Monitoring and analysis method and system for business call relationship

Abstract

The present invention discloses a monitoring and analysis method and system for business call relationships, the method comprising: monitoring and obtaining log information of a business system, dividing the log information into keywords; analyzing the keywords of the log information, and forming independent minimum units according to the keywords; determining the call relationship between the corresponding host and the application process according to the call relationship between the independent minimum units, so as to obtain the connection relationship between the process grids; and using a multi-dimensional visualization method to display the call relationship between the businesses corresponding to the processes. Through the technical solution of the present invention, a business log grid system in a three-dimensional space manner can be formed, so as to visually monitor and analyze the business call relationship of the business system, thereby quickly locating the fault node in the business system, quickly recovering the business fault, and providing data support for system optimization, and can be applicable to all user-oriented business systems.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a monitoring and analysis method for business call relationships and a monitoring and analysis system for business call relationships. Background Art

[0002] At present, operators' distributed systems are becoming increasingly complex. Due to the huge amount of log information, complex system architecture, and complex system call relationships, when problems arise, operation and maintenance personnel and developers are faced with huge distributed systems and have no idea where to start, cannot quickly locate the problem, and have difficulty troubleshooting. Summary of the invention

[0003] In view of the above problems, the present invention provides a monitoring and analysis method and system for business call relationships, which divides the business call log information into smaller units, minimizes each point of the log information, and determines the mutual call relationship through the minimum unit to form a three-dimensional business log grid system, thereby visually monitoring and analyzing the business call relationship of the business system, and then quickly locating the faulty nodes in the business system, quickly recovering from business failures, and providing data support for system optimization. The system is applicable to all user-oriented business systems.

[0004] To achieve the above object, the present invention provides a method for monitoring and analyzing service call relationships, comprising:

[0005] Monitor and obtain log information of business systems, and break down the log information into keywords;

[0006] Analyze the keywords of the log information, and summarize them into independent minimum units according to the keywords;

[0007] Determine the calling relationship between the corresponding host and the application process according to the calling relationship between the independent minimum units, so as to obtain the connection relationship between the process grids;

[0008] A multi-dimensional visualization method is used to display the calling relationship between the businesses corresponding to the processes.

[0009] In the above technical solution, preferably, the specific process of monitoring and obtaining log information of the business system includes:

[0010] The log system collects logs of the business system using the business system code, and obtains log information collected by the log system for the business system, so as to determine the business call relationship according to the relationship between the business system codes in the log information; or,

[0011] An agent probe is deployed in each application process of the business system, and log information in the business system is obtained by calling the agent probe, so as to determine the calling relationship between the application processes in the business system according to the log information.

[0012] In the above technical solution, preferably, the specific process of breaking down the log information into keywords, analyzing the keywords of the log information, and summarizing the keywords to form independent minimum units includes:

[0013] Decomposing the characteristic characters of the log information by using a log analysis tool;

[0014] Automatically analyze keywords based on the decomposed characteristic characters;

[0015] Keywords used to express the same attribute characteristics in a single application process are combined as independent minimum units, so as to determine the relationship between the application processes by analyzing the relationship between the independent minimum units.

[0016] In the above technical solution, preferably, the specific process of determining the calling relationship between the corresponding host and the application process according to the calling relationship between the independent minimum units includes:

[0017] Determine the call summary information of the service access process between the same host or different hosts according to the call process detailed information between the keywords in the independent minimum unit;

[0018] The call process details include the call method, call input parameters, call time, call SQL and call time consumption, and the call summary information includes the access sequence, access time, access URL and access time consumption information.

[0019] In the above technical solution, preferably, the multi-dimensional visualization method includes a call process tree diagram, a service topology diagram and a call process timeline diagram.

[0020] In the above technical solution, preferably, the monitoring and analysis method of the service call relationship also includes:

[0021] Summarize the calling relationships between services into a series of independent events based on the system resource information and role information involved;

[0022] The Bayesian analysis method is used to calculate the occurrence probability between the corresponding process grid and each independent event;

[0023] By comprehensively analyzing the occurrence probability between the process grid and independent events, the independent events corresponding to the call relationships between businesses can be inferred;

[0024] When a fault occurs between process grids, the independent event causing the fault can be quickly located based on the probability relationship between the process grid and the independent event.

