A hyper-converged server control system, construction method, medium and application

Abstract

The present invention belongs to the technical field of resource allocation, and discloses a hyper-converged server control system, a construction method, a medium and an application, including: a hyper-converged server deployment module for deploying hyper-converged servers and server nodes in the first area and the second area; a data storage and sharing module for distributed storage and sharing of sensor data; a resource utilization status table establishment module for establishing a resource utilization status table in the main virtual server node; an integration and migration module of virtual server nodes for integrating virtual server nodes and dynamically migrating different virtual server resources. The hyper-converged server of the present invention unifies and integrates traditional computing resources and storage management functions, directly distributes data to storage for computing, reduces the data transmission volume and bandwidth between sensor devices and the data center, and at the same time integrates a variety of virtual technologies to achieve resource load balancing among multiple services.

Description

Technical Field

[0001] The present invention belongs to the technical field of resource allocation, and particularly relates to a hyper-converged server control system, a construction method, a medium and an application. Background Art

[0002] At present, with the development of ubiquitous Internet of Things sensing and detection technologies, various intelligent systems and applications continuously generate a large amount of data in practice. For different professional sensing devices, data collection, computing, and communication resources at the same site are not fully shared and utilized, resulting in the phenomenon of repeated construction of different professional acquisition terminals and repeated data collection. The storage and network environment are complex, and related data processing work shows a trend of large-scale, distributed, parallelized, and diversified development. Traditional solutions can no longer efficiently meet such requirements. Most sensing devices need to transmit data to the background to achieve cross-professional and cross-system data sharing applications or associated calculations. There is often a delay from the generation of data at the source to cross-professional applications, which cannot meet the requirements for high instantaneity of services. Moreover, the traditional virtualization architecture lacks a business resource scheduling method that takes into account both load balancing and energy-saving optimization, which may lead to the situation where virtual machines on some physical machines occupy excessive resource loads and the performance of virtual machines on these physical machines decreases. At the same time, the amount of resources consumed by some physical machines may be very low, and the overall cluster carrying services will show a serious imbalance in resource consumption. It cannot adapt to different situations (data-intensive and memory-sensitive, computing-intensive and CPU-sensitive) and dynamic changes in resource loads, which may lead to load imbalance and load skew.

[0003] Through the above analysis, the problems and defects existing in the prior art are: the data transmission volume and bandwidth between existing sensor devices and data centers are large, and at the same time, the resource loads among multiple services are unbalanced.

[0004] The difficulty in solving the above problems and defects is: under the traditional virtualization architecture, computing and storage are separated, and there are performance bottlenecks in reading data by sensor devices and data centers. The data transmission volume and bandwidth are large, and it cannot meet the requirements for high timeliness of services. Limited by performance and technical bottlenecks, the current traditional server virtualization technology is inefficient on domestic CPU platforms and lacks a resource scheduling scheme between hyper-converged server nodes, which easily leads to unbalanced resource loads among multiple services.

[0005] The significance of solving the above problems and defects is as follows: The hyper-converged server control system of the present invention adopts an ARM architecture and a domestic CPU system, achieving complete autonomy and controllability of domestic hyper-converged servers. It unifies and integrates traditional computing resources and storage management functions, directly distributes data to storage for computing, reduces the data transmission volume and bandwidth between sensor devices and the data center, reduces resource waste caused by accessing data between multiple services, improves service immediacy, and at the same time integrates a variety of virtual technologies to reduce resource waste caused by accessing data between multiple services, realizing resource sharing and maximizing resource utilization among multiple services. Summary of the Invention

[0006] In view of the problems existing in the prior art, the present invention provides a hyper-converged server control system, a construction method, a medium and an application.

