CN115225733B - Identification analysis method and device based on direct routing and dynamic quantization analysis load - Google Patents
Identification analysis method and device based on direct routing and dynamic quantization analysis load Download PDFInfo
- Publication number
- CN115225733B CN115225733B CN202210162591.7A CN202210162591A CN115225733B CN 115225733 B CN115225733 B CN 115225733B CN 202210162591 A CN202210162591 A CN 202210162591A CN 115225733 B CN115225733 B CN 115225733B
- Authority
- CN
- China
- Prior art keywords
- server
- analysis
- data packet
- identification analysis
- load
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 158
- 238000013139 quantization Methods 0.000 title claims abstract description 64
- 238000000034 method Methods 0.000 claims abstract description 52
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 48
- 238000012545 processing Methods 0.000 claims abstract description 36
- 230000008569 process Effects 0.000 claims abstract description 26
- 238000006243 chemical reaction Methods 0.000 claims abstract description 7
- 238000004445 quantitative analysis Methods 0.000 claims abstract description 7
- 238000004590 computer program Methods 0.000 claims description 12
- 230000005540 biological transmission Effects 0.000 claims description 10
- 238000011156 evaluation Methods 0.000 claims description 10
- 238000013210 evaluation model Methods 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 7
- 238000013178 mathematical model Methods 0.000 claims description 7
- 206010044565 Tremor Diseases 0.000 claims description 3
- 238000013461 design Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 abstract description 3
- 238000004891 communication Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000006855 networking Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000013068 supply chain management Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
- H04L69/22—Parsing or analysis of headers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L61/00—Network arrangements, protocols or services for addressing or naming
- H04L61/09—Mapping addresses
- H04L61/25—Mapping addresses of the same type
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/20—Network architectures or network communication protocols for network security for managing network security; network security policies in general
Abstract
The invention provides an identification analysis method and device based on direct routing and dynamic quantitative analysis load, wherein the method comprises the steps of acquiring a data packet to be analyzed and sending the data packet to an LVS server; executing a destination network address conversion protocol based on the LSV server, and forwarding the data packet to be analyzed to the real network node of the identification analysis service according to a polling algorithm; processing the data packet to be analyzed through an identification analysis server corresponding to the real network node of the identification analysis service, and recording parameter indexes in the analysis process; and returning a data packet processing result to be analyzed through the identification analysis server. The invention realizes the identification analysis mechanism based on the direct routing model and the dynamic quantization analysis load algorithm.
Description
Technical Field
The invention belongs to the technical field of electronic information.
Background
The link load of the industrial network has characteristics which are distinct from those of the traditional network, and the problems of access of a large number of heterogeneous devices, limitation of time delay sensitive scenes, instantaneous high concurrency impact and the like are faced when the industrial device is routed, so that the industrial interconnection identification analysis network is required to have extremely strong robustness and flexible load balancing strategies.
The identification analysis system has similar functions to the domain name analysis system of the traditional network, the industrial Internet also needs to convert the identification of the equipment into the IP address of the server storing the information, so as to obtain the information resource, and the identification analysis system serves as an interconnection nerve hub in the whole industrial Internet architecture. However, since the analysis from the OSI network seven-layer model is performed at the seventh application layer of the network, the operation of the identity resolution protocol needs to be performed on the basis of the establishment of a reliable connection link at the transport layer. While the three-way handshake mechanism based on the TCP protocol can ensure the reliability of a transmission link, huge loss is generated on the system performance. Under the scene of massive concurrency facing the industrial Internet, the complex TCP connection requirement is difficult to ensure, and the performance of the system is fully exerted. Meanwhile, the links of the industrial Internet have complexity, the fluctuation rate of the load condition of a single link is large, and the adjacent production networks possibly have time-sharing conditions, so that the problem of load balance in the industrial Internet scene is urgent because a large amount of idle time resources of the networks are wasted if the link bandwidth of each production network is increased is determined.
At present, a plurality of technical schemes related to identification analysis exist, the industrial Internet identification is similar to an IP address in the Internet, equipment in the network is accurately positioned, and the identification analysis is a retrieval process of equipment resources. According to different analysis architectures, the existing identification analysis schemes can be divided into schemes based on ONS architecture and schemes based on non-ONS architecture, and the schemes such as EPC are included in the identification analysis system based on ONS architecture. The non-ONS architecture-based scheme includes a Handle and a scheme which is most widely implemented based on a distributed hash table (DHT, distributed Hash Table), each parsing node performs networking in a point-to-point manner, and the parsing entries are mapped to different storage addresses according to the DHT. The peer-to-peer networking mode ensures the distributed architecture of the analysis nodes and prevents individual nodes from affecting the overall situation.
