TW201812611A - System and method for automatic construction and horizontal extension of service element based on dynamic cluster rule template the entire process can be accomplished in several seconds to improve the expansion speed of the service element - Google Patents

Abstract

The present invention discloses a system and a method for automatic construction and horizontal extension of a service element based on a dynamic cluster rule template. The system utilizes Linux Container lightweight actuating technique and can actuate a service element in a few seconds, and cluster configuration information can be obtained from a cache memory by a coordination unit and set up immediately. The entire process can be accomplished in several seconds, significantly improving automatic cluster setup and service element expansion speed. In addition, service elements that operate in conjunction with application services can independently extend or reduce service nodes without affecting existing functions and states of cooperated operations between service elements. This rapid and automatic setting method provides flexibility in the combination of service elements and improves cluster construction efficiency.

Description

   System and method for automatic construction and horizontal extension of service elements based on dynamic cluster rule template   

The invention belongs to a system and method for automatically constructing and horizontally extending service elements of a dynamic cluster rule template. In particular, the present invention can independently perform extension or reduction of service nodes without affecting existing functions and states that are cooperating among service elements.

In today's shopping or booking websites, service components such as database services, cache services, or queue services are generally used to support the site's message storage and delivery functions, and cluster architecture is used to provide reliability and availability. Coupled with the scalability of cloud computing technology to deal with the problem of instant influx of large amounts of traffic. However, simply starting the extension does not guarantee the performance of the cluster, and additional configuration information needs to be provided to all cluster members of the service component after the extension.

In view of the various shortcomings derived from the above-mentioned conventional methods, the inventor of this case has been eager to improve and innovate. After years of painstaking and meticulous research, he finally successfully developed the system and system for automatic construction and horizontal extension of service elements based on the dynamic cluster rule template. method.

In order to achieve the above object, the present invention proposes to provide a system and method for automatically constructing and horizontally extending service elements of a dynamic clustering rule template. Compared with the application service which generally expands service elements in a virtual machine manner, it takes several minutes to wait for the virtual machine to replicate. And the tedious time of starting the service element, the present invention utilizes the Linux Container (Linux container) lightweight startup technology, which can start the service element in a few seconds, and the coordination unit obtains cluster configuration information from the cache memory and immediately sets it, The entire process can be completed within a few seconds, which greatly increases the speed of cluster automation settings and service component expansion. In addition, service components that work in conjunction with application services can independently extend or reduce service nodes without affecting existing functions and states that are cooperating among service components. This rapid automated configuration method provides flexibility and improves service component combinations. Cluster construction efficiency.

Another object of the present invention is to provide a mechanism for storing dynamic cluster rule templates in cache memory to achieve rapid retrieval and automatic setting. A module is required to automatically construct clusters according to service component types and connection information and Configuration data, including the number of basic nodes, master-slave architecture mode, initial data settings, security policy distribution, etc., and generate corresponding cluster rule templates to be stored in the cache memory of the coordination unit (key-value) )in. The cluster rule template takes the service component type as the index value and corresponds to three major parts: the first part is the overall cluster information, such as the cluster architecture type (defining master / slave, peer-to-peer, or other types of architecture), the minimum number of cluster members, The maximum number of cluster members, or security policy distribution (such as secure connection rules). The second part is the cluster member information, such as the connection information of individual members, the running status, whether the cluster is a master member (runtime role is master node), and so on. The third part is the service information, such as the account, password, or related settings required by the service component. A service component type corresponds to a cluster rule template (processed in JSON or XML format), and a key-value data is stored in the cache memory. Then, another module periodically queries the storage device for the latest service component information, compares the cluster rule template in the cache memory, and automatically configures the cluster construction and service component cluster member changes. The latest data of the cluster members is then dynamically updated. Into the cluster rule template for the next execution.

A system for automatically constructing and horizontally extending service elements based on a dynamic cluster rule template, which mainly includes a service registration unit, which interfaces with the Container (container) management unit, and is responsible for monitoring and registering cluster member information in the service element , And then record the latest service and connection data of the cluster members in the storage unit; a cluster configuration management unit is to create a cluster rule template to automatically establish the connection information and configuration data required for the cluster according to the type of service component classification , And then store the service component type and the cluster rule template in the cache memory (key-value) of the coordination unit; a coordination unit periodically queries the storage unit for the latest service component information and compares the cache memory with the cache memory. The cluster rule template of. If there are new or changed service component members, the cluster rule template obtains the connection and configuration information, and performs automatic cluster construction and cluster member change settings, and then updates the latest data of the cluster members. Into the cluster rule template for use by the coordination unit at its next execution.

