Method, system and equipment for deep identification of application assets of site servers

A self-learning and adaptive method for server application identification enhances accuracy by collecting and analyzing transmission flow and host data, correcting baselines, and reporting anomalies, effectively managing unattended power stations.

CN115801538BActive Publication Date: 2024-12-06YUNNAN POWER GRID CO LTD
View PDF 13 Cites 0 Cited by

Patent Information

Application Number
CN202211404367.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-11-10
Publication Date
2024-12-06
Estimated Expiration
2042-11-10

AI Technical Summary

Technical Problem

Existing methods for identifying server applications are inaccurate due to the use of hidden or fake port numbers, and differences in traffic characteristics among applications, leading to misidentification and increased management costs, particularly in unattended power stations.

Method used

A self-learning and adaptive application identification method that collects inter-server transmission flow information, performs deep detection on hosts, calculates and modifies application baselines, and reports anomalies in real time using a traffic analysis module and operating system probes.

Benefits of technology

Improves recognition accuracy and enables quick location of non-real business application abnormalities, supporting automatic monitoring in unattended power stations with a visual topology for anomaly detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115801538B_ABST
    Figure CN115801538B_ABST
Patent Text Reader

Abstract

The present invention discloses a method, system and device for deep identification of application assets of site servers, including a deep identification technology for application assets of site servers, which is mainly used for self-learning and self-adaptation of server application identification baselines. The method first collects transmission stream information and host process data between servers, analyzes and identifies transmission stream sessions, and then performs deep detection on the host, calculates the application baseline according to the detection results, corrects the application baseline based on the host deep detection results, and finally monitors the process to check the application baseline in real time, and reports abnormalities to the platform. The method can use multiple ways to identify applications and combine them for cross-validation to improve the identification accuracy, and at the same time establish an application identification self-learning baseline. For senior management personnel, the specific location where the abnormality of non-real business application process occurs can be quickly located through visual topology.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of computer network technology, and in particular to a method for deep identification of application assets of a site server. Background Art

[0002] Server applications have always been the core assets of business operations, and application identification has always been the focus of business monitoring. Generally speaking, the inability to identify and perceive business will lead to an increase in management costs. However, for unattended power stations, this problem is even more prominent. Therefore, providing deep perception of business applications and building a visual, adaptive, and management-friendly field business application monitoring baseline has a value that cannot be ignored.

[0003] The traditional application identification method can be traced back to the early port-based identification. This method uses different port numbers registered by different applications to identify applications. However, as more and more illegal applications use hidden or fake port numbers, the accuracy of port number identification has become unsatisfactory. Therefore, on this basis, DPI and DFI have gradually become the mainstream identification methods.

[0004] However, in reality, there is no consensus on which identification method can achieve the best application identification effect. This is because different application types have different states in session connections or data flows, showing different traffic characteristics. Take HTTP, VPN, and NAT as examples:

[0005] Application 1: HTTP Duration: short Average rate: high Number of bytes transferred: medium;

[0006] Application 2: VPN Duration: Long Average Rate: Low Number of Bytes Transferred: High;

[0007] Application 3: NAT Duration: short Average rate: low Number of bytes transferred: low;

[0008] In fact, even the same type of applications may not have obvious features. For example, the duration of NAT applications is not necessarily very short. Therefore, the insensitive features of a few applications also lead to the possibility of misidentification.

[0009] Compared with the two, DPI technology can identify specific applications more accurately, so it is widely used in various systems that require accurate application identification; while DFI technology can identify traffic that DPI technology cannot identify, such as P2P encrypted traffic, because it uses a traffic model. Summary of the invention

[0010] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0011] In view of the above existing problems, the present invention is proposed.

[0012] Therefore, in order to effectively identify and monitor server application assets, the present invention proposes a self-learning and adaptive application identification baseline method.

[0013] In order to solve the above technical problems, the present invention provides the following technical solutions: a site server applies an asset deep identification method, comprising:

[0014] Collect inter-server transmission flow information and host process data;

[0015] Analyze and identify transport stream sessions;

[0016] Conduct deep detection on the host;

[0017] Calculate application baselines;

[0018] Modify the application baseline based on the host deep detection results;

[0019] The monitoring process checks the application baseline in real time and reports any anomalies to the platform.

