A software-defined integrated monitoring system and implementation method thereof

Abstract

The present invention discloses a software-defined integrated monitoring system and its implementation method. The system includes: a business processing cluster, a message service cluster, a management cluster, and a proxy gateway cluster. Each cluster is composed of nodes of the same type. The node is formed by deploying relevant functional modules on an independent virtual machine of a data center cloud platform. The business node registers services and communicates in real time with other three types of nodes through a multi-channel and multi-link mode, performs business redundancy management through the management node, decouples the service requester from the service provider through the proxy gateway node, and implements the subscription, storage, and distribution of business real-time messages by the corresponding service through the message service node. The present invention does not require the business node to be bound to the data of the station or center. The management cluster dynamically adjusts the business deployment through software algorithms based on the real-time monitoring of all business nodes, realizes the decoupling of the business from the operating environment, and then realizes the elastic expansion and dynamic migration of the business.

Description

Technical Field

[0001] The present invention relates to the field of rail transit integrated monitoring systems, and in particular to a software-defined integrated monitoring system and an implementation method thereof. Background Art

[0002] At present, the domestic integrated monitoring system architecture basically adopts the distributed structure of "center-station". In the field of rail transit, the implementation method of urban rail cloud is to migrate all business systems to the data center cloud platform, and continue the traditional station-center two-level deployment method in terms of system structure and business process. In order to improve reliability and solve the problem of backbone ring network interruption, a comprehensive monitoring business server is generally deployed in the station area as a backup for the station server in the data center.

[0003] At present, the integrated monitoring system is deployed in these ways. The resources deployed on the data center cloud platform include the ISCS (Integrated Supervisory Control System) central server and the ISCS station server. In this way, only the business on the original physical machine is migrated to the virtual machine. The station server in the data center still stores the data related to the station, and the data is still coupled with the operating environment. For example, the station server No. 1 has been defined in the system predefined configuration to only process the data of station No. 1. Assuming that the business of station No. 1 is running in a virtual machine, when the business of station No. 1 surges, the operating environment of the virtual machine is overloaded. Although a new virtual machine can be created by applying the characteristics of the cloud platform, the newly created virtual machine still runs all the business of station No. 1. The business processing of station No. 1 is not evenly distributed to the two virtual machines at the business level because of the addition of a virtual machine. Therefore, this deployment method cannot achieve dynamic expansion of the cloud platform. When the line is extended or a new line is built, the business system cannot achieve smooth expansion at the business level along with the smooth expansion of the cloud platform hardware and software, and the debugging cycle is too long. Moreover, although the ISCS station virtual machine is deployed at the station level, it cannot run for a long time to meet business needs due to its poor performance (generally half or even less than the ISCS station real-time node resources in the data center). In addition, due to the large difference between the debugging state and the normal operation, the configuration needs to be manually modified to make the conversion effective, and the model data in the debugging state needs to be manually synchronized to the data center, which is a large and cumbersome workload, prone to errors, and creates security risks of inconsistent data. Summary of the invention

[0004] Purpose of the invention: In view of the deficiencies in the prior art, the purpose of the present invention is to provide a software-defined integrated monitoring system and its implementation method, so as to realize flexible deployment of data on different business nodes and realize elastic expansion and dynamic migration of business.

[0005] Technical solution: In a first aspect, a software-defined integrated monitoring system is provided, the system comprising:

[0006] A business processing cluster is used to complete the processing of various types of business in the comprehensive monitoring system. The business processing cluster is composed of a number of business nodes, which are formed by deploying relevant business processing modules on independent virtual machines on the data center cloud platform;

[0007] A management cluster, used for service redundancy management, the management cluster is composed of a number of management nodes, and the management nodes are formed by deploying service management functions on independent virtual machines of the data center cloud platform;

[0008] The proxy gateway cluster is used to decouple the service requester and the service provider, and provide various business services and data services on the platform by forwarding. The proxy gateway cluster is composed of several proxy gateway nodes, and the proxy gateway nodes are formed by deploying the proxy gateway function on an independent virtual machine of the data center cloud platform;

[0009] A message service cluster is used to implement the subscription, storage and distribution of business real-time messages by corresponding services. The message service cluster is composed of a number of message service nodes, which are formed by deploying message service functions on independent virtual machines on the data center cloud platform.

[0010] The management node, proxy gateway node, and message service node provide registration services to the business node through the registration service module respectively. The business node performs service registration and real-time communication with the management node, message service node, and proxy gateway node through a multi-channel multi-link mode;

[0011] Among them, after the business node registers with the other three types of nodes, it obtains the predefined information related to the business of the node through the registered management node, downloads the data type according to the obtained information, and starts the corresponding service. After starting the service, it registers the subscription topic with the message bus of the node according to the predefined information, and periodically sends the current node message subscription topic status to the registered message service node through the message bus service; and periodically reports the device set status running on the node and the resource usage of the node to the management node; when it needs to request data service, it requests data service through the registered proxy gateway;

[0012] The management node calculates the redundancy of the device set in real time according to the periodic reports of the service nodes, writes it into the local memory and synchronizes it with other management nodes and proxy gateway nodes;

[0013] The proxy gateway node obtains data from the corresponding business node according to the data request of the business node and the redundancy of the device set synchronized by the management node, caches the data to the local real-time library and marks the time, and returns it to the data requester at the same time;

[0014] The message service node stores the message subscription topic of the business node in the local memory and synchronizes it to other message service nodes. After receiving the message of the corresponding topic, it distributes the message according to the topic subscribed by the corresponding business node in the local memory.