The connection information and configuration data include the number of basic nodes, the master-slave architecture mode, the initial data settings, and security policy distribution. The service component type includes the cluster overall information, cluster member information, and service information. The cluster overall information, The system includes cluster architecture types, which are defined as master / slave, peer-to-peer, minimum number of cluster members, maximum number of cluster members, or security policy distribution, such as secure connection rules; Cluster member information, including connection information of individual members, running status, whether it is the master member of the cluster, its runtime role is master node (master node); service information, including account, password, or related required service components Set value.

A method for automatically constructing and horizontally extending service elements based on a dynamic cluster rule template, which includes: setting a timer to periodically start a coordination unit and a service registration unit; a cluster configuration management unit establishes a cluster rule template according to a service element type; and a service registration unit Collect service and network connection information of all cluster members on a regular basis; a timer triggers the coordination unit to perform automated settings.

The automatic setting includes the following steps: the coordination unit queries the storage unit for the service type and connection data of all service components; whether all service components have been processed; if yes, the process ends; if not, one is selected. The processing service component finds the current cluster member, and then obtains the corresponding cluster rule template by using its service type as the index value; compares the cluster member list during the previous execution and checks whether the cluster member has changed. If not, Whether all service components have been processed and re-judged; if so, the cluster rule template is used to obtain connection and configuration information, and to perform automatic cluster construction and member change settings; update the current cluster member information to the cluster rule In the template; end the process.

The system and method for automatically constructing and horizontally extending service elements based on a dynamic cluster rule template provided by the present invention have the following advantages when compared with other conventional technologies:

1. It is to improve the speed of cluster automation setting and service component expansion, and help to solve the problem of processing a large number of messages in a short time.

2. The service registration component interfaces with the Container management unit, which is responsible for monitoring and registering the connection information within the cluster members, records it in the storage component, and automatically detects changes in the cluster members.

3. Provide a mechanism for storing dynamic cluster rule templates in cache memory to achieve quick retrieval and automatic setting.

102‧‧‧Application Services

104‧‧‧Service element one

106‧‧‧Service Element Two

108‧‧‧ Coordination Unit

110‧‧‧Service Registration Unit

112 ₁ ~ 112 _n ‧‧‧ cluster members

116 ₁ ~ 116 _X ‧‧‧ cluster members

118‧‧‧Storage unit

120‧‧‧ Cluster Configuration Management Unit

122‧‧‧Container Management Unit

124‧‧‧operation unit

202‧‧‧Cluster rule template

204‧‧‧ Service Component Type

206‧‧‧ cluster overall information

208‧‧‧ cluster member information

210‧‧‧Service Information

S302 ~ S308‧‧‧Process

S402 ~ S412‧‧‧‧Automatic setting process

Please refer to the detailed description of the present invention and its accompanying drawings to further understand the technical content and the purpose and effect of the present invention. The related drawings are: FIG. 1 is the automatic construction and horizontal extension of service elements based on the dynamic clustering rule template of the present invention Schematic diagram of the system and method architecture; Figure 2 is a cluster rule template architecture diagram of a system and method for automatically constructing and horizontally extending service elements based on dynamic cluster rule templates of the present invention; Figure 3 is a service element based on dynamic cluster rule templates of the present invention Flow chart of a system and method for automatic construction and horizontal extension; FIG. 4 is a flowchart of automatic setting of the system and method for automatic construction and horizontal extension of service components based on a dynamic cluster rule template according to the present invention.

In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not intended to limit the present invention.

Hereinafter, the present invention will be further described with reference to the accompanying drawings. Please refer to FIG. 1, which is a schematic diagram of a system and method for automatically constructing and horizontally extending service elements based on a dynamic cluster rule template according to the present invention. The system composition includes application service 102 , Service element one 104, service element two 106, coordination unit 108, service registration unit 110, storage unit 118, cluster configuration management unit 120, and container management unit 122, of which service element one 104 and service element two 106 are for support The requirements of application service functions can include database services, cache services, or queue services. The service component one 104 and service component two 106 can be a single program or a clustered service component composed of multiple running programs, such as services. Component 1 104 cluster members 114 ₁ , 114 ₂ , ... 114 _n , service component 2 106 cluster members 116 ₁ , 116 ₂ , 116 _x , Container Management Unit 122 is a technology that uses LinuxContainer (Linux container) to reduce weight and share underlying resources. Provides the functions of resource configuration and service element deployment. The service registration unit 110 is responsible for regularly monitoring services. Add or remove cluster members from service component one 104 and service component two 106, and collect network connection information of all cluster members. Storage unit 118 is used to store connection information collected by service registration unit 110. The cluster The configuration management unit 120 is the connection information and configuration data required for automatically constructing a cluster according to the type classification of service component one 104 and service component two 106, including the number of basic nodes, master-slave architecture mode, initial data settings, security Policy dispatch, etc., and generate corresponding cluster rule templates stored in the cache memory of the coordination unit 108, the cache memory of which is stored in a key-value key-value structure, the coordination unit 108, to the storage unit 118 regularly Query the latest service component information, compare the cluster rule templates in the cache memory. If there are new or changed service component members, the cluster rule template obtains the connection and configuration information, and performs automatic cluster construction and service component member changes. Its settings.