[0007] The present invention is implemented as follows. A hyper-converged server control system, the hyper-converged server control system includes:

[0008] A hyper-converged server deployment module for deploying hyper-converged servers and server nodes in the first area and the second area;

[0009] A data storage and sharing module for distributed storage and sharing of sensor data;

[0010] A resource utilization status table establishment module for establishing a resource utilization status table in the main virtual server node;

[0011] An integration and migration module of virtual server nodes for integrating virtual server nodes and dynamically migrating different virtual server resources according to the resource utilization status of each virtual server node;

[0012] Further, the data storage and sharing module includes:

[0013] A local directory for storing data returned by edge sensors;

[0014] A data sharing directory, which is set on the main virtual server node and is used to store shareable service data using a distributed Ceph architecture.

[0015] Further, the two attributes of the resource utilization status table are the virtual server node name and the current status respectively; wherein, the current status includes: an overloaded state, a high-load state, and a low-load state.

[0016] Another object of the present invention is to provide a hyper-converged server control system construction method for constructing the hyper-converged server control system, the hyper-converged server control system construction method includes:

[0017] Step 1: Deploy a hyper-converged server within the first area range of the sensor, and use virtual technology to create multiple virtual server nodes;

[0018] Step 2: Perform data storage and sharing; determine the running status of the virtual server nodes, and establish a resource utilization status table for the virtual server nodes;

[0019] Step 3: Integrate and migrate the virtual server nodes.

[0020] Furthermore, in Step 1, the deploying of the hyper-converged server within the first area range of the sensor and using virtual technology to create multiple virtual server nodes includes:

[0021] (1) Build a hyper-converged server under the ARM architecture and determine the network topology of the ubiquitous Internet of Things;

[0022] (2) According to the locations of the edge sensor nodes in the perception layer, delimit the first area range, and deploy a hyper-converged server within the first area. The hyper-converged server stores and calculates data for each edge sensor node within the first area range;

[0023] (3) Use virtual technology to create multiple virtual server nodes, delimit the second area range, and deploy one or more basic services within the second area range on each virtual server node.

[0024] Furthermore, the second area range is a subset of the first area range. The virtual server nodes serve as the storage and calculation devices for the sensor nodes within the second area range, and the sum of the services running on the virtual server nodes should correspond to the sum of all services within the first area range.

[0025] Furthermore, in Step 2, the performing of data storage and sharing includes:

[0026] Create a main virtual server node in the hyper-converged server, set read permissions for the specified virtual server nodes according to business requirements; set up a data sharing file directory locally in the main virtual server node to store the data returned by the edge sensors; specify a mount point locally in the specified virtual server node for access by other virtual server nodes with permissions.

[0027] Furthermore, in Step 2, the determining of the running status of the virtual server nodes and establishing a resource utilization status table for the virtual server nodes includes:

[0028] Establish a main virtual server node to monitor the running status of other virtual server nodes, establish a resource utilization status table in the main virtual server node, and set load status thresholds respectively.

[0029] Further, the load status threshold includes:

[0030] The threshold for the overloaded state is the critical value of the hyper-converged server resource utilization rate that causes the virtual server node to crash.

[0031] The threshold for the high-load state is the critical value of the hyper-converged server resource utilization rate that causes the performance of the virtual server node to seriously decline.

[0032] The threshold for the low-load state is the critical value of the hyper-converged server resource utilization rate when the energy efficiency output of the hyper-converged server is relatively low.

[0033] Further, in step three, the integration and migration of the virtual server nodes include:

[0034] 1) Create a separate main virtual server node on the hyper-converged server, monitor the running status of other virtual server nodes, and the main virtual server node periodically judges the current status of other different virtual server nodes.

[0035] 2) Set a sliding time window, and periodically update the resource utilization status table in the main virtual server node; sort in descending order according to the resource utilization status of the virtual server nodes, and reserve a certain amount of CPU and memory resources for each virtual server node in the hyper-converged server.

[0036] 3) Place the virtual server nodes into the hyper-converged server according to the strategy of giving priority to the largest resource occupancy, allocate resources to each virtual server node; integrate the virtual server nodes, and dynamically migrate different virtual server resources according to the resource utilization status of each virtual server node.