[0015] Preferably, the management node, proxy gateway node, and message service node respectively provide a public access network address for access by the business node, and respectively realize load balancing with other nodes of the same type through a load balancing module. The load balancing module is used for newly registered business nodes to preferentially register with management nodes, proxy gateway nodes, and message service nodes with low resource usage, and is used for automatically creating a new virtual machine to realize the corresponding node function and share the load on the original related nodes when the load on the current management node, proxy gateway node, and message service node exceeds a specified threshold.

[0016] Preferably, the message service node receives the message sent by the business node through the message receiving module and stores it in the local memory; synchronizes the subscription topic messages of each business node with other message service nodes through the message synchronization module; distributes the corresponding topic data to the relevant business nodes through the message publishing module;

[0017] The management node and the proxy gateway node synchronize the redundant information of the device set through the message synchronization module;

[0018] The proxy gateway node controls the access and flow of the business node through the flow control module, the access control module and the authority authentication module.

[0019] Preferably, the management node calculates the redundancy of the device set in real time according to the periodic report of the service node, including:

[0020] According to the information reported by each business node, the IP address, CPU and memory usage of the business node are written into the shared memory to determine the device level corresponding to the business node, which is a certain level in "line-stage-station-professional-equipment";

[0021] The equipment levels corresponding to each business node are sorted according to the hierarchical description of "line-stage-station-professional-equipment", and the minimum level under the same business is obtained as the minimum equipment set redundancy of the business. The node groups with data redundancy relationships are identified and written into the shared memory;

[0022] According to the load of the node group with data redundancy relationship, the master node corresponding to the minimum device set is determined, written into the shared memory, and the content in the shared memory is synchronized to other management nodes and proxy gateway nodes.

[0023] Preferably, the service node obtains predefined information related to the service of the current node through the registered management node, and determines the device set running on the current node according to the predefined information; the predefined information is stored in the management node in the form of a configuration file.

[0024] In a second aspect, a method for implementing a software-defined integrated monitoring system includes a deployment phase and a working phase.

[0025] In the deployment phase: according to the virtual machine VM resource pool provided by the data center cloud platform, the business management function is deployed on at least two independent VMs, each VM forms a management node, and at least two management nodes form a management cluster for business redundancy management of the integrated monitoring system; the message service function is deployed on at least two independent VMs, each VM forms a message service node, and at least two message service nodes form a message service cluster, which is used to implement the subscription, storage, and distribution of business real-time messages by the corresponding service; the proxy gateway function is deployed on at least two independent VMs, each VM forms a proxy gateway node, and at least two proxy gateway nodes form a proxy gateway cluster, which is used to decouple the service requester and the service provider; the business processing function is deployed on several independent VMs as needed, each VM forms a business node, and several business nodes form a business cluster, which is used to complete the processing of various types of business in the integrated monitoring system;

[0026] During the working phase: the business node reads the local configuration file, obtains the information of the management node, message service node, and proxy gateway node, and registers with the public addresses of the management node, message service node, and proxy gateway node respectively;

[0027] The business node obtains the predefined information related to the current node business through the registered management node, and after obtaining the information, it downloads the full data model step by step, and starts the front-end service, message bus service and related business services; the front-end service obtains the required operation data and interface location of the node from the management node, collects data through the link, and refreshes the real operation data of the current node; after the relevant business services are started, they register and subscribe to the topic with the message bus of the node according to their respective predefined information; after starting the message bus service, register the link with the message service node, and periodically send the current node message subscription topic status to the registered message service node through the message bus service;

[0028] After the business node is registered, it periodically reports the status of the device set running on the current node and the usage of business node resources to the registered management node; the management node calculates the redundancy of the device set in real time, writes it into the local memory, and synchronizes it with other management nodes and proxy gateways; the proxy gateway synchronizes the device set redundancy information by receiving the management node message and stores it in the local shared memory;

[0029] When a business node needs to request data services, it first sends a request to the proxy gateway registered with the node. The proxy gateway obtains data from the corresponding business node based on the redundancy of the device set synchronized by the management node, caches it in the local real-time library, adds a time stamp, and returns the data to the requester.

[0030] After the message service node is started, it listens to the connection request sent by the business node. When the two parties establish a connection, the message service node receives the message sent by the business node, distributes the message according to the topics subscribed by all business nodes in the shared memory, and sends it to the corresponding business node.

[0031] Beneficial effects: The present invention designs four independent functional modules for the comprehensive monitoring system, namely, a business management module (responsible for redundant management of business systems running on the platform), a business processing module (completing business processing of various business modules running on the platform), a proxy gateway module (decoupling the service requester and the service provider, and providing various business services and data services on the platform by forwarding) and a message service module (completing management functions such as subscription, storage, and distribution of business real-time messages by the business processing module and the corresponding services of the client), and deploys the four modules on the cloud platform virtual machine according to actual needs, forming four types of nodes, namely, business management nodes (referred to as management nodes), business processing nodes (referred to as business nodes), proxy gateways (referred to as proxy gateways) and message service nodes (referred to as message nodes). The business node registers services and communicates in real time with the management node, message service node and proxy gateway through a multi-channel and multi-link manner, thereby decoupling data from the operating node environment and enabling flexible deployment of data on different business nodes. There is no need to bind the business node to the data of the station or center, so that the business data processed and the services provided in the business node are independent of the initial configuration of the business node and the current operating environment of the node. The management cluster dynamically adjusts the business deployment through software algorithms based on real-time monitoring of all business nodes in the platform, thereby decoupling the business from the operating environment and achieving elastic expansion and dynamic migration of the business. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is the overall schematic diagram of the software-defined integrated monitoring system;

[0033] Figure 2 This is a schematic diagram of a business processing cluster. DETAILED DESCRIPTION

[0034] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0035] The present invention proposes a software-defined integrated monitoring system and its implementation method, which are used for the specific implementation and application of rail transit integrated automation system. The integrated monitoring system is divided into four independent functional modules from the software function, namely business management module, business processing module, proxy gateway module and message service module.

