High-performance distributed data center monitoring architecture
Patent Information
- Application Number
- CN202411059652.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-05
- Publication Date
- 2024-12-06
AI Technical Summary
Traditional data center monitoring architectures struggle to efficiently monitor large-scale data centers in a timely manner, leading to increased monitoring delays and reduced performance.
A high-performance distributed data center monitoring architecture utilizing a multi-threaded monitoring system with multi-mode compatible adapters, central control modules, data analysis, and notification modules to dynamically expand and improve monitoring efficiency and accuracy.
The system reduces monitoring delays, enhances system throughput, and enables real-time monitoring, allowing for dynamic expansion and intelligent analysis of data centers, providing intuitive feedback and alarm prompts for timely countermeasures.
Smart Images

Figure CN119088637A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of distributed data monitoring technology, and in particular to a high-performance distributed data center monitoring architecture. Background Art
[0002] A distributed data center refers to a data center that distributes the data collection, processing, and analysis processes across multiple geographical locations to achieve higher availability, reliability, and efficiency. In recent years, with the rapid development of cloud computing, it has become a frontier topic in the field of information technology research. Large-scale cloud data centers have also set off a construction boom around the world, and the monitoring of cloud data centers has gradually attracted people's attention. At present, the monitoring capabilities of traditional data center monitoring architectures are limited. When the scale of data centers increases, it is impossible to better monitor the data centers in a timely manner. Therefore, the present invention proposes a high-performance distributed data center monitoring architecture, which uses a distributed multi-threaded monitoring architecture to monitor data centers through multi-mode compatible adapters. It can not only reduce the monitoring acquisition delay and improve the efficiency of monitoring acquisition, but also dynamically expand the scale of data centers, effectively improving the performance of the distributed data center monitoring architecture. Summary of the invention
[0003] The purpose of the present invention is to provide a high-performance distributed data center monitoring architecture to solve the problems raised in the above background technology.
[0004] To achieve the above object, the present invention provides the following technical solutions: a high-performance distributed data center monitoring architecture, including: a monitoring and acquisition module, a central control module, a data analysis module and a notification and warning module;
[0005] The monitoring and acquisition module is used to use a distributed multi-threaded monitoring architecture to use a multi-mode compatible adapter to perform monitoring access on the monitoring object and obtain monitoring and acquisition information;
[0006] The central control module is used to determine the monitoring acquisition control information based on the polling device and transmit the monitoring acquisition control information to the monitoring acquisition module;
[0007] The data analysis module is used to perform data aggregation and alarm analysis based on the monitoring and collection information to obtain data analysis results;
[0008] The notification and warning module is used to perform graphical output presentation of results and multi-method alarm prompts based on the data analysis results.
[0009] Furthermore, the monitoring and acquisition module includes multiple monitoring machines, and a distributed multi-threaded monitoring architecture between the monitoring and acquisition module and the data center is established through the monitoring machines according to a one-to-many connection rule, and multi-mode compatible adapters are respectively set in the monitoring machines. The monitoring machines are adapted to the monitored objects in the data center through the multi-mode compatible adapters.
[0010] Furthermore, the multi-mode compatible adapter includes multiple monitoring plug-ins, and when the monitoring machine is adapted to the monitored object in the data center through the multi-mode compatible adapter, it includes:
[0011] Monitor the data center, identify the monitored objects in the data center, and obtain information about the monitored objects;
[0012] Determine the type of the monitored object according to the information of the monitored object;
[0013] Performing a first monitoring plug-in match according to the type of the monitored object to obtain a first matching result;
[0014] Combine the monitored object information to perform a second monitoring plug-in match in the first matching result to determine the target monitoring plug-in;
[0015] The monitored object is connected via the target monitoring plug-in to establish a monitoring relationship between the monitored object and the monitoring machine.
[0016] Furthermore, the central control module includes: a host computer unit, a central processing unit and a control transmission unit;
[0017] The host computer unit is used to obtain control instruction information;
[0018] The central processing unit is used to analyze the control indication information and determine the monitoring collection control information based on the polling device according to the analysis result of the control indication information;
[0019] The control transmission unit is used to transmit the monitoring collection control information obtained by the central processing unit in real time by adopting a multi-link parallel processing method.
[0020] Furthermore, the poller includes: a passive monitoring poller and an active monitoring poller, and the central processing unit analyzes the control indication information, and determines the monitoring collection control information based on the poller according to the analysis result of the control indication information, including:
[0021] Analyze whether the control instruction information is quick control instruction information, and determine the analysis and judgment result;
[0022] Performing standardization processing on the control instruction information according to the analysis and judgment results to obtain the standardized control instruction information;
[0023] Extracting a first feature from the control instruction information or the standardized control instruction information to obtain first feature information of the control instruction;
[0024] Identify the first characteristic information of the control indication, determine whether the control indication information is a control indication for a passive monitoring poller or a control indication for an active monitoring poller, obtain a first characteristic identification result of the control indication, and perform control information set matching according to the first characteristic identification result of the control indication to obtain a first control matching result;
[0025] The control indication information or the standardized control indication information is parsed to obtain the second characteristic information of the control indication, and the control information is configured according to the second characteristic information of the control indication in the first control matching result to obtain the monitoring collection control information.