With the rapid development of the identification analysis technology, many industry enterprises realize the supply chain management and life cycle management of products by utilizing the identification analysis technology through butting secondary nodes. The registration amount and the resolution amount of the identifier also reach a mass level, and how to enable the identifier resolution system to effectively process the high-concurrency identifier resolution service request is urgent. Load balancing is one of the main solutions for high concurrency requests.
In a general solution, the request forwarding is performed by prepending the server IP, i.e. placing a server dedicated to distributing the request before the server cluster actually providing the identity resolution service, which is the load balancing server. The load balancing server forwards the received request in a specified mode according to a configured algorithm, such as a random forwarding algorithm, a fast polling algorithm and the like, so as to achieve the effect that the identification analysis request arrives at the identification analysis server in a balanced mode.
The following problems exist in the conventional scheme:
1) Identifying random storage not to utilize aggregated queries
Whether the ONS-based identification analysis scheme or the non-ONS-based identification analysis scheme is adopted, the identification is clustered aiming at industry enterprises, and quick inquiry in the industry is facilitated.
2) Identification data is difficult to manage
With the national emphasis on data security and the advancement of related laws and regulations, regulatory schemes for identifying and analyzing data also need to be changed.
3) Load balancing server single point problem
The process seems to solve the problem of unbalanced load of the servers, but the load borne by a plurality of servers is converged on the load balancing server, concurrent requests which cannot be processed by a single server before, and how the load balancing server supports becomes the key of the problem. The sign of the problem is that the request delay of the identification analysis service is delayed, the identification analysis service is located at the seventh layer application layer in the network model, the identification analysis service is the layer with the lowest efficiency in the communication layer, and the service of the application layer needs to call the transmission control layer for transmission, so that the process is complicated.
4) Load balancing algorithm performance problem
Under the industrial Internet scene, the performances of different devices are very different, and a common load balancing algorithm based on the minimum connection number and polling is difficult to make the best use of the real things of each identification analysis server.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems in the related art to some extent.
Therefore, a first object of the present invention is to provide an identification parsing method based on direct routing and dynamic quantization parsing of load, which is used for solving the problem of load balancing in industrial internet scenarios.
A second object of the present invention is to provide an identification parsing apparatus based on direct routing and dynamic quantization parsing of load.
A third object of the invention is to propose a computer device.
A fourth object of the present invention is to propose a computer readable storage medium.
To achieve the above objective, an embodiment of a first aspect of the present invention provides an identification parsing method based on direct routing and dynamic quantization parsing load, including: acquiring a data packet to be analyzed and sending the data packet to an LVS server; executing a destination network address conversion protocol based on the LSV server, and forwarding the data packet to be analyzed to the real network node of the identification analysis service according to a polling algorithm; processing the data packet to be analyzed through an identification analysis server corresponding to the real network node of the identification analysis service, and recording parameter indexes in the analysis process; and returning a data packet processing result to be analyzed through the identification analysis server.
The identification analysis method based on direct routing and dynamic quantitative analysis load provided by the embodiment of the invention designs a hierarchical identification analysis architecture based on Chord routing protocol based on the evolution of an industrial Internet identification analysis system, reserves a global domain for national supervision identification analysis data, and can realize the safety and controllability of the data. And a dynamic quantitative analysis load balancing model based on a direct routing model is designed aiming at the characteristics of the industrial Internet, and finally, a massive concurrent identification analysis network architecture for bearing the industrial Internet is realized.
In addition, the identification parsing method based on the direct routing and dynamic quantization parsing load according to the above embodiment of the present invention may further have the following additional technical features:
further, in one embodiment of the present invention, when the number of id analyses reaches the set threshold, the polling algorithm is switched to a dynamic quantization analysis load algorithm, where the dynamic quantization analysis load algorithm includes:
constructing a server evaluation model of weighted quantitative indexes and qualitative indexes:
S＝S quantitative +S qualitative ，
wherein S is a quantization index of the server, S quantitative For weighting the quantitative index, S qualitative A weighted result of the qualitative index;
and obtaining a quantization index S of each server according to the parameter index in the analysis process, initializing the weight value of each server, and switching the polling algorithm of the LSV server into a dynamic quantization analysis load algorithm.
Further, in one embodiment of the present invention, the method further includes:
periodically acquiring quantization indexes S of each server and establishing a load value of a mathematical model quantization server;
and calculating the weight value of each server according to the real-time change of the load value to evaluate the real-time processing capacity of the server, distributing the request quantity according to the real-time processing capacity, and balancing the load among the servers in each period.