[0036] Business management module: responsible for the redundancy management of business systems running on the platform;

[0037] Business processing module: completes the business processing of various business modules running on the platform (such as comprehensive monitoring business module, power monitoring business module and train automatic monitoring business module);

[0038] Proxy gateway module: decouples service requesters from service providers, and provides various business services and data services on the platform through forwarding;

[0039] Message service module: completes the business processing module and the client's corresponding service's subscription, storage, distribution and other management functions for business real-time messages.

[0040] By deploying the above four types of modules on the virtual machines of the data center cloud platform, four types of nodes are formed, namely business management nodes (referred to as management nodes), business processing nodes (referred to as business nodes), proxy gateway nodes (referred to as proxy gateways) and message service nodes (referred to as message nodes). Nodes of the same type form corresponding clusters. Therefore, the system contains a total of four functional clusters, namely management clusters, business processing clusters, proxy gateway clusters and message service clusters.

[0041] like Figure 1 As shown, the software-defined integrated monitoring service system provided by the present invention includes:

[0042] The business processing cluster is used to complete the processing of various types of business in the integrated monitoring system. The business processing cluster is composed of several business nodes, which are formed by deploying relevant business processing modules on independent virtual machines on the data center cloud platform. The business nodes run the business services of the integrated monitoring system and are actually the providers of the business.

[0043] A management cluster, used for service redundancy management, the management cluster is composed of a number of management nodes, and the management nodes are formed by deploying service management functions on independent virtual machines of the data center cloud platform;

[0044] The proxy gateway cluster is used to decouple the service requester and the service provider, and provide various business services and data services on the platform by forwarding; the proxy gateway cluster is composed of a number of proxy gateway nodes, and the proxy gateway nodes are formed by deploying the proxy gateway function on an independent virtual machine of the data center cloud platform;

[0045] A message service cluster is used to implement the subscription, storage and distribution of business real-time messages by corresponding services. The message service cluster is composed of a number of message service nodes, which are formed by deploying message service functions on independent virtual machines on the data center cloud platform.

[0046] The management node, proxy gateway node and message service node provide registration services to the business node through the registration service module respectively. The business node performs service registration and real-time communication with the management node, message service node and proxy gateway node through a multi-channel multi-link mode.

[0047] Service requester in Figure 1 The client is shown in the figure. The business node will also run the corresponding process and will also have the need for service requests. Strictly speaking, the service requester can be considered as a client process. If the business node also runs a client process, the business node can be considered as a combination of the client and the business node. For the integrated monitoring system, the client process (service requester) generally runs on an independent node (virtual machine). This type of process is mainly responsible for the human-computer interface interaction and visualization functions. The node running this type of client process is generally called a client.

[0048] The so-called proxy gateway has the function of decoupling the client and the server of the information service, and mainly provides information services to the client's service request. That is, for the client, the proxy gateway is its server. For example, the client needs the BAS fan data of Station 1. The client does not need to know where the main service and related data of the BAS fan data of Station 1 are. It only needs to send a request to the proxy gateway. The proxy gateway obtains the IP and port of the main service of the BAS fan of Station 1 through the redundant information of the device set obtained from the management node, and then obtains the data from the main service of the BAS fan data and returns it to the client. The main service of the BAS fan data here is a kind of information service, and the proxy gateway realizes the decoupling of the client and the information service. There are access control modules and flow control modules inside the proxy gateway module, which are for the purpose of access control of client requests to achieve the role of gateway.

[0049] The deployment method of each functional module is as follows: a resource pool is established in the data center to provide virtual machines VM for system use; four types of functional modules are deployed on the virtual machines VM according to business needs: the business management function is deployed on at least 2 independent VMs, which can be dynamically and adaptively expanded; the message service function is deployed on at least 2 independent VMs, which can be dynamically and adaptively expanded; the proxy gateway function is deployed on at least 2 independent VMs, which can be dynamically and adaptively expanded; the business processing function is deployed on several independent VMs as needed, which can be dynamically expanded.

[0050] For reliability reasons, the management cluster, message service cluster, and proxy gateway cluster generally run on at least two nodes to ensure that if one node fails, the other node can provide services seamlessly. The business cluster runs on several nodes according to demand. The dynamic adaptive expansion mentioned above is the ability to automatically establish new nodes and distribute loads by monitoring node status. It mainly responds to two scenarios, taking the management cluster as an example:

[0051] 1) When one management node fails, the other management node will find that it is the only one working normally. At this time, according to the pre-configuration (the current system has two nodes), the system automatically applies for a virtual machine, and this virtual machine runs the management function module. When this virtual machine is fully operational, there are two nodes in the management cluster again;

[0052] 2) When the two nodes of the management cluster are operating normally, due to the large number of business nodes, the two management nodes process many requests from the business nodes, and the loads of the two management nodes exceed the predefined threshold (such as the predefined threshold is that the CPU utilization rate exceeds 70%). If the load balancing module on the management node cannot reduce the CPU utilization rate of the two nodes to below 70% no matter how it migrates the business on the two nodes, a new virtual machine will be automatically applied for, and this virtual machine will run the management function module. When this virtual machine is fully and normally operating, there will be three nodes in the management cluster. At this time, the original two nodes will migrate the relevant business to the new management node, so that the CPU utilization rate of the original two management nodes will be reduced to below 70%, ensuring the normal operation of the system.

