leaves the time of interchanger 2 in network;T2It indicates at two switch processes
The time difference of reason machine;Tstart_2At the beginning of indicating data flow by interchanger 2 in network;Tleave_1Indicate that data are wandered about as a refugee
Open the time of interchanger 1 in network;Tc-s1And Tc-s2Respectively indicate controller to interchanger 1 and interchanger 2 two-way time;
(3) handling capacity Ui:
Wherein, Cfirst_avgIndicate the packet number of the average data stream of counter in First interchanger in measurement period i
Mesh;Cfinal_avgIndicate the packet number of the average data stream of counter in last interchanger in measurement period i;UfirstGeneration
Table passes through the handling capacity of First interchanger in measurement period i;UfinalIt represents in measurement period i, is handed over by last
The handling capacity changed planes;UiIt represents in measurement period i, passes through the average throughput of First interchanger and the last one interchanger
Amount;
(4) delay variation Jitter:
Wherein, n represents the time-delay series number of measurement network, td-jRepresent j-th of time-delay series when measurement network, Avg
(td-j) represent the time-delay series average value for measuring network;
(5) measurement error RMSE (X):
Wherein, xiFor the streaming rate value of i-th systematic sampling,The true of streaming rate is sent for i-th host
Real value, n are the number of sampling.
Beneficial effects of the present invention:
The present invention proposes a kind of controller performance analysis method based on systematic sampling.It is calculated herein using interchanger selection
Method and systematic sampling method, SDN controller (POX controller, FloodLight controller, RYU control to current mainstream
Device, ONOS controller and OpenDayLight controller) systematically analyze and compares, more comprehensive, system survey is provided
It is special to measure performance of the controller when data packet occurs for the first time by calculating the above network performance index comprehensively for amount method
Sign, and quantitative analysis is carried out to the performance indicator of controller, to provide effective research side for the researcher in future
Case.As can be seen that the present invention merely with systematic sampling method can more each controller performance, this is exactly of the invention
Where the advantage of analysis method.
Importantly, the present invention is to carry out quantitative analysis to the performance indicator of controller.It takes in the measurements at equal intervals
The Method of Stability Analysis of sampling, the data flow measurement error not only obtained is almost the same, illustrates measurement method to the property of controller
Can analyze does not influence, and also ensures the measurement accuracy of data flow in network.And the experimental results showed that, using at equal intervals
The controller performance index of methods of sampling measurement does not influence the performance of analyzer-controller substantially, while also ensuring measurement
Accuracy.
Detailed description of the invention
Fig. 1 is the SDN measurement mechanism that the present invention uses;
Fig. 2 is a plurality of flow measurement mechanism of SDN that the present invention uses;
Fig. 3 is network delay of the present invention composition;
Fig. 4 is the Linear Network topology of present example application;
Fig. 5 is the reticular pattern network topology of present example application;
Fig. 6 is present example simulation result;
It (a) is the average packet loss ratio of network topology 1
It (b) is the average packet loss ratio of network topology 2
It (c) is the average delay of network topology 1
It (d) is the average delay of network topology 2
It (e) is the average throughput of network topology 1
It (f) is the average throughput of network topology 2
(g) it is shaken for the average delay of network topology 1
(h) it is shaken for the average delay of network topology 2
It (i) is root-mean-square error.
Specific embodiment
Network topology structure is established using Mininet emulator, and is connected with controller.Two kinds of network simulations are realized herein
Topological structure: Linear Network topology and mesh network topologies, it is as shown in Figure 4 and Figure 5 respectively;Pass through the network topology knot of foundation
Structure carries out analysis comparison to controller performance index using method proposed in this paper.Controller performance based on systematic sampling
Analysis method, the specific steps are as follows:
Step 1: netinit
Netinit randomly chooses source node and destination node from network topology;In embodiments of the present invention, it applies
Two different topological structures.Wherein, Fig. 4 is simple linear network topology structure, which is controlled by 1
Device, 2 interchangers and 4 host compositions, ultimately form the bus-network of 2 network nodes and 4 host nodes;Fig. 5 is
Reticular pattern network topology structure, the network topology structure are made of 1 controller, 6 interchangers and 6 hosts, each exchange
It is connected with each other between machine, ultimately forms the reticular pattern network of 6 network nodes and 6 host nodes.
Step 2: selection interchanger
Corresponding source node and destination node interchanger are selected, i.e., the First interchanger mentioned in the method for the present invention and most
Latter interchanger selects S1 and S2 as First interchanger and last exchange in linear topology as shown in Figure 4
Machine;Meanwhile in reticular pattern network topology structure as shown in Figure 5, select S1 and S6 as First interchanger and last
Interchanger, and measure related data stream information;
Step 3: data stream statistics request is sent
Whithin a period of time, controller regularly sends stream statistics and requests to source interchanger and purpose interchanger.For linear
Topology is as shown in figure 4, only count the data flow that IP address is host h1 in S1 and S2 interchanger;For netted net topology such as Fig. 5
It is shown, the data flow that IP address is host h1 in S1 and S6 interchanger is only counted, this is also method requirement of the invention.