[0020] As a preferred solution of the site server application asset deep identification method of the present invention, wherein: the transport stream information and host process data include transport stream session information collected through an interface;

[0021] The session information includes source IP, destination IP, destination port address, source port address, and transmission protocol.

[0022] As a preferred solution of the site server application asset deep identification method described in the present invention, the analysis and identification of the transmission stream session includes: the traffic identification module performs cluster analysis on the collected session information and performs rapid application identification based on the characteristics.

[0023] As a preferred solution of the site server application asset deep identification method, system and device described in the present invention, the deep detection includes deep detection of the host through an operating system probe.

[0024] As a preferred solution of the site server application asset deep identification method, system and device described in the present invention, the calculation of the application baseline includes, through the identification results of the traffic analysis module, establishing the application baseline based on the business application traffic information of the interaction between the identified servers.

[0025] As a preferred solution of the site server application asset deep identification method, system and device described in the present invention, wherein: the correction of the application baseline includes comparing the deep detection results with the identification results of the traffic analysis module, and correcting the application baseline model based on the comparison results.

[0026] As a preferred solution of the site server application asset deep identification method, system and device described in the present invention, wherein: the correction of the application baseline also includes reporting non-business applications and services discovered during deep detection to the management platform.

[0027] As a preferred solution of the method, system and device for deep identification of application assets of site servers described in the present invention, wherein: the identification result by the traffic analysis module includes:

[0028] Set up two servers A and B under the host, and identify the traffic between server A and server B as application 1 corresponding to traffic X1, application 2 corresponding to traffic X2, and application n corresponding to traffic Xn, and finally obtain the application information between server A and server B marked by traffic;

[0029] Assume that there is another server C in the host besides A and B. The traffic between server B and server C identifies the traffic X3 corresponding to application 3, the traffic X4 corresponding to application 4, and the traffic Xn corresponding to application n. Finally, the application information marked by traffic between server B and server C is obtained.

[0030] Finally, a baseline model of the entire server application assets based on the site is obtained.

[0031] As a preferred solution of the site server application asset deep identification system described in the present invention, it includes:

[0032] Information collection module, used to collect information including source IP, destination IP, destination port address, source port address, and transmission protocol;

[0033] Traffic identification module, used to perform cluster analysis on the collected session information and quickly identify applications based on features;

[0034] Deep detection module, which performs deep detection on the host through operating system probes

[0035] The correction module is used to compare the deep detection result with the recognition result of the traffic analysis module, and correct the application baseline model based on the comparison result.

[0036] As a preferred solution of the site server application asset deep identification device described in the present invention, wherein: a computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the site server application asset deep identification method and system steps when executing the computer program.

[0037] Beneficial effects of the present invention: The method of the present invention is based on a self-learning and adaptive application identification baseline method, which adopts multiple approaches to application identification and combines them for cross-validation to improve the recognition accuracy. The method also establishes an application identification self-learning baseline, which effectively assists unmanned power stations in automatic monitoring. At the same time, for senior management personnel, the site and its specific server where the non-real business application process abnormality occurs can be quickly located through the visual topology. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0039] Figure 1 A schematic diagram of the overall architecture of a method and system for deep identification of application assets of a site server provided by an embodiment of the present invention;

[0040] Figure 2 A schematic diagram of the process of a method for deep identification of assets applied by a site server provided in one embodiment of the present invention.

[0041] Figure 3 A schematic diagram of a baseline model of a method for deep identification of assets applied to a site server provided by an embodiment of the present invention;

[0042] Figure 4 An example diagram of a traffic prediction model for a site server application asset deep identification method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0044] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0045] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0046] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0047] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0048] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0049] Example 1

[0050] Reference Figure 1-4 , which is the first embodiment of the present invention, and provides a method for deep identification of application assets of a site server, including:

[0051] S1: collects inter-server transmission flow information and host process data;

[0052] Furthermore, the transmission flow information between servers is obtained through the acquisition module, and the acquisition module preliminarily screens the session information and uploads it to the analysis module for processing.

[0053] It should be noted that the host side reads each network access and port information in the host through the operating system probe and records the connection status of the process service.

[0054] S2: Analyze and identify the transport stream session;

[0055] Furthermore, the traffic analysis module makes a preliminary judgment on the session information to check whether the session information has been marked.