Further, in one embodiment of the present invention, the method further includes:
and setting up a global supervision layer on the real network node of the identification analysis service, and tracing and safely monitoring the identification analysis data.
Further, in one embodiment of the present invention, after the LVS server receives the data packet to be parsed, the method further includes:
according to whether the target port number of the data packet is the port number of the appointed identification analysis process or not, if so, forwarding the data packet to be analyzed to an identification analysis service real network node according to a polling algorithm; otherwise, processing according to the set exception handling scheme.
To achieve the above object, a second aspect of the present invention provides an identifier parsing apparatus based on direct routing and dynamic quantization parsing load, which is characterized by comprising the following modules: the acquisition module is used for acquiring the data packet to be analyzed and sending the data packet to the LVS server; the transmission module is used for executing a destination network address conversion protocol based on the LSV server and forwarding the data packet to be analyzed to the real network node of the identification analysis service according to the polling algorithm; the analysis module processes the data packet to be analyzed through an identification analysis server corresponding to the real network node of the identification analysis service, and records the parameter index in the analysis process; and the return module returns the processing result of the data packet to be analyzed through the identification analysis server.
Further, in one embodiment of the present invention, the method further includes a dynamic update module for:
constructing a server evaluation model of weighted quantitative indexes and qualitative indexes:
S＝S quantitative +S qualitative ，
wherein S is a quantization index of the server, S quantitative For weighting the quantitative index, S qualitative A weighted result of the qualitative index;
and obtaining a quantization index S of each server according to the parameter index in the analysis process, initializing the weight value of each server, and switching the polling algorithm of the LSV server into a dynamic quantization analysis load algorithm.
Further, in one embodiment of the present invention, the dynamic update module is further configured to:
periodically acquiring quantization indexes S of each server and establishing a load value of a mathematical model quantization server;
and calculating the weight value of each server according to the real-time change of the load value to evaluate the real-time processing capacity of the server, distributing the request quantity according to the real-time processing capacity, and balancing the load among the servers in each period.
To achieve the above object, an embodiment of a third aspect of the present invention provides a computer device, which is characterized by comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the identification parsing method based on direct routing and dynamic quantization parsing load as described above when executing the computer program.
To achieve the above object, a fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the identification resolution method based on direct routing and dynamic quantization resolution load as described above.
Drawings
The foregoing and/or additional aspects and advantages of the invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:
fig. 1 is a flow chart of an identification analysis method based on direct routing and dynamic quantization analysis load according to an embodiment of the present invention.
Fig. 2 is a schematic flow chart of an identification analysis device based on direct routing and dynamic quantization analysis load according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of an identifier resolution network according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of an identifier resolution network for implementing load balancing based on a DR model according to an embodiment of the present invention.
Fig. 5 is a flowchart of identification analysis based on a direct routing model and a dynamic quantization analysis load algorithm according to an embodiment of the present invention.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative and intended to explain the present invention and should not be construed as limiting the invention.
The following describes an identification parsing method and apparatus based on direct routing and dynamic quantization parsing load according to an embodiment of the present invention with reference to the accompanying drawings.
Fig. 1 is a flow chart of an identification analysis method based on direct routing and dynamic quantization analysis load according to an embodiment of the present invention.
As shown in fig. 1, the identification parsing method based on direct routing and dynamic quantization parsing load comprises the following steps:
s1: acquiring a data packet to be analyzed and sending the data packet to an LVS server;
s2: executing a destination network address conversion protocol based on the LSV server, and forwarding the data packet to be analyzed to the real network node of the identification analysis service according to a polling algorithm;
s3: processing the data packet to be analyzed through an identification analysis server corresponding to the real network node of the identification analysis service, and recording parameter indexes in the analysis process;
s4: and returning a data packet processing result to be analyzed through the identification analysis server.
The invention designs a hierarchical identification analysis architecture based on Chord routing protocol based on the evolution of an industrial Internet identification analysis system, reserves a global domain for national supervision identification analysis data, and can realize the safety and the controllability of the data. And a dynamic quantitative analysis load balancing model based on a direct routing model is designed aiming at the characteristics of the industrial Internet, and finally, a massive concurrent identification analysis network architecture for bearing the industrial Internet is realized.
Further, in one embodiment of the present invention, the method further includes:
and setting up a global supervision layer on the real network node of the identification analysis service, and tracing and safely monitoring the identification analysis data.