[0026] Furthermore, the monitoring and acquisition module includes: a monitoring and acquisition unit and a data export unit; the monitoring and acquisition unit is used to perform target matching on the monitoring and acquisition control information, and input the monitoring and acquisition control information into the passive monitoring poller or the active monitoring poller according to the matching result, and then monitor the data center in the passive monitoring poller or the active monitoring poller based on the distributed multi-threaded monitoring architecture through a multi-mode compatible adapter to obtain the monitoring and acquisition information; the data export unit is used to perform data preprocessing on the monitoring and acquisition information, and import the preprocessed monitoring and acquisition information into the data analysis module.
[0027] Furthermore, the distributed data center monitoring architecture also includes: a query module; the query module is connected to the data analysis module, and when the query module obtains query information, the data analysis module performs a target data query based on the query information, obtains the query result, and feeds back the query result through the query module.
[0028] Furthermore, the data analysis module includes: a real-time receiving unit, a data storage unit, a scheduling query unit and an analysis processing unit;
[0029] The real-time receiving unit is used to receive the monitoring and collecting information, and import the monitoring and collecting information into the data storage unit and the analysis and processing unit respectively;
[0030] The data storage unit is used to store monitoring and collection information;
[0031] The analysis and processing unit is used to combine the monitoring and collection information with the historical monitoring and collection information in the data storage unit to perform data aggregation and alarm analysis to obtain data analysis results;
[0032] The scheduling query unit is used to determine a query target according to the query information, make a query request for the query target, obtain query information based on the query request, and then organize the query information to obtain a query result.
[0033] Furthermore, when the analysis and processing unit combines the monitoring and collection information with the historical monitoring and collection information in the data storage unit for data aggregation and alarm analysis, it includes:
[0034] Perform a preliminary analysis on the monitoring and collection information to determine whether there is any indicator monitoring and collection information exceeding the indicator threshold in the monitoring and collection information, and obtain a first judgment result of the analysis and processing;
[0035] Retrieving historical monitoring and collection information from the data storage unit according to the indicators of the monitoring and collection information, and obtaining historical monitoring data of the indicators;
[0036] Aggregate the monitoring and collection information according to the indicators and the historical monitoring data of the indicators to obtain the indicator monitoring and collection summary information, and at the same time predict the data fluctuations of the historical monitoring data of the indicators to obtain the indicator prediction data, and perform abnormal analysis and judgment on the monitoring and collection information according to the indicator prediction data to obtain the second judgment result of the analysis and processing;
[0037] An alarm analysis is performed based on the analysis and processing of the first judgment result or the analysis and processing of the second judgment result to determine whether to issue an alarm prompt and obtain an alarm analysis result.
[0038] Furthermore, the notification and warning module includes: a graphic output unit, a management and recording unit, and an alarm notification unit;
[0039] The image output unit is used to perform graphical conversion on the data analysis results and output the data analysis results in the form of graphics;
[0040] The management and recording unit is used to update the log records according to the data analysis results;
[0041] The alarm notification unit is used to use multiple means to issue an alarm when the alarm analysis result shows that an abnormality exists according to the data analysis result.
[0042] The present invention realizes the monitoring of the data center, and adopts a distributed multi-threaded monitoring architecture so that the monitoring and acquisition module can acquire monitoring and acquisition information in parallel with multiple threads, thereby improving the utilization rate of the monitoring and acquisition module and the system throughput, reducing the monitoring and acquisition delay, improving the efficiency of monitoring and acquisition, and ensuring the real-time nature of the monitoring and acquisition information.
[0043] The present invention realizes direct association with the data center through a multi-mode compatible adapter in the monitoring and acquisition module, realizes dynamic expansion of the monitoring protocol or the custom monitoring interface, so that when the scale of the data center changes, it can also be dynamically expanded according to the scale of the data center, effectively improving the performance of the distributed data center monitoring architecture, and realizes intelligent analysis of the monitoring and acquisition information through the data analysis module, which can not only discover the abnormal phenomena of the data center, but also realize the organization of the monitoring and acquisition information, providing convenience for relevant personnel. In addition, with the help of the notification and warning module, not only can the relevant personnel understand the status of the data center more intuitively, but also can improve the vigilance of the relevant personnel through alarm prompts when abnormal phenomena exist in the data center, so that the relevant personnel can adopt countermeasures based on the alarm prompts, thereby reducing the impact of the data center anomalies on the normal use of the data center.
[0044] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.
[0045] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0047] Figure 1 A schematic diagram of the distributed data center monitoring architecture of the present invention;
[0048] Figure 2 A schematic diagram of the steps of adapting the monitoring machine and the data center in the distributed data center monitoring architecture of the present invention;
[0049] Figure 3 A specific schematic diagram of a central control module of a distributed data center monitoring architecture according to the present invention;
[0050] Figure 4 A schematic diagram of the steps of the central processing unit of the central control module in the distributed data center monitoring architecture of the present invention;
[0051] Figure 5 A specific schematic diagram of the monitoring and acquisition module of the distributed data center monitoring architecture of the present invention;
[0052] Figure 6 It is another schematic diagram of the distributed data center monitoring architecture of the present invention;
[0053] Figure 7 A specific schematic diagram of a data analysis module in the distributed data center monitoring architecture of the present invention;
[0054] Figure 8 A schematic diagram of the steps of the analysis and processing unit of the data analysis module in the distributed data center monitoring architecture of the present invention;
[0055] Fig. 9 It is a specific schematic diagram of the notification and warning module in the distributed data center monitoring architecture described in the present invention. DETAILED DESCRIPTION
[0056] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.
