CN103414611B - A kind of flow statistical method of high speed laod network equalizing system - Google Patents
A kind of flow statistical method of high speed laod network equalizing system Download PDFInfo
- Publication number
- CN103414611B CN103414611B CN201310365816.XA CN201310365816A CN103414611B CN 103414611 B CN103414611 B CN 103414611B CN 201310365816 A CN201310365816 A CN 201310365816A CN 103414611 B CN103414611 B CN 103414611B
- Authority
- CN
- China
- Prior art keywords
- rule
- high speed
- data
- equalizing system
- laod network
- 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.)
- Expired - Fee Related
Links
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a kind of flow statistical method of high speed laod network equalizing system, by resolving the data in high speed laod network equalizing system, mate with the rule that keeper issues, and judge whether meet with rule, then the length of packet when meeting is carried out the cumulative result as traffic statistics, finally by this result feedback to webmaster; Advantage does not need a jumbo memory device to carry out data cached information, and built-in timer can allow keeper to specify any one time period statistical rules, multiple Different Rule can carry out traffic statistics simultaneously, meet complicated statistical demand.
Description
Technical field
The present invention relates to the flow statistical method in a kind of high speed network environment, especially relate to a kind of flow statistical method of high speed laod network equalizing system.
Background technology
The scale of the Internet each within hundred days, will double, various application constantly increases, especially various video website open-minded, and video flow constantly increases, online working universal etc., and network traffics are also increasing.Network Load Balance market development for the management of network traffics bottleneck is rapid, and first this technology is applied in industries such as telecommunications, movement, bank, securities trading, large-scale websites.And in a lot of enterprise, along with the development of enterprise's key network applied business, the application demand of load balancing is also increasing.High speed laod network equalizing system has become the visual plant of network application, the nucleus equipment of particularly catenet application.
For high speed laod network equalizing system, traffic statistics are the very important functions of its one.The flow statistical method of comparative maturity mainly contains NetFlow etc. now, Netflow technology is invent in DarrenKerr and BarryBruins by Cisco System Co. in 1996 the earliest, network transfer speeds at that time and the scale of network are now so not huge, and propose high requirement to the mass data storage of network traffics and later data analysis now.The method of early stage network flow statistic more and more can not meet the requirement of the traffic statistics under present high speed laod network environment, and the user of the network equipment is for the object more and more refinement of traffic statistics, and the statistical rules therefore formulated is more and more diversification also.
Based in the high speed laod network equalizing system of netflow technology, the workflow of traffic statistics is: NetFlow utilizes first IP bag data of the switch mode data streams of standard, generate NetFlow buffer memory, data same are subsequently transmitted in same data flow based on cache information, no longer mate the strategies such as relevant access control, NetFlow buffer memory contains the statistical information of subsequent data stream simultaneously.But the network equipment on the internet now is ten hundreds of, data traffic in network is very huge, the user of high speed laod network equalizing system more and more wish can in the process of stream quantitative statistics according to self-ordained requirement to complete traffic statistic function, so just inevitably have following drawback when using NetFlow method to realize traffic statistics in high speed laod network environment: the memory device 1) needing capacity huge carrys out data cached stream; 2) based on the equipment of this kind of method, network manager well can not formulate the rule of traffic statistics in use according to the demand of oneself.
Summary of the invention
Technical problem to be solved by this invention is to provide a kind of flow statistical method not needing jumbo memory device to carry out the high speed laod network equalizing system of data cached stream, under high speed network environment, adopt method of the present invention, network manager can formulate the rule of traffic statistics according to the demand of oneself, then carry out traffic statistics according to the rule of specifying.