[0053] In the above two scenarios, management nodes are automatically added, so it is called dynamic adaptive expansion. The same logic applies to message service nodes, proxy gateways, and business nodes, which are dynamically adaptively expanded.

[0054] In the present invention, the virtual machine VM running the comprehensive automation business is not divided according to the traditional center and station, but is set according to user-defined requirements. The smallest set of business processing can be refined to the equipment, hereinafter referred to as the equipment set (devSet); for the equipment set, its characteristics are described by the five levels of "line-stage-station-professional-equipment". For business processing, data is processed through business services. The data of comprehensive monitoring are all based on the points of the equipment, and for the rail transit industry, a point of a device can be represented by the hierarchy of "line-stage-station-professional-equipment-point", where:

[0055] Line: refers to a specific line in rail transit, such as Metro Line 1, Light Rail Line 2, etc.;

[0056] Phase: usually refers to the construction or opening stage of the line, which may be divided into Phase I, Phase II, etc., to distinguish the construction and operation status of the project in different time periods;

[0057] Station: Each station on a rail transit line, such as a subway station or light rail station;

[0058] Major: In the field of rail transit, major usually refers to different technical fields or systems, such as signal system (SIG), power supply system (PSCADA), electromechanical system (BAS), etc.

[0059] Equipment: refers to specific equipment or facilities in a certain professional field. For example, in the power supply system, the equipment may include transformers, circuit breakers, overhead lines, etc.; in the signal system, the equipment may include signal machines, switch machines, ATP systems, etc.

[0060] A point (also called a measuring point or monitoring point) can be a specific monitoring location on a device, such as a temperature sensor of a transformer, a start switch of a fan, etc. Each point corresponds to a specific monitoring parameter or status on the device; it can also be a basic unit of data acquisition: in a monitoring system, each point may correspond to a data acquisition address, which is used to collect real-time data of the point, such as temperature, humidity, current, voltage, switch status, etc.; a point can also be an interface for the control system to send instructions and receive feedback information, such as remotely controlling the start or stop of a device.

[0061] By monitoring these "points", the integrated monitoring system can grasp the operating status of each device and system in the rail transit system in real time, detect abnormalities in time, ensure driving safety, and improve operational efficiency. At the same time, through hierarchical management, it can effectively organize and query a large amount of monitoring data, which is convenient for troubleshooting and maintenance management.

[0062] Considering high availability, the integrated monitoring system needs redundant design, that is, the same business and data are deployed and run on different nodes, but only one business is active at the same time, providing data and services to the outside world. In this way, when a primary node of a certain business has a physical failure or the service running on this node that processes the current business fails, at least one other node related service will be immediately upgraded to the primary node of the current business, providing services and data to the outside world, ensuring the reliable operation of the system.

[0063] Compared with the whole machine redundancy of the traditional comprehensive monitoring scheme, the accuracy of the main service in the present invention can be accurate to the device set, that is, the redundancy can be to the minimum device set for business processing. For example, the traditional scheme has 2 servers running the comprehensive monitoring business of Station 1, but only one server can provide the main service at the same time. At this time, a large number of customer requests are processed by that server, while the other server does not provide services to the outside. If the client needs a BAS professional equipment operation of Station 1 and a PSCADA professional equipment operation of Station 1, both need to be connected to the main server, that is, data access to the station level is required, and data redundancy is also to the station level; in the system of the present invention, business deployment is based on user needs, accurate to the device set, such as the BAS professional equipment operation of Station 1 and the PSCADA professional equipment operation of Station 1 can be connected separately, because the main data can be on different nodes, so different nodes can be connected at the same time, because the redundancy is to the device set, so this scheme, for the operation of a BAS professional jet fan of Station 1 and the operation of a 400V knife switch of Station 1, different main services can be accessed.

[0064] According to the implementation mode of the present invention, based on the hierarchical description of the equipment set, unified supervision of data at different redundancy levels can be achieved. For example, for the equipment set of the jet fan, this equipment can be found through "Nanjing Line 9-Phase 1-Sanpailou Station-BAS Professional-Jet Fan". As stated above, its redundancy object can be any of the five levels, such as Figure 2 As shown, for the redundant object being temperature-sensitive optical fiber, its description is Station 3-BAS-temperature-sensitive optical fiber (the complete description is Nanjing Line 9-Phase 1-Station 3-BAS-temperature-sensitive optical fiber), and its redundancy level is Level 5; for the electric power profession (PSCADA), there are multiple redundant objects, namely, the redundant object is the entire line PSCADA (the complete description is Nanjing Line 9-PSCADA, the redundancy level is Level 2), the first phase PSCADA (the complete description is Nanjing Line 9-Phase 1-PSCADA, the redundancy level is Level 3). It can be seen from the figure that according to business needs, for monitoring data, redundant objects and redundancy levels can be customized, and data of different redundancy levels can be uniformly supervised.