[0057] like Figure 1 As shown, the embodiment of the present invention provides a high-performance distributed data center monitoring architecture, including: a monitoring and acquisition module, a central control module, a data analysis module and a notification and warning module;
[0058] The monitoring and acquisition module is used to use a distributed multi-threaded monitoring architecture to monitor and access the data center using a multi-mode compatible adapter to obtain monitoring and acquisition information;
[0059] The central control module is used to determine the monitoring acquisition control information based on the polling device and transmit the monitoring acquisition control information to the monitoring acquisition module;
[0060] The data analysis module is used to perform data aggregation and alarm analysis based on the monitoring and collection information to obtain data analysis results;
[0061] The notification and warning module is used to perform graphical output presentation of results and multi-method alarm prompts based on the data analysis results.
[0062] In the above technical solution, the monitoring and acquisition module is control-connected to the central control module, the data analysis module is control-connected to both the central control module and the notification and warning module, and the monitoring and acquisition module is also connected to the data analysis module.
[0063] When the high-performance distributed data center monitoring architecture in the above technical solution monitors the data center, the central control module determines the monitoring and acquisition control information based on the poller, and transmits the monitoring and acquisition control information to the monitoring and acquisition module. The monitoring and acquisition module realizes direct association with the data center through a multi-mode compatible adapter, and uses a distributed multi-threaded monitoring architecture based on the association relationship of the monitoring and acquisition control information to monitor and access the data center to obtain the monitoring and acquisition information; then the data analysis module summarizes the data and performs alarm analysis on the monitoring and acquisition information to obtain the data analysis results, and finally outputs the data analysis results in the form of graphical results by informing the early warning module, and also uses multiple means of alarm prompts based on the data analysis results.
[0064] In the above technical solution, the data center includes: hardware resources and software resources, such as: various servers, storage, network equipment, databases and application services, etc.
[0065] In the above technical solution, the poller includes: an active monitoring poller and a passive monitoring poller.
[0066] In the above technical solution, the multi-mode compatible adapter implements IPMI protocol, SNMP protocol, database and application service monitoring plug-ins, etc. to obtain monitoring data by various commands and simulated access methods, and also supports plug-in integration, allowing custom protocols, third-party plug-ins or some future protocols to be seamlessly integrated into the adapter.
[0067] The above technical solution realizes the monitoring of the data center. The distributed multi-threaded monitoring architecture is adopted so that the monitoring and acquisition module can acquire monitoring and acquisition information in parallel through multiple threads, thereby improving the utilization rate and system throughput of the monitoring and acquisition module, reducing the monitoring and acquisition delay, improving the efficiency of monitoring and acquisition, and ensuring the real-time nature of the monitoring and acquisition information. In addition, in the monitoring and acquisition module, a direct association with the data center is realized through a multi-mode compatible adapter, and the dynamic expansion of the monitoring protocol or the custom monitoring interface is realized, so that when the scale of the data center changes, it can also be dynamically expanded according to the scale of the data center, effectively improving the performance of the distributed data center monitoring architecture, and realizing intelligent analysis of the monitoring and acquisition information through the data analysis module, which can not only discover the abnormal phenomena of the data center, but also realize the organization of the monitoring and acquisition information, providing convenience for relevant personnel. In addition, with the help of the notification and warning module, not only can the relevant personnel understand the status of the data center more intuitively, but also can improve the vigilance of the relevant personnel through alarm prompts when there are abnormal phenomena in the data center, so that the relevant personnel can adopt countermeasures based on the alarm prompts, thereby reducing the impact of the data center abnormalities on the normal use of the data center.
[0068] In one embodiment provided by the present invention, the monitoring and acquisition module includes multiple monitoring machines, and a distributed multi-threaded monitoring architecture between the monitoring and acquisition module and the data center is established through the monitoring machines according to a one-to-many connection rule, and multi-mode compatible adapters are respectively set in the monitoring machines. The monitoring machines are adapted to the monitored objects in the data center through the multi-mode compatible adapters.
[0069] In the above technical solution, the monitored objects in the data center are various servers, storage, network equipment, databases and application services.
[0070] In the above technical solution, the one-to-many connection rule is: one monitoring machine monitors multiple monitored objects, and one monitored object is connected to at least two monitoring machines.
[0071] The above technical solution builds a distributed multi-threaded monitoring architecture so that the monitoring and acquisition module can synchronously acquire monitoring and acquisition information in parallel when monitoring the data center, and adopts a one-to-many connection rule so that the monitoring of the monitored object will not cause conflicts, reduce the monitoring and acquisition delay, improve the efficiency of monitoring and acquisition, and ensure the performance of the monitoring and acquisition module. In addition, the monitoring machine is connected to the monitored object through a multi-mode compatible adapter, which is compatible with multiple modes, so that the monitored object can be accessed in any situation, and the dynamic expansion of the monitoring protocol or the custom monitoring interface can be realized, so that when the scale of the data center changes, it can also be dynamically expanded according to the scale of the data center, effectively improving the performance of the distributed data center monitoring architecture.