[0025] The present invention further proposes a monitoring and analysis system for service call relationships, which uses a monitoring and analysis method for service call relationships disclosed in any one of the above technical solutions, including:

[0026] A log decomposition module is used to monitor and obtain log information of the business system and decompose the log information into keywords;

[0027] A log analysis module, used to analyze the keywords of the log information and summarize them into independent minimum units according to the keywords;

[0028] A call analysis module, used to determine the call relationship between the corresponding host and the application process according to the call relationship between the independent minimum units, so as to obtain the connection relationship between the process grids;

[0029] The relationship display module is used to display the calling relationship between the businesses corresponding to the processes in a multi-dimensional visual manner.

[0030] In the above technical solution, preferably, the log analysis module is specifically used for:

[0031] Decomposing the characteristic characters of the log information by using a log analysis tool;

[0032] Automatically analyze keywords based on the decomposed characteristic characters;

[0033] Keywords used to express the same attribute characteristics in a single application process are combined as independent minimum units, so as to determine the relationship between the application processes by analyzing the relationship between the independent minimum units.

[0034] In the above technical solution, preferably, the call analysis module is specifically used to:

[0035] Determine the call summary information of the service access process between the same host or different hosts according to the call process detailed information between the keywords in the independent minimum unit;

[0036] The call process details include the call method, call input parameters, call time, call SQL and call time consumption, and the call summary information includes the access sequence, access time, access URL and access time consumption information.

[0037] In the above technical solution, preferably, the monitoring and analysis system of the business call relationship further includes an event analysis module, and the event analysis module is specifically used for:

[0038] Summarize the calling relationships between services into a series of independent events based on the system resource information and role information involved;

[0039] The Bayesian analysis method is used to calculate the occurrence probability between the corresponding process grid and each independent event;

[0040] By comprehensively analyzing the occurrence probability between the process grid and independent events, the independent events corresponding to the call relationships between businesses can be inferred;

[0041] When a fault occurs between process grids, the independent event causing the fault can be quickly located based on the probability relationship between the process grid and the independent event.

[0042] Compared with the prior art, the beneficial effects of the present invention are as follows: by dividing the business call log information into smaller units, each point of the log information is minimized, and the mutual call relationship is determined through the minimum unit, forming a three-dimensional business log grid system, so as to visualize the business call relationship of the business system, thereby quickly locating the faulty nodes in the business system, quickly recovering from business failures, and providing data support for system optimization, and can be applicable to all user-oriented business systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A flowchart of a method for monitoring and analyzing service call relationships disclosed in an embodiment of the present invention;

[0044] Figure 2 A module diagram of a system for monitoring and analyzing business call relationships disclosed in an embodiment of the present invention.

[0045] In the figure, the corresponding relationship between each component and the reference numeral is as follows:

[0046] 11. Log decomposition module, 12. Log analysis module, 13. Call analysis module, 14. Relationship display module, 15. Event analysis module. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0049] like Figure 1As shown, a monitoring and analysis method for a service call relationship provided by the present invention includes:

[0050] Monitor and obtain log information of business systems, and break down log information into keywords;

[0051] Analyze the keywords of log information and summarize them into independent minimum units based on the keywords;

[0052] According to the calling relationship between the independent minimum units, the calling relationship between the corresponding host and the application process is determined to obtain the connection relationship between the process grids;

[0053] A multi-dimensional visualization method is used to display the calling relationship between the businesses corresponding to the processes.

[0054] In this implementation, the business call log information is divided into smaller units, each point of the log information is minimized, and the mutual call relationship is determined through the minimum unit to form a three-dimensional business log grid system, thereby visually monitoring and analyzing the business call relationship of the business system, thereby quickly locating the faulty nodes in the business system, quickly recovering from business failures, and providing data support for system optimization. It can also be applicable to all user-oriented business systems.