[0037] Further, the main virtual server node periodically judging the current status of other different virtual server nodes includes:

[0038] Through the main virtual server node, obtain the relevant resource dimensions occupied by the virtual server node and its resource utilization situation, calculate the current hyper-converged server resource utilization rate, compare it with the preset load status threshold, and determine the current status of the virtual server node.

[0039] Further, the discrimination dimensions of the resource utilization situation mainly include CPU resources, memory resources, and network IO resources.

[0040] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the processor executes the following steps:

[0041] Step one, deploy a hyper-converged server within the first area range of the sensor, and use virtual technology to create multiple virtual server nodes.

[0042] Step 2: Perform data storage and sharing; determine the running status of the virtual server nodes and establish a resource utilization status table for the virtual server nodes.

[0043] Step 3: Integrate and migrate the virtual server nodes.

[0044] Another object of the present invention is to provide an information data processing terminal, which is used to implement the hyper-converged server control system described above.

[0045] Combining all the above technical solutions, the advantages and positive effects of the present invention are as follows: The hyper-converged server of the present invention unifies and integrates the traditional computing resources and storage management functions, directly distributes data to the storage for computing, reduces the data transmission volume and bandwidth between the sensor devices and the data center, and at the same time integrates a variety of virtual technologies to achieve resource load balancing among multiple services.

[0046] The present invention uses the FT processor with the ARM architecture as the CPU of the physical machine of the hyper-converged server of the present invention, and designs the hyper-converged service on this hardware basis to achieve the autonomy and controllability of the domestic hyper-converged server; realizes the shared storage of data through the Ceph architecture; maximizes the resource allocation and utilization of the hyper-converged server by establishing a resource utilization status table, and realizes the diversity of services.

[0047] The hardware CPU of the hyper-converged server of the present invention uses the FT processor with the ARM architecture to achieve the autonomy and controllability of the domestic hyper-converged server; the data can be shared, and the shared storage of data is realized by using the distributed Ceph architecture, reducing the resource waste caused by accessing data between multiple services; the resource utilization status is visible, and through the main virtual server node, the resource utilization rate of the current hyper-converged server can be calculated; the resources are maximally utilized, and resources are allocated to each virtual server node according to the strategy of giving priority to the largest occupied resources. Description of the Drawings

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required to be used in the embodiments of the present application. Obviously, the following described drawings are only some embodiments of the present application, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0049] Figure 1 It is a block diagram of the hyper-converged server control system provided by the embodiment of the present invention.

[0050] Figure 2 It is a schematic structural diagram of the hyper-converged server control system provided by the embodiment of the present invention;

[0051] In the figure: 1. Hyper-converged server deployment module; 2. Data storage and sharing module; 3. Resource utilization status table establishment module; 4. Integration and migration module of virtual server nodes.

[0052] Figure 3 It is a schematic diagram of the method for constructing a hyper-converged server control system provided by an embodiment of the present invention.

[0053] Figure 4 It is a flowchart of the method for constructing a hyper-converged server control system provided by an embodiment of the present invention.

[0054] Figure 5 It is a deployment schematic diagram in the ubiquitous Internet of Things environment provided by an embodiment of the present invention.

[0055] Figure 6 It is a schematic diagram of the table structure of the resource utilization status table provided by an embodiment of the present invention.

[0056] Figure 7 It is an integration and migration schematic diagram provided by an embodiment of the present invention. Detailed implementation manners

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the described embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0058] In view of the problems existing in the prior art, the present invention provides a hyper-converged server control system, a construction method, a medium, and an application. The characteristics are that by deploying virtual server nodes through the hyper-converged server control system, effective operations of different services in different environments are realized. By establishing a main virtual server node, the status monitoring of other virtual server nodes is realized, and based on the status utilization table, the integration and migration of virtual servers are realized, so as to maximize the utilization of hyper-converged server resources while taking into account service sensitivity, load balancing, and energy-saving optimization.