[0072] like Figure 2 As shown, in one embodiment provided by the present invention, the multi-mode compatible adapter includes multiple monitoring plug-ins, and when the monitoring machine is adapted to the monitored object in the data center through the multi-mode compatible adapter, it includes:
[0073] A1. Monitor the data center, identify the monitored objects in the data center, and obtain information about the monitored objects;
[0074] A2. Determine the type of the monitored object based on the information of the monitored object;
[0075] A3. Perform a first monitoring plug-in match according to the type of the monitored object to obtain a first matching result;
[0076] A4. Perform a second monitoring plug-in match in the first matching result based on the monitored object information to determine the target monitoring plug-in;
[0077] A5. Connect the monitored object through the target monitoring plug-in to establish a monitoring relationship between the monitored object and the monitoring machine.
[0078] In the above technical solution, the monitoring plug-in is divided into hardware monitoring plug-in and software monitoring plug-in according to the different objects and protocols, specifically including: server hardware monitoring plug-in based on IMPI protocol, network equipment monitoring plug-in based on SNMP protocol, operating system resource monitoring plug-in based on SNMP protocol, various database application operation monitoring plug-ins, various Web application server operation status monitoring plug-ins, etc.
[0079] In the above technical solution, the types of monitored objects include: hardware and software.
[0080] In the above technical solution, the monitored object information includes the type, attributes, communication characteristics, etc. of the monitored object.
[0081] The above-mentioned technical scheme monitors the data center so that when a new monitored object is added, the data center can promptly establish a monitoring relationship between the monitoring machines for the monitored object, thereby realizing the monitoring of the monitored object, and then realizing the dynamic expansion of the monitoring protocol or the custom monitoring interface, so that when the scale of the data center changes, it can also be dynamically expanded according to the scale of the data center, thereby improving the performance of the distributed data center monitoring architecture, and when the monitoring machine is adapted to the monitored object in the data center through a multi-mode compatible adapter, the monitored object is matched step by step by matching the first monitoring plug-in and the second monitoring plug-in, which is not only convenient and easy to do, but also does not require more time to establish the monitoring relationship and is not prone to errors.
[0082] like Figure 3 As shown, in one embodiment provided by the present invention, the central control module includes: a host computer unit, a central processing unit and a control transmission unit;
[0083] The host computer unit is used to obtain control instruction information;
[0084] The central processing unit is used to analyze the control indication information and determine the monitoring collection control information based on the polling device according to the analysis result of the control indication information;
[0085] The control transmission unit is used to transmit the monitoring collection control information obtained by the central processing unit in real time by adopting a multi-link parallel processing method.
[0086] In the above technical solution, the control instruction information includes: a control start instruction, a control end instruction and control requirement information.
[0087] In the above technical solution, the host computer unit is connected to the central processing unit, and the central processing unit is connected to the monitoring acquisition module through the control transmission unit.
[0088] In the above technical scheme, the host computer unit is divided into a first acquisition subunit and a second acquisition subunit; wherein, the first acquisition subunit is used to directly obtain control requirement information, and the second acquisition subunit includes: a control start indication shortcut key, a control end indication shortcut key, and a node control shortcut key based on the monitoring process.
[0089] The above technical solution realizes the control of the monitoring and acquisition module through the central control module, so that the monitoring and acquisition module can be controlled according to the control instruction information, so as to better monitor the data center, and through the upper computer unit, not only the start and end monitoring and control of the data center monitoring can be realized, but also monitoring can be carried out in combination with people's control needs, so as to meet people's monitoring intentions for the data center and improve people's satisfaction with the data center monitoring, and the control instruction information is converted through the central processing unit, and the control instruction information is more efficiently converted into monitoring and acquisition control information, which provides convenience for the control of the monitoring and acquisition module. In addition, the control transmission unit can accurately transmit the monitoring and acquisition control information to the monitoring and acquisition module to realize precise control of the monitoring and acquisition module.
[0090] like Figure 4 As shown, in one embodiment provided by the present invention, the poller includes: a passive monitoring poller and an active monitoring poller, the central processing unit analyzes the control indication information, and determines the monitoring collection control information based on the poller according to the analysis result of the control indication information, including:
[0091] B1. Analyze whether the control instruction information is quick control instruction information and determine the analysis and judgment result;
[0092] B2. Performing standardization processing on the control instruction information according to the analysis and judgment results to obtain the standardized control instruction information;
[0093] B3. Extracting a first feature from the control instruction information or the standardized control instruction information to obtain first feature information of the control instruction;
[0094] B4, identifying the first characteristic information of the control instruction, determining whether the control instruction information is a control instruction for a passive monitoring poller or a control instruction for an active monitoring poller, obtaining a first characteristic identification result of the control instruction, and matching the control information set according to the first characteristic identification result of the control instruction to obtain a first control matching result;
[0095] B5. Analyze the control instruction information or the standardized control instruction information, obtain the second characteristic information of the control instruction, and configure the control information according to the second characteristic information of the control instruction in the first control matching result to obtain the monitoring collection control information.
[0096] In the above technical solution, the quick control instruction information refers to the control instruction information obtained by the second acquisition subunit in the host computer unit.
[0097] In the above technical solution, when the control instruction information is standardized according to the analysis and judgment result, when the analysis and judgment result is that the control instruction information is not quick control instruction information, the control instruction information is standardized; when the analysis and judgment result is that the control instruction information is quick control instruction information, there is no need to standardize the control instruction information, and the first feature extraction can be directly performed on the control instruction information.
[0098] In the above technical solution, the first feature refers to the feature in the control indication information that distinguishes the passive monitoring poller from the active monitoring poller. The second feature refers to the control detail information in the control indication information, such as: local monitoring control, monitoring of a single attribute, etc.