[0065] The redundancy of the business processing module is based on the minimum set of business processing. Users can customize the degree of redundancy and the granularity of the set according to business needs. For businesses with high reliability requirements, they can be distributed and deployed on N nodes for redundancy. Preferably, in order to be compatible with the traditional mode, the business is deployed on the business processing node in the data center according to the central station, and the business management module, message service module, and proxy gateway module are also deployed on the business node according to demand. The station edge cloud node provides a virtual machine to deploy the management module, message service module, proxy gateway module and the current station business module. In this way, when the cloud platform fails or the backbone network is interrupted, the station edge cloud node can work normally to ensure that the business is not interrupted. The management module, proxy gateway node, and message service node of each station edge cloud node can be configured in the data center or locally.

[0066] The following is a detailed description of the workflow of each node.

[0067] (I) Management Node Workflow

[0068] (1) The management cluster provides public access network addresses to business nodes. The private addresses of each management node are inaccessible to business nodes.

[0069] (2) After the management node is started, the registration service is opened for business nodes to register;

[0070] (3) After the business node is registered, it periodically reports to the management node the status of the device set running on the current node and the resource usage of the business node. After receiving the information, the management node writes the IP address, CPU and memory usage of the business node into the shared memory according to the device set, determines the minimum device set redundancy, and sorts the nodes according to the load situation. It obtains the active node of the current device set, stores it in the local memory, and synchronizes it to the management cluster and proxy gateway;

[0071] (4) The management node predefines the running services of each service node and stores the configuration files of the relevant service nodes on the management node so that each service node can obtain this information and understand which services to run. In the later stage, the service deployment is dynamically adjusted according to the actual running management node.

[0072] (5) The management node monitors the management cluster in real time to achieve load balancing of the management nodes, giving priority to business nodes to register with management nodes that occupy less resources. When multiple management nodes in the cluster reach the predefined threshold, the management node calls the cloud platform interface to dynamically start a preconfigured management node, which joins the management cluster after it starts normally.

[0073] (II) Proxy Gateway Workflow

[0074] (1) The proxy cluster provides a public access network address to the business node, and the business node is not accessible to the private address of each proxy gateway;

[0075] (2) After the proxy gateway is started, the registration service is opened for business nodes or clients to register. For client registration, two interfaces, C and Java, are provided, which are applicable to both B / S and C / S frameworks.

[0076] (3) The proxy gateway receives messages from the management node, synchronizes the redundant information of the device set in the platform, and stores it in the local shared memory;

[0077] (4) After the client process (which may be running on the client or business node) sends a request, the proxy gateway receives it and obtains the main node information of the device set in real time (synchronized by the management node) based on the query memory, caches it in the local real-time library and timestamps it, and returns the data to the client process at the same time;

[0078] (5) The proxy gateway monitors the proxy gateway cluster in real time to achieve load balancing of the proxy gateway nodes, giving priority to business nodes to register with proxy gateway nodes that occupy less resources. When multiple proxy gateway nodes in the cluster reach the predefined threshold, the proxy gateway node calls the cloud platform interface to dynamically start a preconfigured proxy gateway node. After this node starts normally, it joins the proxy gateway cluster.

[0079] (III) Message service node workflow

[0080] (1) The message service cluster provides a public access network address to the business node. The private address of each message service node is inaccessible to the business node;

[0081] (2) After the message service node is started, the registration service is opened for service registration of business nodes or clients. It also provides C and Java interfaces for client registration, which are applicable to both B / S and C / S frameworks.

[0082] (3) After the message service node is started, it listens for requests. When the business node is started, it sends a connection request. When the two parties establish a connection, the message service node receives the message sent by the business node about the topic subscribed by the current node and stores it in the shared memory. After receiving the message, it distributes the message according to the topic subscribed by the business node in the shared memory and then sends it to the corresponding business node.

[0083] (4) The message service node monitors the message service cluster in real time to achieve load balancing of the message service nodes, giving priority to business nodes to register with message service nodes that occupy less resources. When multiple message service nodes in the cluster reach the predefined threshold, the message service node calls the cloud platform interface to dynamically start a preconfigured message service node. After this node starts normally, it joins the message service cluster.

[0084] (IV) Business Node Workflow

[0085] (1) The service node reads the local configuration file, obtains the management node, message service node information, and proxy gateway information, and registers with the management node, message service node, and proxy gateway public addresses respectively. The local configuration file refers to the configuration file on the virtual machine where the service node runs. The configuration file records the proxy gateway information, message service node information, and management node information accessed by each service node. In the present invention, the configuration file includes the information of all service nodes. Each service node only needs to read the part related to its own service node, so the configuration files of all service nodes are the same.

[0086] (2) Obtain the predefined information related to the current node business through the registered management node (which professional and business points the current business node needs to handle). For example, the predefined information of the current node defines the device set running on this node, including:

[0087] Nanjing Line 9 - the entire line - stations 1\2\3\4 - ACS (level 4 equipment set)

[0088] Nanjing Line 9 - Whole Line - CCTV (Level 3 Equipment Set)

[0089] Nanjing Line 9 - Whole Line - Station 5 - BAS - Fan (Level 5 Equipment Set)

[0090] Nanjing Line 9 - Entire Line - PSCADA-400V Circuit Breaker (4-level Equipment Set)

[0091] Nanjing Line 9 - entire line - stations 2\3\4\5\6\7 - FAS (level 4 equipment set).

[0092] (3) After obtaining the information, the business node downloads the full data model, that is, downloads the data of the entire subway line according to the 5-level equipment set. The full data model is downloaded from a dedicated database. The database IP address can be obtained from the predefined information related to the current node business obtained by registering the management node in step (2).