[0059] The technical solutions of the present invention will be described in detail below with reference to the accompanying drawings.

[0060] As Figure 1 - Figure 2 shown, the hyper-converged server control system provided by an embodiment of the present invention includes:

[0061] The hyper-converged server deployment module 1 is used for deploying hyper-converged servers and server nodes in the first area and the second area;

[0062] The data storage and sharing module 2 is used for distributed storage and sharing of sensor data;

[0063] The resource utilization status table establishment module 3 is used for establishing a resource utilization status table in the main virtual server node;

[0064] The integration and migration module 4 of virtual server nodes is used for integrating virtual server nodes and dynamically migrating different virtual server resources according to the resource utilization status of each virtual server node;

[0065] The data storage and sharing module 2 provided by the embodiment of the present invention includes:

[0066] The local directory is used for storing the data returned by the edge sensors;

[0067] The data sharing directory is set on the main virtual server node and is used for storing shareable service data by adopting a distributed Ceph architecture.

[0068] The two attributes of the resource utilization status table provided by the embodiment of the present invention are the virtual server node name and the current status; wherein, the current status includes: overloaded status, high load and low load status.

[0069] Such as Figure 3 - Figure 4 As shown, the method for constructing a hyper-converged server control system provided by the embodiment of the present invention includes the following steps:

[0070] S101, deploying a hyper-converged server within the first area range of the sensors and creating multiple virtual server nodes by using virtual technology;

[0071] S102, performing data storage and sharing; determining the running status of the virtual server nodes and establishing a resource utilization status table of the virtual server nodes;

[0072] S103, performing integration and migration of the virtual server nodes.

[0073] In step S101, the deploying a hyper-converged server within the first area range of the sensors and creating multiple virtual server nodes by using virtual technology provided by the embodiment of the present invention includes:

[0074] (1) Constructing a hyper-converged server under the ARM architecture and determining the network topology structure of the ubiquitous Internet of Things;

[0075] (2) According to the positions of the edge sensor nodes in the perception layer, delimiting the first area range and deploying a hyper-converged server within the first area. The hyper-converged server performs data storage and calculation for each edge sensor node within the first area range;

[0076] (3) Create multiple virtual server nodes using virtual technology, delimit the scope of the second area, and deploy one or more basic services in the second area on each virtual server node.

[0077] The scope of the second area provided by the embodiments of the present invention is a subset of the scope of the first area. The virtual server node serves as the storage and computing device for the sensor nodes within the scope of the second area, and the total sum of the services running on the virtual server node should correspond to the total sum of all services within the scope of the first area.

[0078] In step S102, the data storage and sharing provided by the embodiments of the present invention include:

[0079] Create a main virtual server node in the hyper-converged server, set readable permissions for the specified virtual server node according to business requirements; set up a data sharing file directory locally in the main virtual server node to store the data returned by the edge sensors; specify a mount point locally in the specified virtual server node for access by other virtual server nodes with permissions.

[0080] In step S102, the determination of the running state of the virtual server node and the establishment of the resource utilization status table of the virtual server node provided by the embodiments of the present invention include:

[0081] Establish a main virtual server node to monitor the running states of other virtual server nodes, establish a resource utilization status table in the main virtual server node, and set the load status thresholds respectively.

[0082] The load status thresholds provided by the embodiments of the present invention include:

[0083] The threshold for the overloaded state is the critical value of the resource utilization rate of the hyper-converged server that causes the virtual server node to crash;

[0084] The threshold for the high-load state is the critical value of the resource utilization rate of the hyper-converged server that causes the performance of the virtual server node to seriously decline;

[0085] The threshold for the low-load state is the critical value of the resource utilization rate of the hyper-converged server when the energy efficiency output of the hyper-converged server is relatively low.