[0055] Specifically, combining the characteristics of the business system architecture and the essential characteristics of the business, a large amount of end-to-end business call relationship log information is classified and divided, relying on the system log, application log and trajectory log data obtained above, and breaking them down into smaller pieces, and reasonably establishing logical relationships between them. The main purpose is to divide the business log information into smaller pieces, minimize each point of the log information, and form a business log grid system in a three-dimensional space through mutual call relationships, which can be applied to all user-oriented business systems.

[0056] Preferably, the specific process of monitoring and obtaining log information of the business system includes:

[0057] The log system collects logs of the business system using the business system code, and obtains the log information collected by the log system for the business system, so as to determine the business call relationship according to the relationship between the business system codes in the log information; or,

[0058] Agent probes are deployed in each application process of the business system, and log information in the business system is obtained by calling the agent probes, so as to determine the calling relationship between the application processes in the business system according to the log information.

[0059] Among them, during the deployment of the agent probe, the agent collection point information includes the name of the display application, the version number of the collection agent, the ID number of the collection agent, the start time of collection, etc. The business host status information includes the IP address of the application host, the account name of the application host, the application server name, the jvm version number, the stack usage, the CPU usage, the memory usage and the system throughput.

[0060] In the above implementation, preferably, the specific process of breaking down the log information into keywords, analyzing the keywords of the log information, and summarizing the keywords to form independent minimum units includes:

[0061] Use log analysis tools to decompose the characteristic characters of log information;

[0062] Automatically analyze keywords based on the decomposed characteristic characters;

[0063] Keywords used to express the same attribute characteristics in a single application process are combined as independent minimum units, so as to determine the relationship between the application processes by analyzing the relationship between the independent minimum units.

[0064] In the above implementation, preferably, the specific process of determining the calling relationship between the corresponding host and the application process according to the calling relationship between the independent minimum units includes:

[0065] Determine the call summary information of the service access process between the same host or different hosts according to the call process detail information between the keywords in the independent minimum unit;

[0066] Among them, the details of the calling process of a single business request are displayed through an access path URL to view the detailed information of the code calling process inside the application system when accessing this URL. The calling process details include the calling method, calling input parameters, calling time, calling SQL and calling time. The access details include the summary information of the single business request call, the details of the single business request calling process and the calling process information. The access details are mainly obtained by analyzing a certain visit of the user. The call summary information includes the access sequence, access time, access URL, and access time information, which is used to display all application path file information visited by the user this time.

[0067] Among them, preferably, the multi-dimensional visualization method includes a call process tree diagram, a service topology diagram and a call process timeline diagram.

[0068] For example, use a topology diagram to display the call relationship, and use a graph to display the access time distribution and call time distribution. The business call relationship topology is used to display the call relationship between various applications and databases in the business system, and the number of mutual calls. The topology diagram uses different graphs to distinguish between users, applications, and database nodes.

[0069] In the above implementation, preferably, the monitoring and analysis method of the service call relationship further includes:

[0070] Summarize the calling relationships between services into a series of independent events based on the system resource information and role information involved;

[0071] Using Bayesian Analysis, the probability of occurrence between the corresponding process grid and each independent event is calculated;

[0072] By comprehensively analyzing the occurrence probability between the process grid and independent events, the independent events corresponding to the call relationships between businesses can be inferred;

[0073] When a fault occurs between process grids, the independent event causing the fault can be quickly located based on the probability relationship between the process grid and the independent event.

[0074] Specifically, based on historical data and the circumstances of events, the probability of occurrence of related events is predicted, and the Bayesian analysis method provides a method for calculating the probability of a hypothesis, which is based on the prior probability of the hypothesis, the probability of observing different data under the given hypothesis, and the observed data itself. The method is to combine the prior information about the unknown parameters with the sample information, and then obtain the posterior information according to the Bayesian formula, and then infer the unknown parameters based on the posterior information.

[0075] Among them, according to the log information obtained above, it is divided into small parts, and reasonable logical relationships are established between them, and divided according to the ideas of people, things, objects, and states.

[0076] People: All types of operation and maintenance-related personnel are grouped into independent roles. Each role is associated with a series of events based on the division of responsibilities. The roles are independent and inclusive.