[0093] The present invention downloads data level by level, starting from level 5, such as first downloading Nanjing Line 9-Phase 1-Station 1-PSCADA-400V circuit breaker data, then downloading Nanjing Line 9-Phase 1-Station 1-PSCADA-1500V circuit breaker data, etc. After the download of level 5 data is completed, continue to download level 4 data, that is, download Nanjing Line 9-Phase 1-Station 1-BAS-jet fan data, then download Nanjing Line 9-Phase 1-Station 1-BAS-axial flow fan data, etc., until the download of level 1 equipment set data is completed and the corresponding services are started, including business-related services (such as data processing, alarm remote control and other services), as well as front-end services and message bus services.

[0094] (4) After the front-end service is started, the front-end service obtains the required operation data and interface location of the node from the management node, collects data through the link, and refreshes the real operation data of the current node. The services such as data processing, alarm, remote control, etc. run by the business node completely process the real-time data of the tag through the real-time library;

[0095] After the message bus service is started, the message bus service registers the link with the message service node. After the relevant business services on the business node are started, they register the subscription topic with the message bus of this node according to their predefined requirements. The message service node receives and stores it in the local memory and synchronizes it to all nodes in the message cluster. When the message service node receives the message of the corresponding topic, it will forward it to the message bus service of the business node in real time, and the message bus service will distribute the message to the service that subscribes to the topic. Figure 2 Taking the business node VM in the upper left corner as an example, the data currently running on the VM includes BAS fans at stations 1-5, CCTV station hall gun camera at station 5, PIS for the entire line, PA at station 4, ATS for the entire line, and the BAS large system at station 2. After the BAS fan service is started on the business node, it directly subscribes to data with the topic BAS-FAN to the message bus of this node. After the CCTV station hall gun camera service is started on the business node, it directly subscribes to data with the topic CCTV-CAMERA to the message bus of this node. The business node message bus module periodically sends the current node subscription topic to the message service node. After receiving the message service node, it records the current node's subscription topics BAS-FAN and CCTV-CAMERA in the shared memory. When the message of these two topics is received next time, the topic data is directly distributed to the relevant business nodes.

[0096] (5) All business nodes register with the management node, which arbitrates, calculates the redundancy of the current minimum set of services, and synchronizes information with other management nodes and proxy gateways. Figure 2Taking PSCADA as an example, the PSCADA business has a total of 4 business node VMs, which run the full-line PSCADA data (level 2 device set), called VM1; the first-phase PSCADA data (level 3 device set), called VM2; the second-phase PSCADA data (level 3 device set), called VM3; the third-phase PSCADA data (level 3 device set), called VM4. These 4 VMs send the device sets running on the current nodes to the management node. VM1 sends Nanjing Line 9-PSCADA, VM2 sends Nanjing Line 9-Phase 1-PSCADA, VM3 sends Nanjing Line 9-Phase 2-PSCADA, and VM4 sends Nanjing Line 9-Phase 3-PSCADA. The management node receives the data information sent by each VM and first sorts them according to the device set level. VM1 sends level 2 device set data, and VM2\VM3\VM4 send level 3 device data respectively. Therefore, the level 2 device set data sent by VM1 is greater than the level 3 device set data of VM2\VM3\VM4. Therefore, for PSCADA data, the level 3 device set is the minimum device set redundancy, that is, the data redundancy of VM1 and VM2 for Nanjing Line 9-PSCADA-1 phase; the data redundancy of VM1 and VM3 for Nanjing Line 9-PSCADA-2 phase; the data redundancy of VM1 and VM4 for Nanjing Line 9-PSCADA-3 phase, and then sorted according to the load situation on VM1 and VM2, calculated the main node information of this level 3 device set for Nanjing Line 9-PSCADA-1 phase, and synchronized this information to other management nodes and proxy gateways, the main node information of the level 3 device set for Nanjing Line 9-PSCADA-2 phase and Nanjing Line 9-PSCADA-2 phase is also calculated by the same method, and synchronized to other management nodes and proxy gateways.

[0097] (6) When a business node needs to request data service, it first sends a request to the proxy gateway registered with this node (it has already registered with the proxy gateway). The request information includes the point label information of the target data. The proxy gateway extracts the device set information of the target data based on the point label data, and finds the primary node information of the target data device set based on the device set redundancy synchronized by the management node in the current shared memory. It obtains data from the primary node, caches it in the local real-time library, adds a time stamp, and returns it to the business node at the same time. Taking the remote control opening of the 102nd switch of the 400V power circuit breaker of Station 2 on the business node as an example, the power remote control service (operate_server_PSCADA) first sends a request to the proxy gateway according to the device point label that needs to be remotely controlled (line9-all-2-PSCADA-400Vdlq-102kg, which means the 400V power circuit breaker 102 switch of Station 2 of the entire Line 9). After receiving the request, the proxy gateway queries the device set to which this device belongs based on this device point label, which belongs to Station 2-PSCADA-400V circuit breaker. After querying the host of this device set in the current cache, it obtains the IP and port, sends an operation command to the business host of the corresponding IP and port, and returns the operation result to the operate_server_PSCADA service of the business node.