[0056] It should be noted that if the session information has been marked and its features have been learned, it will directly enter the protocol identification module;

[0057] If the session information is not labeled, feature learning is required.

[0058] Furthermore, the traffic characteristics of the session, such as the connection rate, the relevance of each link, and the data transmission byte distribution, are recorded, and then predicted through classification training, and the prediction results are verified with the process detection results.

[0059] S3: Perform deep detection on the host;

[0060] Furthermore, the operating system probe periodically obtains information such as the host CPU, memory, hard disk, network card, operating system, network configuration, installed software, account status, creation time, the latest login time of the account, login IP, etc. through the monitoring module and writes it into the information file. The application process information is extracted for comparison with the prediction results.

[0061] S4: Calculate application baseline.

[0062] Furthermore, the key to calculating the application baseline lies in the classification training prediction model in S2. This requires using the labeled session information as learning samples and using the application characteristics of these known services as classification criteria;

[0063] Furthermore, the training is to periodically adjust the classification standards according to the changes in the learning samples, and the corresponding entire prediction model will also be revised; as several learning cycles are completed, the prediction model will tend to stabilize and obtain relatively accurate business application classification standards, thereby improving prediction accuracy and realizing self-learning of application baselines.

[0064] It should be noted that, for the convenience of observation, the external presentation of the application baseline adopts a visual topology structure. In the visual topology of business mutual access between servers, each line displays the application identification result of its corresponding traffic, and uses color difference to mark abnormal application processes in massive connections.

[0065] S5: Modify the application baseline based on the host deep detection result.

[0066] It should be noted that the application identification results after deep detection of the host are compared with the identification results of the traffic analysis model. After filtering out the non-real business application identification results, the newly added business applications are added to the traffic baseline to achieve self-learning of the baseline.

[0067] S6: The monitoring process checks the application baseline in real time and reports any anomalies to the platform.

[0068] It should be noted that any connection will open a network port, and the network ports used by fixed services are also relatively fixed, so abnormal network access can be discovered through abnormal network ports. By reading each network access and port information in the host and comparing it with the trusted network port data, abnormal access connections are discovered and alarms are issued, and the alarm results are reported to the platform.

[0069] Example 2

[0070] Reference Figure 1 , which is an embodiment of the present invention, provides a method for deep identification of site server application assets. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0071] The self-learning adaptive application recognition baseline method proposed in this embodiment is as shown in the attached Figure 1 As shown, it mainly consists of a management platform, a traffic identification device and a host deep detection device;

[0072] The traffic analysis module includes an application signature library and an identification record table.

[0073] The feature database stores the features of various grid services:

[0074] Application 1, application layer feature Y1; application 2, application layer feature Y2; application 3, application layer feature Y3... and so on

[0075] The identification record table includes basic five-tuple information (source IP, destination IP, source port, destination port, protocol type) and identification code X;

[0076] Preprocessing: The traffic analysis module compares the five-tuple information of the newly collected session with the known five-tuple information. If they are consistent, the newly collected session is marked with the same identification code in the record table, and the application is identified in the protocol processing module based on the identification code;

[0077] If no matching five-tuple information is found in the record table, the newly collected traffic enters the traffic identification model;

[0078] The traffic analysis module further performs feature analysis on the application layer of the collected traffic, and compares the analyzed application layer features with the feature library to see whether they are consistent with the features of known services. If they are consistent, the traffic analysis module marks the corresponding identification code;

[0079] If no matching feature information is found in the feature library, the collected traffic will enter the prediction model for classification;

[0080] The prediction model is learned by taking the traffic that can be accurately identified as learning samples and classifying these samples;

[0081] Calculate the features of each sample and mark the samples to determine their corresponding classification;

[0082] Similarly, all samples that can be learned are classified and learned. After all the learning and classification are completed, the feature prediction matrix M of the prediction model is obtained. The prediction matrix M includes necessary field information such as category name, feature name, feature value, etc.

[0083] The training process of the model also corresponds to the self-learning process of the baseline;

[0084] When all traffic has been learned and predicted, it enters the protocol processing module for final application identification result processing;

[0085] In-depth detection of hosts plays an important role in model optimization. As mentioned earlier, the features of a few applications are insensitive. Although matrix learning can improve the sensitivity as much as possible, there is still uncertainty. The detection results of the hosts can effectively correct the model. On the management platform, a visual topology is used to present the application baseline, and colors are used to distinguish between normal business application traffic interactions and abnormal application processes.