[0077] Events: The business log analysis results are summarized into a series of independent events. Each type of event is associated with a corresponding series of resources according to the degree of resource impact. The events are independent, dependent, and inclusive.

[0078] Things: Classify and divide the system resource information in the business log according to their respective characteristics. It can be divided by IaaS, PaaS, SaaS layers, or by infrastructure, middleware, and application systems. The resource grid is divided according to the smallest monitoring unit;

[0079] Infrastructure: host, storage, network, resource pool;

[0080] Middleware: Tomcat, WebLogic, etc.

[0081] Status: A measure that reflects the health of business operation status, including technology, service, and business indicators, among which business indicators are independent and have an influencing relationship.

[0082] For example, business indicators include:

[0083] (1) Technical categories: CPU usage, memory usage, file system usage, network ping packet loss rate;

[0084] (2) Service: fault handling time, complaint handling time, and complaint handling timeliness rate;

[0085] (3) Business category: business acceptance time, business activation time, business change time, etc.

[0086] Finally, based on the association relationship, the mutual calling relationship is formed into a business log grid system using three-dimensional space.

[0087] like Figure 2 As shown, the present invention further proposes a monitoring and analysis system for business call relationships, which applies the monitoring and analysis method for business call relationships disclosed in any one of the above embodiments, including:

[0088] The log decomposition module 11 is used to monitor and obtain the log information of the business system and decompose the log information into keywords;

[0089] The log analysis module 12 is used to analyze the keywords of the log information and summarize them into independent minimum units according to the keywords;

[0090] The call analysis module 13 is used to determine the call relationship between the corresponding host and the application process according to the call relationship between the independent minimum units, so as to obtain the connection relationship between the process grids;

[0091] The relationship display module 14 is used to display the calling relationship between the services corresponding to the processes in a multi-dimensional visual manner.

[0092] In this implementation, the business call log information is divided into smaller units, each point of the log information is minimized, and the mutual call relationship is determined through the minimum unit to form a three-dimensional business log grid system, thereby visually monitoring and analyzing the business call relationship of the business system, thereby quickly locating the faulty nodes in the business system, quickly recovering from business failures, and providing data support for system optimization. It can also be applicable to all user-oriented business systems.

[0093] In the above implementation, preferably, the log analysis module 12 is specifically used for:

[0094] Use log analysis tools to decompose the characteristic characters of log information;

[0095] Automatically analyze keywords based on the decomposed characteristic characters;

[0096] Keywords used to express the same attribute characteristics in a single application process are combined as independent minimum units, so as to determine the relationship between the application processes by analyzing the relationship between the independent minimum units.

[0097] In the above implementation, preferably, the call analysis module 13 is specifically used for:

[0098] Determine the call summary information of the service access process between the same host or different hosts according to the call process detail information between the keywords in the independent minimum unit;

[0099] The call process details include the call method, call input parameters, call time, call SQL and call duration, and the call summary information includes the access sequence, access time, access URL and access duration information.

[0100] In the above implementation, preferably, the monitoring and analysis system of the business call relationship further includes an event analysis module 15, and the event analysis module 15 is specifically used for:

[0101] Summarize the calling relationships between services into a series of independent events based on the system resource information and role information involved;

[0102] The Bayesian analysis method is used to calculate the occurrence probability between the corresponding process grid and each independent event;

[0103] By comprehensively analyzing the occurrence probability between the process grid and independent events, the independent events corresponding to the call relationships between businesses can be inferred;

[0104] When a fault occurs between process grids, the independent event causing the fault can be quickly located based on the probability relationship between the process grid and the independent event.

[0105] According to the monitoring and analysis system of business call relationships disclosed in the above implementation mode, the functions to be implemented by each module thereof correspond to the steps of the monitoring and analysis method of business call relationships disclosed in the above implementation mode, respectively. The specific implementation process refers to the above implementation mode and will not be repeated here.