[0086] In step S103, the integration and migration of the virtual server node provided by the embodiments of the present invention include:

[0087] 1) Create a separate main virtual server node on the hyper-converged server to monitor the running states of other virtual server nodes, and the main virtual server node periodically determines the current states of other different virtual server nodes;

[0088] 2) Set a sliding time window to periodically update the resource utilization status table in the main virtual server node; sort the virtual server nodes in descending order according to their resource utilization status, and reserve a certain amount of CPU and memory resources for each virtual server node in the hyper-converged server;

[0089] 3) Place the virtual server nodes into the hyper-converged server according to the strategy of giving priority to the node with the largest occupied resources, and allocate resources to each virtual server node; integrate the virtual server nodes, and dynamically migrate different virtual server resources according to the resource utilization status of each virtual server node.

[0090] The main virtual server node provided by the embodiment of the present invention periodically determines the current status of other different virtual server nodes, including:

[0091] Through the main virtual server node, obtain the relevant resource dimensions occupied by the virtual server node and its resource utilization situation, calculate the resource utilization rate of the current hyper-converged server, compare it with the preset load status threshold, and determine the current status of the virtual server node.

[0092] The discrimination dimensions of the resource utilization situation provided by the embodiment of the present invention mainly include CPU resources, memory resources, and network IO resources.

[0093] The technical effects of the present invention will be further described below in conjunction with specific embodiments.

[0094] Embodiment:

[0095] The hyper-converged server system provided by the embodiment of the present invention for the ubiquitous Internet of Things environment uses the hyper-converged server under the ARM architecture as the physical server node, realizes the effective operation of different services in the ubiquitous Internet of Things environment by deploying virtual server nodes, realizes the status monitoring of other virtual server nodes by establishing a main virtual server node, and realizes the integration and migration of virtual servers based on the status utilization table, so as to maximize the utilization of hyper-converged server resources.

[0096] As shown in the attached Figure 1 figure, the system block diagram provided by the embodiment of the present invention is specifically:

[0097] As a high-performance physical server required for data storage and computing, the hyper-converged server uses virtualization technology to create several virtual server nodes and selects one of them as the primary virtual server node. The hyper-converged server provides services for edge sensor nodes within the first regional scope, and one or more basic services within the second regional scope can run on each virtual server node. The primary virtual server node is different from other virtual server nodes. The primary virtual server node can establish a resource utilization status table for each virtual server node, judge the current status of the virtual server node through information such as memory resources and network IO resources, and place the virtual server node into the hyper-converged server according to the strategy of giving priority to the node with the largest occupied resources, and integrate and migrate the system resources of the virtual server node. In addition, a data sharing directory is established in the primary virtual server node to store shareable business data for use by other virtual server nodes, while other virtual server nodes specify the storage of data returned by the edge sensors in the local directory by establishing a local directory.

[0098] As shown in the Figure 3 appendix, the hyper-converged server provided by the embodiment of the present invention includes the following steps:

[0099] Step 1: Deploy the hyper-converged server: Build a hyper-converged server under the ARM architecture through a processor of the domestic Feiteng 2000 model, divide the regional scope according to the network topology of the current application environment and the positions of edge nodes in the sensing layer, deploy physical nodes of the hyper-converged server, and use virtualization technology to allocate virtual server nodes for different services;

[0100] Step 2: Store and share data among multiple services: One or more services run on each virtual server node. By setting up a shared file directory in the hyper-converged server and using the distributed Ceph architecture, the data storage and sharing of the virtual server node at the local specified mount point are completed;

[0101] Step 3: Establish a resource utilization status table for virtual server nodes: Establish a resource utilization status table for each virtual server node, and set the status value thresholds for the overloaded state, high-load state, and low-load state respectively, so that the established primary virtual server node can compare information such as CPU resources, memory resources, and network IO resources, judge the current status of the virtual server node, and monitor the running status of the virtual server node;

[0102] Step 4: Integrate and migrate virtual server nodes: The primary virtual server node sets a sliding time window, periodically updates the resource utilization status table in the primary virtual server node, sorts the virtual server nodes in descending order according to their resource utilization status, reserves a certain amount of CPU and memory resources for each virtual server node in the hyper-converged server, and places the virtual server nodes into the hyper-converged server according to the strategy of giving priority to the nodes with the most occupied resources, so as to integrate and migrate the system resources of the virtual server nodes.