[0098] Based on the workflow of each node described above, the implementation method of the software-defined integrated monitoring system provided by the present invention includes a deployment phase and a working phase. In the deployment phase, the following steps are included:

[0099] S11. According to the virtual machine VM resource pool provided by the data center cloud platform, the business management function is deployed on at least two independent VMs, each VM forms a management node, and at least two management nodes form a management cluster for business redundancy management of the comprehensive monitoring system;

[0100] S12, deploying the message service function on at least two independent VMs, each VM forming a message service node, and at least two message service nodes forming a message service cluster, for implementing the subscription, storage, and distribution of business real-time messages by the corresponding service;

[0101] S13, deploying the proxy gateway function on at least two independent VMs, each VM forming a proxy gateway node, and at least two proxy gateway nodes forming a proxy gateway cluster, which is used to decouple the service requester and the service provider;

[0102] S14, deploying the business processing functions on several independent VMs as required, each VM forming a business node, and several business nodes forming a business cluster, which is used to complete the processing of various types of business in the comprehensive monitoring system;

[0103] During the working phase, the following steps are included:

[0104] S21. The service node reads the local configuration file, obtains the information of the management node, the message service node, and the proxy gateway node, and registers with the public addresses of the management node, the message service node, and the proxy gateway node respectively;

[0105] S22, the business node obtains the predefined information related to the current node business through the registered management node, and after obtaining the information, downloads the full data model step by step, and starts the front-end service, message bus service and related business services; the front-end service obtains the required operation data and interface location of the node from the management node, collects data through the link, and refreshes the real operation data of the current node; after starting the message bus service, register the link with the message service node; after the related business services are started, according to their respective predefined requirements, register the subscription topic with the message bus of the node, and periodically send the current node message subscription topic status to the message service node through the message bus service;

[0106] S23. After the business node is registered, it periodically reports to the management node the status of the device set running on the current node and the resource usage of the business node; the management node calculates the redundancy of the device set in real time, writes it into the local memory, and synchronizes it to the management cluster and the proxy gateway; the proxy gateway synchronizes the device set redundancy information by receiving the management node message and stores it in the local shared memory;

[0107] S24. When a business node needs to request data service, it first sends a request to the proxy gateway registered with the node. The proxy gateway obtains data from the corresponding business node according to the redundancy of the device set synchronized by the management node, caches it in the local real-time library and timestamps it, and returns the data to the requester at the same time;

[0108] S25. After the message service node is started, it listens to the link request sent by the business node. When the two parties have established a link, the message service node receives the message sent by the business node, distributes the message according to the topics subscribed by all business nodes in the shared memory, and then sends it to the corresponding business node.

[0109] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, devices (systems), computer equipment or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0110] The present invention is described with reference to a flowchart of a method according to an embodiment of the present invention. It should be understood that each process in the flowchart and a combination of processes in the flowchart can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart. Figure 1 A device that specifies functions in a process or multiple processes.

[0111] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A function specified in a process or multiple processes.

[0112] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 The steps of a specified function in a process or multiple processes.

Claims

1. A software-defined integrated monitoring system, characterized in that: The system comprises: A business processing cluster is used to complete the processing of various types of business in the comprehensive monitoring system. The business processing cluster is composed of a number of business nodes, which are formed by deploying relevant business processing modules on independent virtual machines on the data center cloud platform; A management cluster, used for service redundancy management, the management cluster is composed of a number of management nodes, and the management nodes are formed by deploying service management functions on independent virtual machines of the data center cloud platform; The proxy gateway cluster is used to decouple the service requester and the service provider, and provide various business services and data services on the platform by forwarding. The proxy gateway cluster is composed of several proxy gateway nodes, and the proxy gateway nodes are formed by deploying the proxy gateway function on an independent virtual machine of the data center cloud platform; A message service cluster is used to implement the subscription, storage and distribution of business real-time messages by corresponding services. The message service cluster is composed of a number of message service nodes, which are formed by deploying message service functions on independent virtual machines on the data center cloud platform. The management node, proxy gateway node, and message service node provide registration services to the business node through the registration service module respectively. The business node performs service registration and real-time communication with the management node, message service node, and proxy gateway node through a multi-channel multi-link mode; Among them, after the business node registers with the other three types of nodes, it obtains the predefined information related to the business of the node through the registered management node, determines the data type according to the predefined information, and starts the corresponding service. After starting the service, it registers the subscription topic with the message bus of the node according to the predefined information, and periodically sends the current node message subscription topic status to the registered message service node through the message bus service; and periodically reports the device set status running on the node and the resource usage of the node to the management node; when it needs to request data service, it requests data service through the registered proxy gateway; The management node calculates the redundancy of the device set in real time according to the periodic reports of the service nodes, writes it into the local memory and synchronizes it with other management nodes and proxy gateway nodes; The proxy gateway node obtains data from the corresponding business node according to the data request of the business node and the redundancy of the device set synchronized by the management node, caches the data to the local real-time library and marks the time, and returns it to the data requester at the same time; The message service node stores the message subscription topic of the business node in the local memory and synchronizes it to other message service nodes, and after receiving the message of the corresponding topic, distributes the message according to the topic subscribed by the corresponding business node in the local memory; The management node calculates the redundancy of the equipment set in real time according to the periodic reports of the service nodes, including: According to the information reported by each business node, the IP address, CPU and memory usage of the business node are written into the shared memory to determine the device level corresponding to the business node. The device level is the preset level in "Line-Phase-Station-Professional-Equipment"; The equipment levels corresponding to each business node are sorted according to the hierarchical description of "line-stage-station-professional-equipment", and the minimum level under the same business is obtained as the minimum equipment set redundancy of the business. The node groups with data redundancy relationships are identified and written into the shared memory; According to the load of the node group with data redundancy relationship, the master node corresponding to the minimum device set is determined, written into the shared memory, and the content in the shared memory is synchronized to other management nodes and proxy gateway nodes.