The present invention solves the problems of the technologies described above adopted technical scheme: a kind of flow statistical method of high speed laod network equalizing system system, comprises the steps:
S1: the high speed that brings into operation laod network equalizing system, be 0 by all initialization of register being used for traffic statistics in high speed laod network equalizing system system, available machine time in high speed laod network equalizing system is loaded in a timer, enables this timer and start timing;
S2: the data that the PPP solution frame module in high speed laod network equalizing system gets off are resolved, according to the form that data specify, extract in data the characteristic information comprised: the length value of source IP address, object IP address, protocol type, source port address, destination interface address and these data, the characteristic information extracted is left in respectively in different register;
S3: the various rules issued by webmaster leave in different regular registers, and according to the demand of the reality of traffic statistics, these registers are classified, the quantity of the same type rule issued according to webmaster decides the quantity that correspondence deposits this rule-like register;
S4: the characteristic information extracted in S2 is mated with rule in S3, judge whether to meet rule, to legal data, the length value of its correspondence is added in output register corresponding to its that rule, and the numerical value finally obtained is reported webmaster as the result of these rule traffic statistics.
Compared with prior art, the invention has the advantages that by by Data Analysis, the relevant characteristic information issued to keeper in rule is extracted according to the format characteristic of data itself, then the rule formulated according to keeper carries out traffic statistics, thus, data stream is added up as required, a jumbo memory device is not needed to carry out all data messages of buffer memory, the existence of built-in timer can allow keeper to formulate the statistical rules of any one time period, and all corresponding enable process of each rule, thus make the formulation of rule become more flexible, multiple Different Rule can carry out traffic statistics simultaneously, meet complicated statistical demand.
Accompanying drawing explanation
Fig. 1 is the FB(flow block) of flow statistical method in high speed laod network equalizing system.
Embodiment
Below in conjunction with accompanying drawing embodiment, the present invention is described in further detail.
Embodiment: as shown in Figure 1, a kind of flow statistical method of high speed laod network equalizing system, function is the flow information of the fixing source IP address of statistics one, comprises the steps:
S1: the high speed that brings into operation laod network equalizing system, be 0 by all initialization of register being used for traffic statistics in high speed laod network equalizing system system, available machine time in high speed laod network equalizing system is loaded in a timer, enables this timer and start timing:
The realization of traffic statistic function has been come by the traffic statistics module be embedded in high speed laod network equalizing system.All registers for traffic statistics in high speed laod network equalizing system system are all present in this traffic statistics module.Traffic statistics module, when initialization, by the parameter value in registers all in this module, comprises input and output and is all initialized as 0.When opening high speed laod network equalizing system, this system can issue an available machine time, inform the system time that this traffic statistics module is current, traffic statistics inside modules has a timer, we load this available machine time to this timer as initial value, enable this timer and start timing.Supposing the system clock is 100MHz, so every 1*10
8individual clock is exactly 1 second.According to the clock number that high speed laod network equalizing system is run, we just can know the time of whole high speed laod network equalizing system in traffic statistics inside modules.When the initial time supposing high speed laod network equalizing system is here 17 days 10 June in 2013 34 points 59 seconds;
S2: the data that the PPP solution frame module in high speed laod network equalizing system gets off are resolved, according to the form that data specify, extract in data the characteristic information comprised: the length value of source IP address, object IP address, protocol type, source port address, destination interface address and these data, the characteristic information extracted left in respectively in different register:
Because IP stem form is fixing, it is fixing for which results in the data format that in high speed laod network equalizing system, PPP solution frame module gets off.We according to this form by data decomposition, can extract some characteristic values that we need: as the length value etc. of source IP address, object IP address, protocol type, source port address, destination interface address, data.These information all before each data 4*64 position data bit in: in the 1st 64,0 ~ 15 bit representation be the length of data, left in the register of length by name; In 2nd 64,48 ~ 55 bit representations be the type of agreement, left in the register of protocol by name; In 3rd 64,0 ~ 31 bit representation source IP address, is left in the register of ip_src_addr by name, and 32 ~ 63 bit representation object IP addresses, are left in the register of ip_dst_addr by name; In 4th 64,32 ~ 47 bit representations be source port address, left in the register of port_src_addr by name, 48 ~ 63 bit representations be destination interface address, left in the register of port_dst_addr by name by name.Separate after the frame module data of getting off are resolved the PPP in high speed laod network equalizing system and obtain M group data, the length value of M group data is respectively L
1, L
2, L
3..., L
m;
Note: the data bit left side is high-order, and the right is low level; Data bit counts from 0.