The industrial Internet identification analysis architecture realized by the invention is shown in figure 1. The architecture performs local networking based on a point-to-point protocol, performs hash operation on the IP of each node in the network and a port for starting an identification analysis service to obtain the label of the node, builds a DHT network according to a Chord routing protocol, and performs networking according to the logic by each industry. And selecting a node as a boundary node, wherein the boundary node is connected to a global domain network and is connected with a certain node of the global domain, and meanwhile, the global domain records by using a log when analysis occurs so as to prepare for the national data supervision requirement, and the specific identification analysis processing flow is shown in figure 3.
Further, in one embodiment of the present invention, after the LVS server receives the data packet to be parsed, the method further includes:
according to whether the target port number of the data packet is the port number of the appointed identification analysis process or not, if so, forwarding the data packet to be analyzed to an identification analysis service real network node according to a polling algorithm; otherwise, processing according to the set exception handling scheme.
The LVS can achieve communication performance close to network line transmission, and a user is connected with the LVS server without three-way handshake connection. After receiving the data packet sent by the client, the LVS server can check whether the target port number of the data packet is the port number of the designated identification analysis process according to the setting, if so, a built-in load balancing algorithm is operated, and the data packet is forwarded to a cluster for actually processing the identification analysis service according to rules; otherwise, the data packet is directly forwarded. The communication between the client and the LVS is always controlled at the kernel level, no data processing is performed, and the real identification analysis service is processed by a certain server in the identification analysis cluster connected with the LVS, wherein the real identification analysis service comprises a data packet of TCP three-way handshake. This is essentially different from reverse proxy based load balancing of nmginx, where the nmginx server needs to establish TCP connections with clients, the maximum concurrency given by the authorities is at the 5 ten thousand QPS level, but the maximum load of LVS can be extended indefinitely with server performance. Therefore, the nodes in the identification analysis network can be clustered as follows, then the load is carried out by using the load balancing server, the network structure diagram is shown in fig. 4, all requests of users are directly sent to the public network IP address of the load balancing server, and then the requests are forwarded to the identification analysis server by the load balancing server for processing.
The client request is loaded to a server in the identity resolution server cluster, but this presents a problem. Because the destination IP address of the id resolution request packet sent from the client is VIP of the load balancing server, but not RIP of the server loaded, the host will not process a packet whose destination address is not itself. This requires NAT protocol for network address translation, and in most cases the IP addresses used are private network addresses that can only be used in one local area network, which are not actually real IP addresses, only the IP addresses on the router are real public network addresses. These private addresses are not visible on the internet, and there are two addresses on the router, one is a public network address and one is a private address. The router connects to the operator ISP through a public network address and ultimately to the host on the internet to be accessed. This also means that a public network address is necessary to access a host on the internet.
The LVS is utilized to forward the load to the identification analysis server, and the identification analysis server can return by itself after processing the identification analysis request. The load balancing server is transparent to the client, the client does not know that all but the forwarding function is actually available on the IP address requested by the client, and the real service is realized in the following cluster. Meanwhile, the NAT protocol is not needed to be used for converting the IP address in the model, so that the pressure of the load balancing server is greatly reduced.
Further, in one embodiment of the present invention, when the number of id analyses reaches the set threshold, the polling algorithm is switched to a dynamic quantization analysis load algorithm, where the dynamic quantization analysis load algorithm includes:
constructing a server evaluation model of weighted quantitative indexes and qualitative indexes:
S＝S quantitative +S qualitative ，
wherein S is a quantization index of the server, S quantitative For weighting the quantitative index, S qualitative A weighted result of the qualitative index;
and obtaining a quantization index S of each server according to the parameter index in the analysis process, initializing the weight value of each server, and switching the polling algorithm of the LSV server into a dynamic quantization analysis load algorithm.
Further, in one embodiment of the present invention, the method further includes:
periodically acquiring quantization indexes S of each server and establishing a load value of a mathematical model quantization server;
and calculating the weight value of each server according to the real-time change of the load value to evaluate the real-time processing capacity of the server, distributing the request quantity according to the real-time processing capacity, and balancing the load among the servers in each period.