2. The system according to claim 1, characterized in that The management node, proxy gateway node, and message service node respectively provide a public access network address for access by the business node, and respectively realize load balancing with other nodes of the same type through a load balancing module. The load balancing module is used for newly registered business nodes to preferentially register with management nodes, proxy gateway nodes, and message service nodes with low resource occupancy, and is used for automatically creating a new virtual machine to realize the corresponding node function and share the load on the original related nodes when the load on the current management node, proxy gateway node, and message service node exceeds a specified threshold.

3. The system according to claim 1, characterized in that The message service node receives the message sent by the business node through the message receiving module and stores it in the local memory; synchronizes the subscription topic messages of each business node with other message service nodes through the message synchronization module; Distribute corresponding topic data to relevant business nodes through the message publishing module; The management node and the proxy gateway node synchronize the redundant information of the device set through the message synchronization module; The proxy gateway node controls the access and flow of the business node through the flow control module, the access control module and the authority authentication module.

4. The system according to claim 1, characterized in that The service node obtains predefined information related to the service of the current node through the registered management node, and determines the device set running on the current node according to the predefined information; the predefined information is stored in the management node in the form of a configuration file.

5. A method for implementing a software-defined integrated monitoring system, applied to the software-defined integrated monitoring system as claimed in any one of claims 1 to 4, characterized in that: The method comprises a deployment phase and a working phase. In the deployment phase: according to the virtual machine VM resource pool provided by the data center cloud platform, the business management function is deployed on at least two independent VMs, each VM forms a management node, and at least two management nodes form a management cluster for business redundancy management of the integrated monitoring system; the message service function is deployed on at least two independent VMs, each VM forms a message service node, and at least two message service nodes form a message service cluster, which is used to implement the subscription, storage, and distribution of business real-time messages by the corresponding service; the proxy gateway function is deployed on at least two independent VMs, each VM forms a proxy gateway node, and at least two proxy gateway nodes form a proxy gateway cluster, which is used to decouple the service requester and the service provider; Deploy business processing functions on several independent VMs as required. Each VM forms a business node, and several business nodes form a business cluster to complete the processing of various types of business in the comprehensive monitoring system. During the working phase: the business node reads the local configuration file, obtains the information of the management node, message service node, and proxy gateway node, and registers with the public addresses of the management node, message service node, and proxy gateway node respectively; The business node obtains the predefined information related to the current node business through the registered management node, downloads the full data model step by step after obtaining the information, and starts the front-end service, message bus service and related business services; The front-end service obtains the required operation data and interface location of the node from the management node, collects data through the link, and refreshes the real operation data of the current node; After the relevant business services are started, they register and subscribe to topics with the message bus of this node according to their respective predefined information; After starting the message bus service, register the link with the message service node, and periodically send the current node message subscription topic status to the registered message service node through the message bus service; After the business node is registered, it periodically reports the status of the current node's running equipment set and the business node resource usage to the registered management node; The management node calculates the redundancy of the device set in real time, writes it into the local memory, and synchronizes it to other management nodes and proxy gateways; The proxy gateway receives messages from the management node, synchronizes the redundant information of the device set, and stores it in the local shared memory; When a business node needs to request data services, it first sends a request to the proxy gateway registered with the node. The proxy gateway obtains data from the corresponding business node based on the redundancy of the device set synchronized by the management node, caches it in the local real-time library, adds a time stamp, and returns the data to the requester. After the message service node is started, it listens to the connection request sent by the business node. When the two parties establish a connection, the message service node receives the message sent by the business node, distributes the message according to the topics subscribed by all business nodes in the shared memory, and sends it to the corresponding business node.

6. The method according to claim 5, characterized in that The management node, proxy gateway node, and message service node respectively provide a public access network address for access by the business node, and respectively realize load balancing with other nodes of the same type through a load balancing module. The load balancing module is used for newly registered business nodes to preferentially register with management nodes, proxy gateway nodes, and message service nodes with low resource occupancy, and is used for automatically creating a new virtual machine to realize the corresponding node function and share the load on the original related nodes when the load on the current management node, proxy gateway node, and message service node exceeds a specified threshold.

7. The method according to claim 5, characterized in that The message service node receives the message sent by the business node through the message receiving module and stores it in the local memory; synchronizes the subscription topic messages of each business node with other message service nodes through the message synchronization module; Distribute corresponding topic data to relevant business nodes through the message publishing module; The management node and the proxy gateway node synchronize the redundant information of the device set through the message synchronization module; The proxy gateway node controls the access and flow of the business node through the flow control module, the access control module and the authority authentication module.

8. The method according to claim 5, characterized in that The management node calculates the redundancy of the device set in real time, writes it to the local memory, and synchronizes it to other management nodes and proxy gateways, including: According to the information reported by each business node, the IP address, CPU and memory usage of the business node are written into the shared memory to determine the device level corresponding to the business node. The device level is the preset level in "Line-Phase-Station-Professional-Equipment"; The equipment levels corresponding to each business node are sorted according to the hierarchical description of "line-stage-station-professional-equipment", and the minimum level under the same business is obtained as the minimum equipment set redundancy of the business. The node groups with data redundancy relationships are identified and written into the shared memory; According to the load of the node group with data redundancy relationship, the master node corresponding to the minimum device set is determined, written into the shared memory, and the content in the shared memory is synchronized to other management nodes and proxy gateway nodes.

9. The method according to claim 5, characterized in that The service node obtains predefined information related to the service of the current node through the registered management node, and determines the device set running on the current node according to the predefined information; the predefined information is stored in the management node in the form of a configuration file.

Images