An application service 102, whose cluster members are 112 ₁ , 112 ₂ , ... 112 _m , uses service element 104 and service element 106 to support the function of the application service, and cluster members 114 ₁ , 114 _{2 of} service element 104 … 114 _n , cluster members can have master / slave or peer-to-peer architectures to provide specific services, such as databases, caches, or queues; cluster architectures and service types are only examples, but Not limited to this.

The service element two 106 includes cluster members 116 ₁ , 116 ₂ , ... 116 _x , and the cluster may have the same or different architecture as the service element one 106 and provide another specific service to the outside. Service element one 104 and service element two 106 may have a cooperative relationship with each other, such as client / server (client / server mode), publish / subscribe (publish / subscribe mode) or producer / consumer (producer / consumer mode) ), Etc. to provide more complex functions.

The Container Management Unit 122 utilizes Linux Container technology, and can also use Docker engine technology. It can start Service Element 104 and Service Element 106, and trigger an event at startup to notify Service Registration Unit 110 about Service Element 1 104 and Service Element. Two 106 network connection data. An API is provided to accept the request from the request coordination unit 108 and execute the configuration settings for the service element one 104 and the service element two 106.

The service registration unit 110 is responsible for querying and recording the cluster members 114 ₁ , 114 ₂ , ... 114 _n , 116 ₁ , 116 ₂ , ... 116 _x real-time network through the Container management unit 122 Connection information, including service types.

The storage unit 118 is mainly used for storing service and connection information collected by the service registration unit 110.

The cluster configuration management unit 120 categorizes the connection information and configuration data required for automatically constructing a cluster according to the type of service component, including the number of basic nodes, the master-slave architecture mode, initial data settings, security policy distribution, etc., and generates corresponding The cluster rule template is stored in a cache memory of the coordination unit 108, and the cache memory thereof is stored in a key-value (key-value) key-value structure.

The coordination unit 108 queries the storage unit 118 for the transaction data of the service element one 104 and the service element two 106 regularly, and can be linked to the service element one 106 and the service element two 108, and automatically adjusts it with the cluster members 114 ₁ , 114 ₂ , ... 114 _n , 116 ₁ , 116 ₂ , ... 116 _x related configuration settings can also be indirectly communicated with service element 104 and service element 106 through the API provided by the Container management unit 122.

The service element one 104 and the service element two 106 correspond to a group, and the number of service elements is only an example, but is not limited thereto.

The operation unit 124 is a physical host, and the container management unit 122, the service element 1, 104, the service element 108, and the service registration unit 110 can be run thereon. If necessary, multiple hosts can also operate together to provide more computing resources.

Please refer to FIG. 2, which is a cluster rule template architecture diagram of a system and method for automatically constructing and horizontally extending service elements based on a dynamic cluster rule template according to the present invention. A cluster rule template 202 uses service element type 204 as an index value, corresponding to There are three major parts: the first part is the overall cluster information 206, such as the cluster architecture type, which is defined as master / slave, peer-to-peer or other types of architecture, the minimum number of cluster members, the maximum number of cluster members, or security policy delivery, Such as secure wiring rules.

The second part is the cluster member information 208, such as the connection information of individual members, the running status, whether it is the master member of the cluster, and the runtime role as master / leader node.

The third part is the service information 210, such as the account, password, or related settings required by the service component. The data contained in the three parts of the cluster overall information 206, cluster member information 208, and service information 210 may be stored in JSON, XML, or other formats that can define structured data.

Please refer to FIG. 3, which is a flowchart of a system and method for automatically constructing and horizontally extending service elements based on a dynamic cluster rule template according to the present invention. The steps are as follows: S302 sets a timer to start the coordination unit and the service registration unit regularly; S304 cluster configuration management unit creates cluster rule templates according to service component types; S306 service registration unit regularly collects service and network connection information of all cluster members; S308 timer triggers the coordination unit to perform automatic settings.