[0086] For the inter-access lines marked with special colors, you can quickly view the abnormal application results between server A and server B contained in them, and support setting automatic early warnings for them.

[0087] The comparison between the method of the present invention and the conventional technology is shown in the following table:

[0088] The technology of the present invention Traditional Technology Recognition accuracy 86.2% 55.4% Abnormal occurrence location rate 79.9% 47.1%

[0089] The present invention is a deep identification technology for site server application assets, which is mainly used for self-learning and self-adaptation of server application identification baseline. The method first collects the transmission flow information and host process data between servers, analyzes and identifies the transmission flow session, and then deeply detects the host, calculates the application baseline according to the detection result, corrects the application baseline based on the host deep detection result, and finally monitors the process to check the application baseline in real time, and reports abnormalities to the platform.

[0090] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0091] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0092] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0093] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0094] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0095] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0096] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0097] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for deep identification of assets applied by a station server, characterized in that: include, Collect inter-server transmission flow information and host process data; Analyze and identify transport stream sessions; Conduct deep detection on the host; Establish application baselines; Modify the application baseline based on the host deep detection results; The monitoring process checks the application baseline in real time, finds anomalies and reports them to the platform; The transport stream information and host process data include transport stream session information collected through the interface; The session information includes source IP, destination IP, destination port address, source port address, and transport protocol; The analyzing and identifying the transport stream session includes the traffic identification module performing cluster analysis on the collected session information and identifying the application according to the characteristics; The deep detection includes performing deep detection on the host through an operating system probe; The establishing of the application baseline includes obtaining identification results through a traffic analysis module, and establishing the application baseline according to the identified business application traffic information of the interaction between servers; The correction of the application baseline includes comparing the deep detection results with the identification results of the traffic analysis module, filtering the identification results of non-real business applications, adding new business applications to the traffic baseline, and reporting the non-business applications and services found in the deep detection to the management platform; The identification result obtained by the traffic analysis module includes setting two servers A and B under the host, identifying the traffic X1 corresponding to application 1, the traffic X2 corresponding to application 2, and the traffic Xn corresponding to application n based on the traffic between server A and server B, and finally obtaining the application information marked by the traffic between server A and server B; Set up another server C in addition to A and B in the host subordinates, and identify the traffic X3 corresponding to application 3, the traffic X4 corresponding to application 4, and the traffic Xn corresponding to application n based on the traffic between server B and server C, and finally obtain the application information marked by traffic between server B and server C; Finally, a baseline model of the entire server application assets based on the site is obtained.

2. The station server applies the asset deep identification system, which is characterized by: include, An information transmission module is used to transmit information including source IP, destination IP, destination port address, source port address, and transmission protocol; Traffic identification module, used to perform cluster analysis on the collected session information and identify applications based on features; The traffic analysis module is used to set two servers A and B under the host, identify the traffic X1 corresponding to application 1, the traffic X2 corresponding to application 2, and the traffic Xn corresponding to application n based on the traffic between server A and server B, and finally obtain the application information marked by traffic between server A and server B; Set up another server C in addition to A and B in the host subordinates, identify the traffic X3 corresponding to application 3, the traffic X4 corresponding to application 4, and the traffic Xn corresponding to application n based on the traffic between server B and server C, and finally obtain the application information marked by traffic between server B and server C, and finally obtain the baseline model of the entire server application assets in units of sites; Deep detection module, which performs deep detection on the host through operating system probes; The correction module is used to compare the deep detection results with the identification results of the traffic analysis module, filter the identification results of non-real business applications, add new business applications to the traffic baseline, and report non-business applications and services discovered in the deep detection to the management platform.

3. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method according to claim 1 when executing the computer program.

Citation Information

Patent Citations

  • Intelligent network service identification method based on positive transfer learning

    CN104468252A

  • Method and system for detecting network security threat based on trusted business flow

    CN105141604A

  • Industrial control abnormal behavior analysis method and system based on self-learning white list

    CN108055282A

  • Monitoring method, device, equipment and medium

    CN109995555A

  • Dark web encryption application service identification method and system based on network behavior characteristics

    CN114124463A