Quantitative evaluation is given according to the performance of a processor, the performance of a server system and the performance of a high-performance computer such as SPEC (specific performance class), HPCC (high performance CC) and the like, and the evaluation schemes mainly aim at basic performance parameters of the server or independently aim at testing a certain performance, but the indexes are far different from evaluation indexes of the performance of an industrial Internet identification analysis server, so that the invention firstly sets a model aiming at the performance evaluation of the identification analysis server. Firstly, a plurality of test parameters for reflecting various resources and performances of a server are needed to judge the advantages and disadvantages of the performances of the server, and two types of parameters are mainly introduced by referring to a server performance evaluation model: quantitative and qualitative indexes. Wherein the quantitative index comprises: identifying the concurrency of the analysis request, analyzing the response bandwidth of the request, identifying the analysis time delay, the delay jitter and the packet loss rate, and identifying the effective utilization rate of a CPU (Central processing Unit) of an analysis server and the average waiting time of I/O (input/output) read-write operation of the server; the qualitative indexes comprise: reliability, scalability, availability. For the first class of indexes, because accurate values can be obtained through service operation, only the weighting coefficients which are determined by the requirements of different industries on the identification analysis service are required to be weighted linearly. And for the second type of index, the fuzzy processing can be performed, so that a server evaluation model for weighting quantitative indexes and qualitative indexes can be obtained:
S＝S quantitative +S qualitative (1)
Wherein, the positive and negative indexes of the quantitative index, such as time delay and packet loss rate, are negative indexes, so that the method can processNegative index treatment of q i ＝1/q i The forward index is directly multiplied by the weight coefficient, and a further expression can be obtained:
S quantitative ＝∑w i q i (i in qps, bandwidth, delay, tremble, loss, use, wait), (formula 2)
Whereas for qualitative index, by designing an evaluation set v= (V) for each term 1 ,v 2 ,v 3 ) When the corresponding index of the server is v 1 When the membership degree is ambiguous as:
simultaneously, according to a relative comparison method, three index weight coefficients are respectively set as follows:
w 2 ＝[0.5,0.3,0.2](4)
Meanwhile, assuming that the objective evaluation result of the identification analysis server is P, the quantitative result of the qualitative index is:
S qualitative ＝w 2 and (5) RP. (5)
According to DQRB (Dynamic Quantitative Resolution Balance) algorithm realized by the dynamic quantization index, the performance of the server is quantized according to the key performance index of the server, and a single weight value is not set subjectively; the load is quantified by the amount of pressure required to be caused to each performance index of the server, and the number of connections is no longer used as a standard for load measurement. The specific flow of the DQRB algorithm is as follows:
1) Firstly, each parameter needs to be calculated according to a quantization formula of (formula 2), so that a polling algorithm is adopted to receive an identification analysis request and respond, and after the operation meets a calculation standard period, the quantization index S of each server is obtained.
2) And initializing weight values of all servers according to the collected index S, and switching the load balancing of the LVS to a DQRB algorithm.
3) Periodically acquiring load parameters of each server and establishing a reasonable mathematical model to quantify the load values of the servers; and calculating a weight value of each server according to the real-time change of the quantized load to evaluate the real-time processing capacity of the server, and distributing the request quantity according to the real-time processing capacity to balance the load among the servers in each period.
The whole load balancing algorithm operation flow of the identification analysis system is shown in fig. 5.
According to the identification analysis method based on the direct routing and the dynamic quantitative analysis load, in the first aspect, the identification analysis network architecture realized by means of Chord routing protocol is provided, and the currently applied identification analysis architecture does not consider the supervision scheme of clustering the identification and identification analysis data of industry enterprises; in the second aspect, an identification analysis node load balancing scheme realized based on a Linux virtual server technology is designed, an identification analysis service is located in a seventh layer application layer in a network model, the identification analysis service is a layer with the lowest efficiency in a communication layer, and the service of the application layer needs to call a transmission control layer for transmission, but an LVS technology can realize load balancing in a network layer, so that the concurrency capacity of the whole network is enhanced; in the third aspect, a model aiming at the performance evaluation of the identification resolution server is designed first, and two types of parameters are mainly introduced: quantitative and qualitative indexes; for the first type of indexes, because accurate values can be obtained through service operation, only the linear weighting of the weighting coefficients is required to be determined according to the requirements of different industries on the identification analysis service, and for the second type of indexes, the fuzzy processing can be performed, so that a server evaluation model for weighting quantitative indexes and qualitative indexes can be obtained; in the fourth aspect, according to the DQRB (Dynamic Quantitative Resolution Balance) algorithm implemented by the dynamic quantization index, the performance of the server is quantized according to the key performance index of the server, instead of subjectively setting a single weight value, so as to request the pressure caused by each performance index of the server to quantize the load, and the connection number is not used as the standard of load measurement, thereby finally implementing the identification analysis mechanism based on the direct routing model and the dynamic quantization analysis load algorithm.
In order to realize the embodiment, the invention also provides an identification analysis device based on direct routing and dynamic quantization analysis load.