S3: the various rules issued by webmaster leave in different regular registers, and according to the demand of the reality of traffic statistics, these registers are classified, the quantity of the same type rule issued according to webmaster decides the quantity that correspondence deposits this rule-like register:
The information content in the rule that webmaster issues mainly contains two kinds: the characteristic information of time control information and objects of statistics.Temporal information comprises the initial time of traffic statistics and the end time of traffic statistics.The characteristic information of objects of statistics comprises single characteristic information and assemblage characteristic information.Single characteristic information be exactly extract in S2 source IP address, object IP address, protocol type, source port address, one in this five type of destination interface address, assemblage characteristic information is the combining form in these five kinds of characteristic informations, may be the combination of two kinds or more kinds of situation.Here to arrange when initial time is 17 days 10 June in 2013 34 points 59 seconds, when the end time is 17 days 10 June in 2013 35 points 0 second, the time of statistics is a second.The characteristic information of statistics is source IP address, and once add up 32 such characteristic informations, the register deposited is ip_src_addr0 ~ ip_src_addr31, and corresponding regular enable signal is rule_src_info_en [31].Wherein ip_src_addr0 value is " c708b746 " (hexadecimal), other registers depositing rule are all arranged to 0, strictly all rules is set to effectively, i.e. rule_src_info_en [0] ~ rule_src_info_en [31] value is 1 entirely.
S4: the characteristic information extracted in S2 is mated with Rule Information in S3, determine whether to meet rule, to legal data, the length value of its correspondence is added in output register corresponding to its that rule, and the numerical value finally obtained is reported webmaster as the result of these rule traffic statistics:
In S2, we often can be organized the source IP address of data, and can read from register ip_src_addr, judge that whether this IP address is equal with " c708b746 " in S3 rule, if equal, the length value length of these group data is added to the result of a register as the traffic statistics of this rule, here arranging this register is cnt, so cnt=length+cnt
*; Wherein cnt traffic statistics result is currency, cnt
*it is the value in previous moment.In this step, how many output is had to arrange as required, if need the system of a lot of rule being carried out to traffic statistics, the traffic statistics result that multiple such register is used for storing Different Rule can be set, here we are provided with 32 according to the quantity of the rule once received in S3, are ip_src_addr0_cnt0 ~ ip_src_addr31_cnt31 respectively.
Because other rule is empty, so only have ip_src_addr0_cnt0 to have output, ip_src_addr1_cnt1 ~ ip_src_addr31_cnt31 should be 0, in M group data, legal data have n group, and the position in M group data is respectively N1, N2, N3,, Nn, output valve should be L
n1+ L
n2+ L
n3+ ... + L
nn, L
n1, L
n2, L
n3..., L
nnrepresent respectively N1, N2, N3 ..., Nn group data length.
The PPP solution frame module mentioned in the present invention is the stuck-module in network equalizing system, and the form of relevant IP data is according to the stipulative definition of IEEE agreement.