Step 4: measurement data stream information
In the previous step, controller regularly sends stream statistics and requests to source interchanger and purpose interchanger, that is, Fig. 4
In S1 and S2 interchanger, S1 the and S6 interchanger in Fig. 5;Interchanger passes through stream after the stream statistics request for receiving controller
Matching rule and matching algorithm, are collected respective streams and processing for statistical analysis, and the statistical result of respective streams is sent
To controller, controller collects the traffic statistics from each measurement interchanger.
Step 5: network performance index is calculated
Network packet loss rate, network delay, handling capacity, delay variation and measurement error are calculated, by calculating the above internetworking
It can the next performance characteristic for measuring controller comprehensively when data packet occurs for the first time of index.Its result such as Fig. 6 is shown.
Claims (6)
1. a kind of controller performance analysis method based on systematic sampling, the specific steps are as follows:
Step 1: netinit
Netinit, and source node and destination node are randomly choosed from network topology;
Step 2: selection interchanger
Using interchanger selection algorithm, source interchanger and purpose interchanger are selected, to measure the source interchanger and mesh selected
Exchange data stream information and calculate the performance indicator of network;
Step 3: sending data stream statistics request
Whithin a period of time, controller regularly sends stream statistics and requests to source interchanger and purpose interchanger;
Step 4: measurement data stream information
The information of the corresponding data flow of data stream statistics request in measuring process 3;
Step 5: calculating network performance index
Network packet loss rate, network delay, handling capacity, delay variation and measurement error are calculated, is referred to by calculating the above network performance
Mark measures performance characteristic of the controller when data packet occurs for the first time comprehensively.
2. a kind of controller performance analysis method based on systematic sampling according to claim 1, it is characterised in that institute
The netinit in step 1 is stated, particular content includes: controller, the non-edge interchanger for connecting controller, connection controller
Edge switch, connect edge switch the currently transmitted active flow of host, host, source node, destination node, interchanger
Measurement period, the time of measuring upper limit of source interchanger and purpose inter-exchange.
3. a kind of controller performance analysis method based on systematic sampling according to claim 1, it is characterised in that institute
The selection interchanger of step 2 is stated, detailed process is as follows:
Entire supervised path is traversed, if edge switch is source node and is not destination node, this interchanger is selected to make
For source interchanger;If edge switch is destination node and is not source node, this interchanger is selected to exchange as a purpose
Machine.
4. a kind of controller performance analysis method based on systematic sampling according to claim 1, it is characterised in that institute
It states and sends data stream statistics request in step 3, detailed process is as follows: at the appointed time, since the currently measurement moment, every
For a period of time, controller gives source interchanger and purpose interchanger to send stream statistics request respectively.
5. a kind of controller performance parser based on systematic sampling according to claim 1, it is characterised in that institute
Measurement data stream information in step 4 is stated, detailed process is as follows:
By stream matching rule and matching algorithm, requested according to the stream statistics that controller in step 3 is sent, interchanger grabs network
In flow through the data flow that the corresponding data flow of the stream statistics request sent with controller in step 3 of the interchanger matches,
It obtains the statistical information of the data flow and the metrical information is returned to controller.
6. a kind of controller performance parser based on systematic sampling according to claim 1, it is characterised in that step
Calculating network performance index described in rapid 5, detailed process is as follows:
The traffic flow information surveyed according to step 4 calculates packet loss, time delay, link utilization, delay variation, and passes through calculating
The above network performance index measures performance characteristic of the controller when data packet occurs for the first time comprehensively.