[0099] In the above technical scheme, when performing the first feature extraction on the control indication information or the control indication information after standardization, feature extraction is performed on the control indication information or the control indication information after standardization based on the feature extraction template, and the information at the specified position in the control indication information or the control indication information after standardization is screened out to obtain preliminary extracted feature information and filtered control indication information, and then the filtered control indication information is reviewed to analyze and determine whether the first feature-related information exists in the filtered control indication information. When the first feature-related information exists in the filtered control indication information, the first feature-related information is extracted and merged into the preliminary extracted feature information. At the same time, the preliminary extracted feature information is checked to analyze whether there is irrelevant redundant information in the preliminary feature extraction information. When there is irrelevant redundant information in the preliminary feature extraction information, the irrelevant redundant information is eliminated to obtain the first feature information of the control indication.
[0100] In the above technical solution, control information sets based on passive monitoring poller or active monitoring poller identification are formed in advance for passive monitoring poller and active monitoring poller respectively, and control instruction templates based on passive monitoring poller or active monitoring poller are pre-stored in the control information set.
[0101] In the above technical solution, when the control information is configured according to the second characteristic information of the control indication in the first control matching result, the main characteristic information is screened and determined in the second characteristic information of the control indication; then the control instructions are screened according to the main characteristic information to obtain the matching screening results, and then the control instructions obtained from the matching screening results are combined with the second characteristic information of the control indication to fill in and improve the control instructions to obtain complete monitoring and acquisition control information.
[0102] In the above technical solution, the passive monitoring poller and the active monitoring poller are arranged in the monitoring collection module.
[0103] In the above technical solution, the control transmission unit includes: a transmission identification unit, a first transmission unit and a second transmission unit, wherein when the transmission identification unit is for monitoring and acquisition control information, the transmission identification unit performs target identification on the monitoring and acquisition control information, determines whether the monitoring and acquisition control information is based on a control instruction of a passive monitoring poller or a control instruction of an active monitoring poller, obtains a transmission identification result, and then transmits the monitoring and acquisition control information to the monitoring and acquisition module through the first transmission unit according to the transmission identification result to perform monitoring and acquisition based on a passive monitoring poller, and transmits the monitoring and acquisition control information to the monitoring and acquisition module through the second transmission unit to perform monitoring and acquisition based on an active monitoring poller. Moreover, the first transmission unit and the second transmission unit are independent of each other, and while the first transmission unit is transmitting the monitoring and acquisition control information, the second transmission unit can also synchronously transmit different monitoring and acquisition control information.
[0104] The above technical solution realizes the determination of monitoring and acquisition control information through the central processing unit, so that the control indication information can be converted into monitoring and acquisition control information, so as to better monitor data for the data center, and through standardized processing, the control indication information obtained by the first acquisition unit in the upper computer unit is standardized, reducing the interference of irrelevant information, providing convenience for subsequent steps, not only ensuring the efficiency of subsequent steps, but also reducing the error probability of subsequent steps, and through the first feature extraction and the second feature analysis, it is possible to screen, match and configure step by step according to the first feature recognition result and the second feature information of the control indication, thereby realizing the determination of monitoring and acquisition control information.
[0105] like Figure 5 As shown, in one embodiment provided by the present invention, the monitoring acquisition module includes: a monitoring acquisition unit and a data export unit; the monitoring acquisition unit is used to perform target matching on the monitoring acquisition control information, and input the monitoring acquisition control information into the passive monitoring poller or the active monitoring poller according to the matching result, and then monitor the data center in the passive monitoring poller or the active monitoring poller based on the distributed multi-threaded monitoring architecture through a multi-mode compatible adapter to obtain the monitoring acquisition information; the data export unit is used to perform data preprocessing on the monitoring acquisition information, and import the preprocessed monitoring acquisition information into the data analysis module.
[0106] In the above technical solution, the monitoring collection unit is connected to the data export unit. After the monitoring collection unit obtains the monitoring collection information, the data export unit directly processes and exports the monitoring collection information obtained by the monitoring collection unit.
[0107] In the above technical solution, data preprocessing includes: data information optimization, data format standardization, transmission processing, etc.
[0108] In the above technical scheme, when the passive monitoring poller or the active monitoring poller monitors the data center through a multi-mode compatible adapter based on a distributed multi-threaded monitoring architecture according to the monitoring and acquisition control information, when the monitoring and acquisition control information is a control instruction of the passive monitoring poller, the monitoring and acquisition unit imports the monitoring and acquisition control information into the passive monitoring poller, thereby performing a control response according to the monitoring and acquisition control information in the passive monitoring poller, and realizing that the passive monitoring poller performs monitoring access to the monitoring object based on the distributed multi-threaded monitoring architecture; when the monitoring and acquisition control information is a control instruction of the active monitoring poller, the monitoring and acquisition unit imports the monitoring and acquisition control information into the active monitoring poller, thereby performing a control response according to the monitoring and acquisition control information in the active monitoring poller, and realizing that the active monitoring poller performs monitoring access to the monitoring object based on the distributed multi-threaded monitoring architecture.
[0109] The above technical solution monitors the data center according to the monitoring and acquisition control information through the monitoring and acquisition module, and also monitors the data center through a multi-mode compatible adapter based on a distributed multi-threaded monitoring architecture in the monitoring and acquisition unit, thereby improving the performance of the monitoring and acquisition unit, and processing and exporting data through the data export unit, thereby reducing the redundancy of interference information in the monitoring and acquisition information and ensuring the export efficiency and effectiveness of the monitoring and acquisition information.