Fig. 2 is a schematic structural diagram of an identifier parsing device based on direct routing and dynamic quantization parsing load according to an embodiment of the present invention.
As shown in fig. 2, the identification parsing apparatus based on direct routing and dynamic quantization parsing load includes: the system comprises an acquisition module 10, a transmission module 20, an analysis module 30 and a return module 40, wherein the acquisition module is used for acquiring a data packet to be analyzed and sending the data packet to the LVS server; the transmission module is used for executing a destination network address conversion protocol based on the LSV server and forwarding the data packet to be analyzed to the real network node of the identification analysis service according to the polling algorithm; the analysis module processes the data packet to be analyzed through an identification analysis server corresponding to the real network node of the identification analysis service, and records the parameter index in the analysis process; and the return module returns the processing result of the data packet to be analyzed through the identification analysis server.
Further, in one embodiment of the present invention, the method further includes a dynamic update module for:
constructing a server evaluation model of weighted quantitative indexes and qualitative indexes:
S＝S quantitative +S qualitative ，
wherein S is a quantization index of the server, S quantitative For weighting the quantitative index, S qualitative A weighted result of the qualitative index;
and obtaining a quantization index S of each server according to the parameter index in the analysis process, initializing the weight value of each server, and switching the polling algorithm of the LSV server into a dynamic quantization analysis load algorithm.
Further, in one embodiment of the present invention, the dynamic update module is further configured to:
periodically acquiring quantization indexes S of each server and establishing a load value of a mathematical model quantization server;
and calculating the weight value of each server according to the real-time change of the load value to evaluate the real-time processing capacity of the server, distributing the request quantity according to the real-time processing capacity, and balancing the load among the servers in each period.
To achieve the above object, an embodiment of a third aspect of the present invention provides a computer device, which is characterized by comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the identification parsing method based on direct routing and dynamic quantization parsing load as described above when executing the computer program.
To achieve the above object, a fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the identification resolution method based on direct routing and dynamic quantization resolution load as described above.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present invention, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise.
While embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the invention, and that variations, modifications, alternatives and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the invention.
Claims (6)
1. The identification analysis method based on direct routing and dynamic quantization analysis load is characterized by comprising the following steps:
acquiring a data packet to be analyzed and sending the data packet to an LVS server;
based on the LVS server executing a destination network address conversion protocol, forwarding the data packet to be analyzed to an identification analysis service real network node according to a polling algorithm;
processing the data packet to be analyzed through an identification analysis server corresponding to the real network node of the identification analysis service, and recording parameter indexes in the analysis process;
returning the processing result of the data packet to be analyzed through the identification analysis server;
when the identification analysis times reach a set threshold value, switching the polling algorithm into a dynamic quantitative analysis load algorithm, wherein the dynamic quantitative analysis load algorithm comprises the following steps:
constructing a server evaluation model of weighted quantitative indexes and qualitative indexes:
S＝S quantitative +S qualitative ，
wherein S is a quantization index of the server, S quantitative For weighting the quantitative index, S qualitative A weighted result of the qualitative index; wherein the negative index is treated as q i ＝1/q i The forward index is directly multiplied by the weight coefficient, and a further expression can be obtained:
S quantitative ＝Σw i q i (i in qps,bandwidth,delay,tremble,loss,usage,wait)，
for qualitative indications, by providing eachItem design evaluation set v= (V) 1 ,v 2 ,v 3 ) When the corresponding index of the server is v 1 When the membership degree is ambiguous as:
simultaneously, according to a relative comparison method, three index weight coefficients are respectively set as follows:
w 2 ＝[0.5,0.3,0.2]，
meanwhile, assuming that the objective evaluation result of the identification analysis server is P, the quantitative result of the qualitative index is:
S qualitative ＝w 2 RP；
according to the parameter indexes in the analysis process, obtaining quantization indexes S of each server, initializing weight values of each server, and switching a polling algorithm of the LVS server into a dynamic quantization analysis load algorithm;
periodically acquiring quantization indexes S of each server and establishing a load value of a mathematical model quantization server;
and calculating the weight value of each server according to the real-time change of the load value to evaluate the real-time processing capacity of the server, distributing the request quantity according to the real-time processing capacity, and balancing the load among the servers in each period.
2. The method as recited in claim 1, further comprising:
and setting up a global supervision layer on the real network node of the identification analysis service, and tracing and safely monitoring the identification analysis data.
3. The method of claim 1, further comprising, after the LVS server receives the data packet to be parsed:
according to whether the target port number of the data packet is the port number of the appointed identification analysis process or not, if so, forwarding the data packet to be analyzed to an identification analysis service real network node according to a polling algorithm; otherwise, processing according to the set exception handling scheme.