Specific calculation formula is as follows:
(1) packet loss li:
Cfirst=Cfirst_j-Cfirst_(j-i) (1)
Cfinal=Cfinal_j-Cfinal_(j-i) (2)
Li=Cfirst-Cfinal (3)
Wherein, Cfirst_jIndicate the packet number of the data flow of counter in moment j record First interchanger;Cfirst_(j-i)It indicates
Moment j-i records the packet number of the data flow of counter in First interchanger;Cfinal_jIndicate that moment j records last friendship
Change planes middle counter data flow packet number;Cfinal_(j-i)Indicate that moment j-i records counter in last interchanger
The packet number of data flow;LiIt indicates in measurement period i, the increment of counter in First interchanger and last interchanger
The packet number difference of data flow;
(2) network delay tdelay:
T1=Tstart_1-Tleave_2 (5)
T2=Tstart_2-Tleave_1 (6)
Wherein, T1Indicate the time difference of two switch processes data flows;Tstart_1Indicate data flow by interchanger in network
At the beginning of 1;Tleave_2Indicate that data flow leaves the time of interchanger 2 in network;T2Indicate two switch processes processors
Time difference;Tstart_2At the beginning of indicating data flow by interchanger 2 in network;Tleave_1Indicate that data flow leaves net
The time of interchanger 1 in network;Tc-s1And Tc-s2Respectively indicate controller to interchanger 1 and interchanger 2 two-way time;
(3) handling capacity Ui:
Wherein, Cfirst_avgIndicate the packet number of the average data stream of counter in First interchanger in measurement period i;
Cfinal_avgIndicate the packet number of the average data stream of counter in last interchanger in measurement period i;UfirstIt represents
In measurement period i, pass through the handling capacity of First interchanger;UfinalIt represents in measurement period i, is exchanged by last
The handling capacity of machine;UiIt represents in measurement period i, passes through the average throughput of First interchanger and the last one interchanger;
(4) delay variation Jitter:
Wherein, n represents the time-delay series number of measurement network, td-jRepresent j-th of time-delay series when measurement network, Avg
(td-j) represent the time-delay series average value for measuring network;
(5) measurement error RMSE (X):
Wherein, xiFor the streaming rate value of i-th systematic sampling,The true of streaming rate is sent for i-th host
Value, n are the number of sampling.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811079839.3A CN109088795B (en) | 2018-09-17 | 2018-09-17 | Controller performance analysis method based on equal-interval sampling |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811079839.3A CN109088795B (en) | 2018-09-17 | 2018-09-17 | Controller performance analysis method based on equal-interval sampling |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109088795A true CN109088795A (en) | 2018-12-25 |
CN109088795B CN109088795B (en) | 2021-11-02 |
Family
ID=64841671
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811079839.3A Active CN109088795B (en) | 2018-09-17 | 2018-09-17 | Controller performance analysis method based on equal-interval sampling |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109088795B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103795596A (en) * | 2014-03-03 | 2014-05-14 | 北京邮电大学 | Programmable control SDN measuring system and method |
CN105357068A (en) * | 2015-11-03 | 2016-02-24 | 华中科技大学 | OpenFlow network flow control method for QoS assurance of application |
CN105515793A (en) * | 2014-09-23 | 2016-04-20 | 中国电信股份有限公司 | On-line charging processing method and system, Openflow controller and Openflow switch |
-
2018
- 2018-09-17 CN CN201811079839.3A patent/CN109088795B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103795596A (en) * | 2014-03-03 | 2014-05-14 | 北京邮电大学 | Programmable control SDN measuring system and method |
CN105515793A (en) * | 2014-09-23 | 2016-04-20 | 中国电信股份有限公司 | On-line charging processing method and system, Openflow controller and Openflow switch |
CN105357068A (en) * | 2015-11-03 | 2016-02-24 | 华中科技大学 | OpenFlow network flow control method for QoS assurance of application |
Also Published As
Publication number | Publication date |
---|---|
CN109088795B (en) | 2021-11-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108600102B (en) | Flexible data transmission system based on intelligent cooperative network | |
DE102015101370A1 (en) | Managing big data in process control systems | |
CN104935476B (en) | A kind of network traffics matrix measuring method based on SDN | |
CN103795596A (en) | Programmable control SDN measuring system and method | |
CN105933184A (en) | SDN (Software Defined Network) link delay measuring method based on LLDP (Link Layer Discovery Protocol) | |
CN109831320A (en) | A kind of auxiliary flow forecast Control Algorithm, storage medium and electronic equipment | |
Zhou et al. | Chinese Internet AS-level topology | |
CN106059830A (en) | Automatic analysis method for traffic performance of PTN (Packet Transport Network) ring network | |
CN109474023A (en) | Intelligent distribution network section real time updating method, system, storage medium and terminal | |
CN106899503A (en) | The route selection method and network manager of a kind of data center network | |
CN108809752A (en) | A kind of adaptive process monitoring method, apparatus of network flow, NPB equipment and medium | |
Liu et al. | A SDN-based active measurement method to traffic QoS sensing for smart network access | |
CN101610246B (en) | Method for inferring source-target flow component of link layer of local area network | |
CN107995121A (en) | Flow statistical method in software defined network based on asterisk wildcard | |
CN108964958A (en) | path evaluation method and device | |
CN107749805A (en) | A kind of virtual machine deployment method and device | |
CN106850253A (en) | A kind of method of the transmission time reliability measurement based on multimode network | |
CN109088795A (en) | A kind of controller performance analysis method based on systematic sampling | |
CN104363142B (en) | A kind of automation data central site network performance bottleneck analysis method | |
CN107528731A (en) | Network applied to NS3 parallel artificials splits optimized algorithm | |
CN103259731B (en) | A kind of network key node Self-Similar Traffic based on ON/OFF source model generates method for simplifying | |
CN105959167B (en) | A kind of global optimization SDN measurement method based on greedy algorithm | |
CN109495315B (en) | Metropolitan area network analysis and prediction method under big data environment and readable storage medium | |
CN108108473A (en) | Data query method and server | |
Bolodurina et al. | Comprehensive approach for optimization traffic routing and using network resources in a virtual data center |
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 |