[0110] like Figure 6 As shown, in one embodiment provided by the present invention, the distributed data center monitoring architecture also includes: a query module; the query module is connected to the data analysis module, and when the query module obtains query information, the data analysis module performs a target data query based on the query information, obtains the query result, and feeds back the query result through the query module.
[0111] In the above technical solution, when the query module obtains the query information, it obtains the status query request or data viewing request of the relevant personnel for any component in the distributed data center monitoring architecture.
[0112] In the above technical solution, when the query result is fed back through the query module, the query result is directly presented in the query module, thereby realizing the feedback of the query result.
[0113] The above technical solution realizes the query of the progress and data of any module in the distributed data center monitoring architecture through the query module, so that relevant personnel can understand the status and data of the distributed data center monitoring architecture of the data center at any time, improves the interactivity of the distributed data center monitoring architecture and external connections, and provides convenience for relevant personnel.
[0114] like Figure 7 As shown, in one embodiment provided by the present invention, the data analysis module includes: a real-time receiving unit, a data storage unit, a scheduling query unit and an analysis processing unit;
[0115] The real-time receiving unit is used to receive the monitoring and collecting information, and import the monitoring and collecting information into the data storage unit and the analysis and processing unit respectively;
[0116] The data storage unit is used to store monitoring and collection information;
[0117] The analysis and processing unit is used to combine the monitoring and collection information with the historical monitoring and collection information in the data storage unit to perform data aggregation and alarm analysis to obtain data analysis results;
[0118] The scheduling query unit is used to determine a query target according to the query information, make a query request for the query target, obtain query information based on the query request, and then organize the query information to obtain a query result.
[0119] In the above technical solution, the real-time receiving unit is connected to the data storage unit and the analysis and processing unit, and the analysis and processing unit is also connected to the data storage unit.
[0120] In the above technical solution, the operation of the scheduling query unit and the analysis and processing unit do not interfere with each other, and the data aggregation and alarm analysis of the analysis and processing unit will not be affected when the scheduling query unit responds to the query information.
[0121] In the above technical solution, after receiving the monitoring and acquisition information, the real-time receiving unit simultaneously distributes the received monitoring and acquisition information to the data storage unit and the analysis and processing unit.
[0122] The above technical solution not only responds to query information through the data analysis module, but also processes monitoring and acquisition information, thereby improving the performance of the data analysis module, so that the distributed data center monitoring architecture can better feedback the monitoring results of the data center's monitoring situation. At the same time, it can also realize query responses with relevant staff, providing convenience for relevant personnel to understand the process of distributed data center monitoring architecture monitoring the data center. In addition, the storage of monitoring and acquisition information is realized through the data storage unit, which not only facilitates the retrieval and verification of historical monitoring and acquisition information, but also provides data support for data aggregation and alarm analysis of the analysis and processing unit.
[0123] like Figure 7 As shown, in one embodiment provided by the present invention, when the analysis and processing unit combines the monitoring and collection information with the historical monitoring and collection information in the data storage unit to perform data aggregation and alarm analysis, it includes:
[0124] C1. Perform a preliminary analysis on the monitoring and collection information to determine whether the monitoring and collection information exceeds the indicator threshold, and obtain the first judgment result of the analysis and processing;
[0125] C2. Retrieving historical monitoring and collection information from the data storage unit according to the indicators of the monitoring and collection information to obtain historical monitoring data of the indicators;
[0126] C3. Aggregate the monitoring and collection information according to the indicators and the historical monitoring data of the indicators to obtain the indicator monitoring and collection summary information. At the same time, predict the data fluctuations of the historical monitoring data of the indicators to obtain the indicator prediction data. Perform abnormal analysis and judgment on the monitoring and collection information according to the indicator prediction data to obtain the second judgment result of the analysis and processing;
[0127] C4. Perform alarm analysis based on the analysis and processing of the first judgment result or the analysis and processing of the second judgment result to determine whether to issue an alarm prompt and obtain an alarm analysis result.
[0128] In the above technical solution, the data analysis results are divided into two types: monitoring collection summary information and alarm analysis results.
[0129] In the above technical scheme, when predicting data fluctuations for the historical monitoring data of indicators, it is analyzed whether the time attribute of the historical monitoring data of indicators is multiple complete working cycles of the data center. When the time attribute of the historical monitoring data of indicators is not multiple complete working cycles of the data center, the historical monitoring data of indicators are divided into multiple time periods, and the monitoring data of the next time is predicted according to the time period to obtain indicator prediction data. When the time attribute of the historical monitoring data of indicators is multiple complete working cycles of the data center, the time nodes of the monitoring collection information in the complete working cycle of the data center are obtained, the historical monitoring data of indicators are screened based on the time nodes to obtain the target historical monitoring data, and the time node data is predicted based on the target historical monitoring data to obtain the indicator prediction data.
[0130] In the above technical scheme, when performing abnormal analysis and judgment on the monitoring and collection information based on the indicator prediction data, the difference data information between the indicator prediction data and the monitoring and collection information of the corresponding indicator in the monitoring and collection information is compared, and it is analyzed whether the difference data information is within the preset range. When the difference data information is within the preset range, the second judgment result of the analysis and processing is that the monitoring data is normal, otherwise, the second judgment result of the analysis and processing is that the monitoring data is abnormal.