4. An identification analysis device based on direct routing and dynamic quantization analysis load is characterized by comprising the following modules:
the acquisition module is used for acquiring the data packet to be analyzed and sending the data packet to the LVS server;
the transmission module is used for executing a destination network address conversion protocol based on the LVS server and forwarding the data packet to be analyzed to an identification analysis service real network node according to a polling algorithm;
the analysis module processes the data packet to be analyzed through an identification analysis server corresponding to the identification analysis service real network node, and records parameter indexes in the analysis process;
the return module returns the processing result of the data packet to be analyzed through the identification analysis server;
the system further comprises a dynamic updating module for:
constructing a server evaluation model of weighted quantitative indexes and qualitative indexes:
S＝S quantitative +S qualitative ，
wherein S is a quantization index of the server, S quantitative For weighting the quantitative index, S qualitative A weighted result of the qualitative index;
according to the parameter indexes in the analysis process, obtaining quantization indexes S of each server, initializing weight values of each server, and switching a polling algorithm of the LVS server into a dynamic quantization analysis load algorithm; wherein the negative index is treated as q i ＝1/q i The forward index is directly multiplied by the weight coefficient, and a further expression can be obtained:
S quantitative ＝Σw i q i (i in qps,bandwidth,delay,tremble,loss,usage,wait)，
for qualitative index, by designing an evaluation set v= (V) 1 ,v 2 ,v 3 ) When the corresponding index of the server is v 1 When the membership degree is ambiguous as:
simultaneously, according to a relative comparison method, three index weight coefficients are respectively set as follows:
w 2 ＝[0.5,0.3,0.2]，
meanwhile, assuming that the objective evaluation result of the identification analysis server is P, the quantitative result of the qualitative index is:
S qualitative ＝w 2 RP；
according to the parameter indexes in the analysis process, obtaining quantization indexes S of each server, initializing weight values of each server, and switching a polling algorithm of the LVS server into a dynamic quantization analysis load algorithm;
periodically acquiring quantization indexes S of each server and establishing a load value of a mathematical model quantization server;
and calculating the weight value of each server according to the real-time change of the load value to evaluate the real-time processing capacity of the server, distributing the request quantity according to the real-time processing capacity, and balancing the load among the servers in each period.
5. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of any of claims 1-3 when executing the computer program.
6. A computer readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, implements the method according to any of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210162591.7A CN115225733B (en) | 2022-02-22 | 2022-02-22 | Identification analysis method and device based on direct routing and dynamic quantization analysis load |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210162591.7A CN115225733B (en) | 2022-02-22 | 2022-02-22 | Identification analysis method and device based on direct routing and dynamic quantization analysis load |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115225733A CN115225733A (en) | 2022-10-21 |
| CN115225733B true CN115225733B (en) | 2024-04-05 |
Family
ID=83606272
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210162591.7A Active CN115225733B (en) | 2022-02-22 | 2022-02-22 | Identification analysis method and device based on direct routing and dynamic quantization analysis load |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115225733B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103220354A (en) * | 2013-04-18 | 2013-07-24 | 广东宜通世纪科技股份有限公司 | Method for achieving load balancing of server cluster |
| WO2017050117A1 (en) * | 2015-09-25 | 2017-03-30 | 网宿科技股份有限公司 | Network load balance processing system, method, and apparatus |
| CN108092798A (en) * | 2017-11-27 | 2018-05-29 | 西安财经学院 | A kind of cloud service preferred method, Cloud Server based on change granularity |
| CN111930511A (en) * | 2020-08-24 | 2020-11-13 | 北京工业大学 | Identifier resolution node load balancing device based on machine learning |
| CN113014689A (en) * | 2021-03-01 | 2021-06-22 | 杭州迪普科技股份有限公司 | Method and device for network address translation |
| WO2021218068A1 (en) * | 2020-04-28 | 2021-11-04 | 网络通信与安全紫金山实验室 | Icn-based industrial internet identifier analysis system and data access method |
