Application topological graph drawing device for mixed environment

Abstract

The invention discloses an application topological graph drawing device for a mixed environment, and the device comprises a collection module which collects business data and generates a TCP message;the identification module is used for identifying sources and destinations of the TCP messages transmitted by the acquisition module and acquiring IP addresses and port information corresponding to the TCP messages; the operation analysis module is used for analyzing the corresponding relationship between the IP address and the port according to the TCP connection; the processing module is used for confirming the corresponding application process according to the IP address and the port, acquiring the name of the actually called application module and confirming the calling relationship between the application modules; and the automatic association module calls a drawing function to draw a call chain relation graph of the application module. According to the invention, the rule base is added for automatic matching, and redundant data transmission is reduced; tCP messages are automatically identified, and an automatic analysis process is realized; and realizing quick data cleaning and matching with a CMDB library, and generating a call chain application relationship topological graph.

Description

technical field [0001] The invention relates to a device for drawing an application topology map, in particular to a device for drawing an application topology map for a mixed environment. Background technique [0002] Due to the popularization of virtualization and containerization, various applications of IT systems have been migrated to virtualized and containerized environments, and various business systems are constantly updated and updated, and the architecture is continuously innovated and optimized, which makes the system module call chain relationship more complicated. It is more difficult to use links to identify, which makes it impossible for module-related IT operation and maintenance personnel to comprehensively analyze the application exceptions of specific links. However, the measurement and display capabilities of automatic application topology are more prominent, which is more conducive to identifying the application module where the system failure is located...

AI Technical Summary

Problems solved by technology

[0009] In the acquisition of the network traffic mirroring method, the service side with a huge amount of storage is required. Based on the analysis of the business flow, the analysis focuses on the traffic, but not reflected in the association relationship and topology data aggregation on the application side, and deploys application probes to obtain IP ports. Interaction of TCP messages causes resource occupation. This solution is insufficient in data filtering, and consumes a lot of storage machine and probe resources. However, the drawn network topology diagram does not involve the call relationship of the application side, and cannot analyze the scaling topology relationship list.

[0010] Based on the intricate invocation methods of minicomputers, clustering, containerization, cross-region, modularization, and microservices, a large number of traditional IT and new IT applications are integrated and new services are constantly expanding, and the trend of segmented and modularized operation and maintenance is increasing. Some solutions are difficult to control the application call chain of all system nodes, resulting in missing links and temporarily added links that cannot be identified in real time. The above three methods mainly have the following shortcomings:

[0011] Solution 1 lacks an automatic identification application process, and cannot locate service failure nodes. Solution 1 mainly relies on TCP packet detection, which can identify network relationship calls, and can also draw network relationships on the switch and router sides. However, when an application module passes load balancing N applications appear on the end, this solution cannot automatically identify the server calling the application, let alone the detection of application failures

[0012] Solution 2 cannot simplify and identify specific application modules in a mixed environment. Solution 2 manages network nodes through the SNMP protocol, relies on network card side and fixed port monitoring and provides data analysis. For example, there are multiple different application deployments in a cluster or container, but internal Call relationship, this solution is difficult to identify whether it is the same application in such a complex environment and analyze the specific topology relationship;

[0013] Solution 3 has insufficient filtering and analysis capabilities for real-time message information, which cannot satisfy various application business systems. Solution 3 uses mirroring traffic, but with huge traffic analysis, more machine resources need to be provided for storage, and there is a lack of filtering and analysis for discarding redundant data. process, it is impossible to quickly and accurately identify key call chain information on the service side of a huge business group

[0014] There are various deficiencies in the existing technical solutions, which cannot meet the application analysis requirements of huge IT application systems, cannot fully control business production failures, lack intuitive detection of abnormal point of application failures, and are inconvenient to expand and enhance other functions. Therefore, this The proposal proposes an intelligent application topology map drawing device that is suitable for mixed environments. After analyzing TCP messages, it filters and aggregates useful information to realize intelligent drawing of application topology maps. At the same time, it introduces function extensions such as log analysis result sets and capacity control. Realize function enhancement based on application topology to improve operation and maintenance efficiency

Images