[0106] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for monitoring and analyzing business call relationships, characterized in that: include: Monitor and obtain log information of business systems, and break down the log information into keywords; Analyze the keywords of the log information, and summarize them into independent minimum units according to the keywords; Determine the calling relationship between the corresponding host and the application process according to the calling relationship between the independent minimum units, so as to obtain the connection relationship between the process grids; Use multi-dimensional visualization to show the calling relationship between the businesses corresponding to the processes; Summarize the calling relationships between services into a series of independent events based on the system resource information and role information involved; The Bayesian analysis method is used to calculate the occurrence probability between the corresponding process grid and each independent event; By comprehensively analyzing the occurrence probability between the process grid and independent events, the independent events corresponding to the call relationships between businesses can be inferred; When a fault occurs between process grids, the independent event causing the fault can be quickly located based on the probability relationship between the process grid and the independent event.

2. The monitoring and analysis method of business call relationship according to claim 1 is characterized in that: The specific process of monitoring and obtaining log information of the business system includes: The log system collects logs of the business system using the business system code, and obtains log information collected by the log system for the business system, so as to determine the business call relationship according to the relationship between the business system codes in the log information; or, An agent probe is deployed in each application process of the business system, and log information in the business system is obtained by calling the agent probe, so as to determine the calling relationship between the application processes in the business system according to the log information.

3. The monitoring and analysis method of business call relationship according to claim 2 is characterized in that: The specific process of breaking down the log information into keywords, analyzing the keywords of the log information, and summarizing the keywords to form independent minimum units includes: Decomposing the characteristic characters of the log information by using a log analysis tool; Automatically analyze keywords based on the decomposed characteristic characters; Keywords used to express the same attribute characteristics in a single application process are combined as independent minimum units, so as to determine the relationship between the application processes by analyzing the relationship between the independent minimum units.

4. The monitoring and analysis method of business call relationship according to claim 3 is characterized in that: The specific process of determining the calling relationship between the corresponding host and the application process according to the calling relationship between the independent minimum units includes: Determine the call summary information of the service access process between the same host or different hosts according to the call process detailed information between the keywords in the independent minimum unit; The call process detailed information includes the call method, call input parameters, call time, call SQL and call time consumption, and the call summary information includes the access sequence, access time, access URL and access time consumption information.

5. The monitoring and analysis method of business call relationship according to claim 1 is characterized in that: The multi-dimensional visualization method includes a call process tree diagram, a service topology diagram and a call process timeline diagram.

6. A monitoring and analysis system for business call relationships, characterized in that: The monitoring and analysis method of the business call relationship according to any one of claims 1 to 5 is applied, comprising: A log decomposition module is used to monitor and obtain log information of the business system and decompose the log information into keywords; A log analysis module, used to analyze the keywords of the log information and summarize them into independent minimum units according to the keywords; A call analysis module, used to determine the call relationship between the corresponding host and the application process according to the call relationship between the independent minimum units, so as to obtain the connection relationship between the process grids; The relationship display module is used to display the calling relationship between the services corresponding to the processes in a multi-dimensional visual way; Event analysis module, specifically used for: Summarize the calling relationships between services into a series of independent events based on the system resource information and role information involved; The Bayesian analysis method is used to calculate the occurrence probability between the corresponding process grid and each independent event; By comprehensively analyzing the occurrence probability between the process grid and independent events, the independent events corresponding to the call relationships between businesses can be inferred; When a fault occurs between process grids, the independent event causing the fault can be quickly located based on the probability relationship between the process grid and the independent event.

7. The monitoring and analysis system for business call relationships according to claim 6, characterized in that: The log analysis module is specifically used for: Decomposing the characteristic characters of the log information by using a log analysis tool; Automatically analyze keywords based on the decomposed characteristic characters; Keywords used to express the same attribute characteristics in a single application process are combined as independent minimum units, so as to determine the relationship between the application processes by analyzing the relationship between the independent minimum units.

8. The monitoring and analysis system for business call relationships according to claim 6, characterized in that: The call analysis module is specifically used for: Determine the call summary information of the service access process between the same host or different hosts according to the call process detailed information between the keywords in the independent minimum unit; The call process detailed information includes the call method, call input parameters, call time, call SQL and call time consumption, and the call summary information includes the access sequence, access time, access URL and access time consumption information.

Images