| CN113794739A (en) * | 2021-11-16 | 2021-12-14 | 北京邮电大学 | Double-layer active defense method and device for man-in-the-middle attack |
| CN113938488A (en) * | 2021-09-24 | 2022-01-14 | 浙江理工大学 | Load balancing method based on dynamic and static weighted polling |
| CN114051049A (en) * | 2021-02-10 | 2022-02-15 | 中国信息通信研究院 | Proxy forwarding method of identifier, server and computer readable storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6665702B1 (en) * | 1998-07-15 | 2003-12-16 | Radware Ltd. | Load balancing |
| US8635350B2 (en) * | 2006-06-12 | 2014-01-21 | Icontrol Networks, Inc. | IP device discovery systems and methods |
| CA3015486A1 (en) * | 2016-02-23 | 2017-08-31 | Level 3 Communications, Llc | Systems and methods for content server rendezvous in a dual stack protocol network |
| US10911570B2 (en) * | 2017-11-02 | 2021-02-02 | Utech, Inc. | System and method for content parsing |
-
2022
- 2022-02-22 CN CN202210162591.7A patent/CN115225733B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103220354A (en) * | 2013-04-18 | 2013-07-24 | 广东宜通世纪科技股份有限公司 | Method for achieving load balancing of server cluster |
| WO2017050117A1 (en) * | 2015-09-25 | 2017-03-30 | 网宿科技股份有限公司 | Network load balance processing system, method, and apparatus |
| CN108092798A (en) * | 2017-11-27 | 2018-05-29 | 西安财经学院 | A kind of cloud service preferred method, Cloud Server based on change granularity |
| WO2021218068A1 (en) * | 2020-04-28 | 2021-11-04 | 网络通信与安全紫金山实验室 | Icn-based industrial internet identifier analysis system and data access method |
| CN111930511A (en) * | 2020-08-24 | 2020-11-13 | 北京工业大学 | Identifier resolution node load balancing device based on machine learning |
| CN114051049A (en) * | 2021-02-10 | 2022-02-15 | 中国信息通信研究院 | Proxy forwarding method of identifier, server and computer readable storage medium |
| CN113014689A (en) * | 2021-03-01 | 2021-06-22 | 杭州迪普科技股份有限公司 | Method and device for network address translation |
| CN113938488A (en) * | 2021-09-24 | 2022-01-14 | 浙江理工大学 | Load balancing method based on dynamic and static weighted polling |
| CN113794739A (en) * | 2021-11-16 | 2021-12-14 | 北京邮电大学 | Double-layer active defense method and device for man-in-the-middle attack |
Non-Patent Citations (2)
| Title |
|---|
| 基于工作负载预测的虚拟机整合算法;魏亮;黄韬;陈建亚;刘韵洁;;电子与信息学报(第06期);全文 * |
| 基于软件定义网络的服务器集群负载均衡技术研究;于天放;芮兰兰;邱雪松;;电子与信息学报(第12期) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115225733A (en) | 2022-10-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8065395B2 (en) | System and method for queuing to a cloud via a queuing proxy | |
| CN111600936B (en) | Asymmetric processing system based on multiple containers and suitable for ubiquitous electric power internet of things edge terminal | |
| CN107135268B (en) | Distributed task computing method based on information center network | |
| CA2489142C (en) | Distributed computer | |
| KR20130070500A (en) | Method and apparatus for processing server load balancing with the result of hash function | |
| Di Modica et al. | Resource and service discovery in SOAs: A P2P oriented semantic approach | |
| US10061621B1 (en) | Managing resources in a configurable computing environment | |
| CN112839081A (en) | Load balancing method of cloud cluster | |
| WO2021120633A1 (en) | Load balancing method and related device | |
| Soleimanzadeh et al. | SD‐WLB: An SDN‐aided mechanism for web load balancing based on server statistics | |
| CN115225733B (en) | Identification analysis method and device based on direct routing and dynamic quantization analysis load | |
| Ke et al. | Load balancing using P4 in software-defined networks | |
| Lai et al. | A scalable multi-attribute hybrid overlay for range queries on the cloud | |
| US11075840B1 (en) | Disaggregation of network traffic | |
| CN110049081A (en) | For build and using high availability Docker private library method and system | |
| KR100475668B1 (en) | Apparatus and method of providing information for grid resources in grid network and computer readable recording medium of recording process thereof | |
| Gu et al. | Optimization of service addition in multilevel index model for edge computing | |
| EP3685567B1 (en) | Load shedding of traffic based on current load state of target capacity | |
| Chen | Design of computer big data processing system based on genetic algorithm | |
| Altangerel et al. | Study on emerging applications on data plane and optimization possibilities | |
| CN113946587A (en) | Handle identifier analysis caching method, query method and handle identifier analysis system | |
| Fan et al. | An edge computing service model based on information-centric networking | |
| CN112104566B (en) | Processing method and device for load balancing | |
| CN103905249B (en) | A kind of mobile Internet network method for managing and monitoring based on JXME | |
| Wang | Content, topology and cooperation in in-network caching |
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 |