Claims (1)
1. a flow statistical method for high speed laod network equalizing system, is characterized in that comprising the steps:
S1: the high speed that brings into operation laod network equalizing system, be 0 by all initialization of register being used for traffic statistics in high speed laod network equalizing system system, available machine time in high speed laod network equalizing system is loaded in a timer, enables this timer and start timing;
S2: the data that the PPP solution frame module in high speed laod network equalizing system gets off are resolved, according to the form that data specify, extract in data the characteristic information comprised: the length value of source IP address, object IP address, protocol type, source port address, destination interface address and these data, the characteristic information extracted is left in respectively in different register;
S3: the various rules issued by webmaster leave in different regular registers, and according to the demand of the reality of traffic statistics, these registers are classified, the quantity of the same type rule issued according to webmaster decides the quantity that correspondence deposits this rule-like register;
S4: the characteristic information extracted in S2 is mated with Rule Information in S3, determine whether to meet rule, to legal data, the length value of its correspondence is added in output register corresponding to its that rule, and the numerical value finally obtained is reported webmaster as the result of these rule traffic statistics.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310365816.XA CN103414611B (en) | 2013-08-21 | 2013-08-21 | A kind of flow statistical method of high speed laod network equalizing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310365816.XA CN103414611B (en) | 2013-08-21 | 2013-08-21 | A kind of flow statistical method of high speed laod network equalizing system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103414611A CN103414611A (en) | 2013-11-27 |
| CN103414611B true CN103414611B (en) | 2016-04-20 |
Family
ID=49607597
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310365816.XA Expired - Fee Related CN103414611B (en) | 2013-08-21 | 2013-08-21 | A kind of flow statistical method of high speed laod network equalizing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103414611B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110022248A (en) * | 2019-04-19 | 2019-07-16 | 山东浪潮云信息技术有限公司 | Link flow statistical method and system, traffic statistics host and statistics request end |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1564547A (en) * | 2004-03-25 | 2005-01-12 | 上海复旦光华信息科技股份有限公司 | High speed filtering and stream dividing method for keeping connection features |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5621781B2 (en) * | 2009-10-06 | 2014-11-12 | 日本電気株式会社 | Network system, controller, method and program |
-
2013
- 2013-08-21 CN CN201310365816.XA patent/CN103414611B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1564547A (en) * | 2004-03-25 | 2005-01-12 | 上海复旦光华信息科技股份有限公司 | High speed filtering and stream dividing method for keeping connection features |
Non-Patent Citations (1)
| Title |
|---|
| 徐绍崇.用NetFlow技术进行大型IP网络流量统计分析.《铁道通信信号》.2005,(第10期), * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103414611A (en) | 2013-11-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106209506B (en) | A kind of virtualization deep-packet detection flow analysis method and system | |
| US9531620B2 (en) | Control plane packet traffic statistics | |
| Pontarelli et al. | Traffic-aware design of a high-speed FPGA network intrusion detection system | |
| CN101656677B (en) | Message diversion processing method and device | |
| US10740027B2 (en) | High speed logging system | |
| EP2429128B1 (en) | Flow statistics aggregation | |
| CN103379002B (en) | The adaptive monitoring of telecommunication network | |
| CN108259371A (en) | A kind of network flow data analysis method and device based on stream process | |
| CN105099916B (en) | Open flows route exchange device and its processing method to data message | |
| CN102882746B (en) | Measuring method and measuring system of flow table adding rate in Openflow switchboard system | |
| US10469326B1 (en) | Discovering a computer network topology for an executing application | |
| CN109547288B (en) | Programmable flow measuring method for protocol independent forwarding network | |
| CN101753639B (en) | Service role recognition method based on flow communication mode | |
| CN104486116A (en) | Multidimensional query method and multidimensional query system of flow data | |
| CN105681125A (en) | Method for counting traffic of virtual machine extranet of cloud platform | |
| CN103618733A (en) | Data filtering system and method applied to mobile internet | |
| US8654654B2 (en) | Traffic distribution control | |
| CN107273554A (en) | Elevator intelligent monitoring system and method | |
| CN105359472A (en) | Data processing method and apparatus for OpenFlow network | |
| CN102932285A (en) | Message packaging method and message analysis method and device | |
| CN105871803A (en) | Flow-based network state rapid sensing system | |
| CN106027397B (en) | A kind of distributed measurement device network communication method of star-like expansion | |
| CN103414611B (en) | A kind of flow statistical method of high speed laod network equalizing system | |
| US10263807B2 (en) | Hierarchical statistics acceleration | |
| CN100566266C (en) | Having the belt TCP streambuf of ageing dynamic bidirectional sets up and manner of execution |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160420 Termination date: 20170821 |