[0131] In the above technical solution, when alarm analysis is performed according to the analysis and processing of the first judgment result or the analysis and processing of the second judgment result, either the analysis and processing of the first judgment result or the analysis and processing of the second judgment result can determine whether to issue an alarm prompt, thereby obtaining an alarm analysis result. When the analysis and processing of the first judgment result is that the monitoring collection information contains indicator monitoring collection information that exceeds the indicator threshold, the alarm analysis result is determined to issue an alarm prompt, and the alarm information is determined based on the indicator monitoring collection information that exceeds the indicator threshold in the monitoring collection information. When the analysis and processing of the first judgment result is that the monitoring collection information does not contain indicator monitoring collection information that exceeds the indicator threshold, the alarm analysis result is not to issue an alarm prompt. When the analysis and processing of the second judgment result is that the monitoring data is abnormal, the alarm analysis result is determined to issue an alarm prompt, and the alarm information is determined based on the abnormal monitoring data. When the analysis and processing of the second judgment result is that the monitoring data is normal, the alarm analysis result is not to issue an alarm prompt.
[0132] In the above technical scheme, when the alarm analysis result determined by the analysis and judgment of the first result is to issue an alarm prompt, there is no need to perform data fluctuation prediction after obtaining the historical monitoring data of the indicator. When the alarm analysis result determined by the analysis and judgment of the first result is not to issue an alarm prompt, data fluctuation prediction is performed on the historical monitoring data of the indicator after obtaining the historical monitoring data of the indicator, and the alarm analysis result is determined based on the analysis and processing of the second judgment result after obtaining the second judgment result of the analysis and processing.
[0133] The above technical solution realizes data aggregation and alarm analysis of monitoring and collection information. It can not only analyze the monitoring and collection information, but also issue an alarm when the monitoring and collection information is abnormal. Moreover, when performing alarm analysis, it can determine in the preliminary analysis that the alarm analysis result is for an alarm prompt, and no longer predict data fluctuations based on the historical monitoring data of the indicator, and analyze and process the determination of the second judgment result, thereby reducing unnecessary processing, improving the efficiency of the analysis and processing unit, and when summarizing data, combining the historical monitoring data of the indicator for data aggregation, so that the indicator monitoring and collection summary information can be compared with the historical monitoring data of the indicator, which makes it easier to reflect the current level of the monitoring and collection information in the indicator.
[0134] like Fig. 9 As shown, in one embodiment provided by the present invention, the notification and warning module includes: a graphic output unit, a management and recording unit, and an alarm notification unit;
[0135] The image output unit is used to perform graphical conversion on the data analysis results and output the data analysis results in the form of graphics;
[0136] The management and recording unit is used to update the log records according to the data analysis results;
[0137] The alarm notification unit is used to use multiple means to issue an alarm when the alarm analysis result shows that an abnormality exists according to the data analysis result.
[0138] In the above technical solution, when the image output unit performs graphical conversion on the data analysis results, the index monitoring and collection summary information is converted into a chart for presentation, and when there is alarm information, it is marked in the presented chart according to the alarm information.
[0139] In the above technical solution, the alarm means include: color alarm, voice alarm, SMS alarm, email alarm, light alarm and so on.
[0140] In the above technical solution, when the management record unit updates the log record according to the data analysis result, it parses the operation information performed by the distributed data center monitoring architecture according to the data analysis result, and updates the log record according to the operation information.
[0141] The above technical solution realizes the feedback of data analysis results in the distributed data center monitoring architecture by informing the early warning module, so that the data analysis results can be presented more intuitively, and the alarm notification unit uses different alarm means to issue alarms, thereby increasing the vigilance of relevant personnel and avoiding large deviations in the distributed data center monitoring architecture caused by neglect by relevant personnel, so that relevant personnel can deal with the alarm situation in a timely manner and reduce the impact of abnormalities. The log update record is realized through the management recording unit, so that the operation information performed in the distributed data center monitoring architecture is recorded, and the monitoring operation of the distributed data center monitoring architecture is made transparent, which also provides a reference for the maintenance of the distributed data center monitoring architecture.
[0142] Those skilled in the art should understand that the first and second in the present invention merely refer to different application stages.
[0143] Other embodiments of the present disclosure will be readily apparent to those skilled in the art after considering the specification and practicing the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.
[0144] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims
1. A high-performance distributed data center monitoring architecture, It is characterized in that include: Monitoring and acquisition module, central control module, data analysis module and notification and warning module; The monitoring and acquisition module is used to use a distributed multi-threaded monitoring architecture to use a multi-mode compatible adapter to perform monitoring access on the monitoring object and obtain monitoring and acquisition information; The central control module is used to determine the monitoring acquisition control information based on the polling device and transmit the monitoring acquisition control information to the monitoring acquisition module; The data analysis module is used to perform data aggregation and alarm analysis based on the monitoring and collection information to obtain data analysis results; The notification and warning module is used to perform graphical output presentation of results and multi-method alarm prompts based on the data analysis results.
2. The distributed data center monitoring architecture according to claim 1, It is characterized in that The monitoring and acquisition module includes multiple monitoring machines, and a distributed multi-threaded monitoring architecture between the monitoring and acquisition module and the data center is established through the monitoring machines according to a one-to-many connection rule. In addition, multi-mode compatible adapters are respectively set in the monitoring machines. The monitoring machines are adapted to the monitored objects in the data center through the multi-mode compatible adapters.