Please refer to FIG. 4, which is a flowchart of automatic setting of a system and method for automatically constructing and horizontally extending service elements based on a dynamic cluster rule template according to the present invention. The steps are as follows: S402 The coordination unit queries the storage unit for the services of all service elements. Type and connection data; S404 Whether all service components have been processed, if yes, then the process ends; S406 If no, select an unprocessed service component, check the current cluster member, and then use its service type The corresponding cluster rule template is obtained from the index value; S408 compares the cluster member list during the previous execution, and checks whether there are any changes in the cluster members. If not, return to S404 whether all service components have been processed and re-judge; If so, S410 obtains the connection and configuration information from the cluster rule template, and performs automatic cluster construction and member change settings; S412 updates the current cluster member information to the cluster rule template; and ends the process.

The above detailed description is a specific description of a feasible embodiment of the present invention, but this embodiment is not intended to limit the patent scope of the present invention. Any equivalent implementation or change that does not depart from the technical spirit of the present invention should be included in Within the scope of the patent in this case.

To sum up, this case is not only innovative in terms of technical ideas, but also has many of the above-mentioned effects that are not used by traditional methods. It has fully met the requirements of statutory invention patents that are novel and progressive. To approve this invention patent application, to encourage invention, to the utmost convenience.

Claims (8)

    A system for automatically constructing and horizontally extending service elements based on a dynamic cluster rule template, which mainly includes: a service registration unit, which interfaces with the Container (container) management unit, and is responsible for monitoring and registering cluster members in the service element. Information, and then record the latest service and connection data of the cluster members in the storage unit; a cluster configuration management unit, which is to create a cluster rule template to automatically construct the cluster connection information and configuration required by the type of service component Data, and then store the service component type and the cluster rule template in the cache memory of the coordination unit; a coordination unit periodically queries the storage unit for the latest service component information, and compares the cluster rule template in the cache memory If there are new or changed service component members, the cluster rule template obtains the connection and configuration information, and performs the cluster automatic construction and cluster member change settings, and then the latest data of the cluster members is updated to the cluster rule template. Medium for use by the coordination unit at its next execution.         A system for automatically constructing and horizontally extending service components based on a dynamic cluster rule template as described in item 1 of the scope of patent application, wherein the connection information and configuration data include the number of basic nodes, master-slave architecture mode, and initial data settings 2, security policy delivery.         The system for automatically constructing and horizontally extending service components based on a dynamic cluster rule template as described in item 1 of the scope of the patent application, wherein the service component type includes cluster overall information, cluster member information, and service information.         A system for automatically constructing and horizontally extending service components based on a dynamic cluster rule template as described in item 3 of the scope of the patent application, wherein the overall cluster information includes the cluster architecture type, which is defined as master / slave (master / slave) , Peer-to-peer (point-to-point), minimum number of cluster members, maximum number of cluster members, or security policy delivery, such as secure connection rules.         A system for automatically constructing and horizontally extending service components based on a dynamic cluster rule template as described in item 3 of the scope of the patent application, wherein the cluster member information includes connection information of individual members, operating status, and whether the cluster is a master member. Its runtime role is master node.         The system for automatically constructing and horizontally extending service elements based on a dynamic cluster rule template as described in item 3 of the scope of patent application, wherein the service information includes account, password, or related setting values required by the service element.         A method for automatically constructing and horizontally extending service elements based on a dynamic cluster rule template, which includes: setting a timer to periodically start a coordination unit and a service registration unit; a cluster configuration management unit establishes a cluster rule template according to a service element type; and a service registration unit Collect service and network connection information of all cluster members on a regular basis; a timer triggers the coordination unit to perform automated settings.         The method for automatically constructing and horizontally extending service elements based on a dynamic cluster rule template as described in item 7 of the scope of the patent application, wherein the process of the automatic setting includes: the coordination unit queries the storage unit for the service types and connections of all service elements. Online data; whether all service components have been processed, if yes, the process ends; if not, select an unprocessed service component, find the current cluster member, and then use the service type as the index value to obtain the corresponding The cluster rule template of; compare the cluster member list at the previous execution, check whether the cluster members have changed, and if not, return all service components to complete the processing, and re-judgment; if yes, the cluster rule template Obtain connection and configuration information, perform automatic cluster construction and member change settings; update the current cluster member information to the cluster rule template; end the process.