3. The distributed data center monitoring architecture according to claim 2, It is characterized in that The multi-mode compatible adapter includes multiple monitoring plug-ins. When the monitoring machine is adapted to the monitored object in the data center through the multi-mode compatible adapter, it includes: Monitor the data center, identify the monitored objects in the data center, and obtain information about the monitored objects; Determine the type of the monitored object according to the information of the monitored object; Performing a first monitoring plug-in match according to the type of the monitored object to obtain a first matching result; Combine the monitored object information to perform a second monitoring plug-in match in the first matching result to determine the target monitoring plug-in; The monitored object is connected via the target monitoring plug-in to establish a monitoring relationship between the monitored object and the monitoring machine.
4. The distributed data center monitoring architecture according to claim 1, It is characterized in that The central control module includes: a host computer unit, a central processing unit and a control transmission unit; The host computer unit is used to obtain control instruction information; The central processing unit is used to analyze the control indication information and determine the monitoring collection control information based on the polling device according to the analysis result of the control indication information; The control transmission unit is used to transmit the monitoring collection control information obtained by the central processing unit in real time by adopting a multi-link parallel processing method.
5. The distributed data center monitoring architecture according to claim 4, It is characterized in that The poller includes: a passive monitoring poller and an active monitoring poller. The central processing unit analyzes the control indication information and determines the monitoring collection control information based on the poller according to the analysis result of the control indication information, including: Analyze whether the control instruction information is quick control instruction information, and determine the analysis and judgment result; Performing standardization processing on the control instruction information according to the analysis and judgment results to obtain the standardized control instruction information; Extracting a first feature from the control instruction information or the standardized control instruction information to obtain first feature information of the control instruction; Identify the first characteristic information of the control indication, determine whether the control indication information is a control indication for a passive monitoring poller or a control indication for an active monitoring poller, obtain a first characteristic identification result of the control indication, and perform control information set matching according to the first characteristic identification result of the control indication to obtain a first control matching result; The control indication information or the standardized control indication information is parsed to obtain the second characteristic information of the control indication, and the control information is configured according to the second characteristic information of the control indication in the first control matching result to obtain the monitoring collection control information.
6. The distributed data center monitoring architecture according to claim 5, It is characterized in that The monitoring and acquisition module includes: a monitoring and acquisition unit and a data export unit; the monitoring and acquisition unit is used to perform target matching on the monitoring and acquisition control information, and input the monitoring and acquisition control information into a passive monitoring poller or an active monitoring poller according to the matching result, and then monitor the data center in the passive monitoring poller or the active monitoring poller based on the distributed multi-threaded monitoring architecture through a multi-mode compatible adapter to obtain monitoring and acquisition information; the data export unit is used to perform data preprocessing on the monitoring and acquisition information, and import the preprocessed monitoring and acquisition information into the data analysis module.
7. The distributed data center monitoring architecture according to claim 1, It is characterized in that The distributed data center monitoring architecture also includes: a query module; the query module is connected to the data analysis module. When the query module obtains query information, the data analysis module performs a target data query based on the query information, obtains the query result, and feeds back the query result through the query module.
8. The distributed data center monitoring architecture according to claim 7, It is characterized in that The data analysis module includes: a real-time receiving unit, a data storage unit, a scheduling query unit and an analysis processing unit; The real-time receiving unit is used to receive the monitoring and collecting information, and import the monitoring and collecting information into the data storage unit and the analysis and processing unit respectively; The data storage unit is used to store monitoring and collection information; The analysis and processing unit is used to combine the monitoring and collection information with the historical monitoring and collection information in the data storage unit to perform data aggregation and alarm analysis to obtain data analysis results; The scheduling query unit is used to determine a query target according to the query information, make a query request for the query target, obtain query information based on the query request, and then organize the query information to obtain a query result.
9. The distributed data center monitoring architecture according to claim 8, It is characterized in that When the analysis and processing unit combines the monitoring and collection information with the historical monitoring and collection information in the data storage unit to perform data aggregation and alarm analysis, it includes: Perform a preliminary analysis on the monitoring and collection information to determine whether there is any indicator monitoring and collection information exceeding the indicator threshold in the monitoring and collection information, and obtain a first judgment result of the analysis and processing; Retrieving historical monitoring and collection information from the data storage unit according to the indicators of the monitoring and collection information, and obtaining historical monitoring data of the indicators; Aggregate the monitoring and collection information according to the indicators and the historical monitoring data of the indicators to obtain the indicator monitoring and collection summary information, and at the same time predict the data fluctuations of the historical monitoring data of the indicators to obtain the indicator prediction data, and perform abnormal analysis and judgment on the monitoring and collection information according to the indicator prediction data to obtain the second judgment result of the analysis and processing; An alarm analysis is performed based on the analysis and processing of the first judgment result or the analysis and processing of the second judgment result to determine whether to issue an alarm prompt and obtain an alarm analysis result.
10. The distributed data center monitoring architecture according to claim 1, It is characterized in that The notification and warning module includes: a graphic output unit, a management and recording unit, and an alarm notification unit; The image output unit is used to perform graphical conversion on the data analysis results and output the data analysis results in the form of graphics; The management and recording unit is used to update the log records according to the data analysis results; The alarm notification unit is used to use multiple means to issue an alarm when the alarm analysis result shows that an abnormality exists according to the data analysis result.