[0103] The system block diagram provided by the embodiment of the present invention is specifically as follows:

[0104] The hyper-converged server, as a high-performance physical server required for data storage and computing, uses virtual technology to create several virtual server nodes and selects one of them as the primary virtual server node; the hyper-converged server provides services for each edge sensor node within the first area range, and one or more basic services within the second area range can run on each virtual server node; the primary virtual server node is different from other virtual server nodes. The primary virtual server node can establish a resource utilization status table for each virtual server node, judge the current status of the virtual server node through information such as memory resources and network IO resources, place the virtual server nodes into the hyper-converged server according to the strategy of giving priority to the nodes with the most occupied resources, and integrate and migrate the system resources of the virtual server nodes; in addition, a data sharing directory is established in the primary virtual server node for storing shareable business data, which is used by other virtual server nodes, while other virtual server nodes specify the data returned by the edge sensors to be stored in the local directory by establishing a local directory.

[0105] The hardware configuration of the hyper-converged server provided by the embodiment of the present invention, compared with other hyper-converged servers on the market, the CPU used in the hyper-converged server described in the embodiment of the present invention is configured as: Feiteng FT-2000, adopting the ARM64 architecture, with 63 FTC662 processor cores, as shown below.

[0106]

[0107] The deployment method of the hyper-converged server provided by the embodiment of the present invention in an application scenario, the method is specifically as follows:

[0108] First, determine the network topology of the ubiquitous Internet of Things. According to the locations of the edge sensor nodes in the perception layer, delimit the first area range, so that the hyper-converged server serves as a high-performance physical server required for data storage and computing of each edge sensor node within the first area range;

[0109] Then, virtual technology is used to create multiple virtual server nodes, and the scope of the second area is delimited so that one or more basic services in the scope of the second area can run on each virtual server node;

[0110] Finally, on the premise that the scope of the second area is a subset of the scope of the first area, ensure that the total sum of the services running on the virtual server nodes is greater than or equal to the total sum of all services within the scope of the first area.

[0111] The table structure of the resource utilization status table provided by the embodiments of the present invention is specifically as follows:

[0112] The resource utilization status table is established by the main virtual server node. Each record in the table represents the current status of a virtual node, and each record has two attributes, which respectively represent the name of the virtual node and the calculated current status index; among them, according to the resource utilization of each virtual server node, the resource utilization status can be divided into three types: overloaded status, high load, and low load status.

[0113] The method for integrating and migrating virtual server nodes provided by the embodiments of the present invention is specifically as follows:

[0114] First, set the size of the sliding time window and periodically update the resource utilization status table in the main virtual server node. The instantaneous value of the resource utilization rate is not used during the update process of the status table. Instead, through the setting of the size of the sliding time window, it shows the average resource utilization rate of a physical machine in a certain resource dimension within a certain period of time. Only when the average resource utilization rate within a certain period of time exceeds or is lower than the status threshold, will the integration and migration of virtual server nodes in the hyper-converged server be carried out;

[0115] Next, perform a descending order sorting according to the resource utilization status of the virtual server nodes, and reserve a certain amount of CPU and memory resources for each virtual server node in the hyper-converged server;

[0116] Finally, place the virtual server nodes into the hyper-converged server according to the strategy of giving priority to the largest occupied resources, so as to realize the integration and migration of virtual server nodes.

[0117] The hyper-converged server control system, construction method, medium and application provided in the embodiments of the present invention include: a hyper-converged server deployment module for deploying hyper-converged servers and server nodes in the first and second regions; a data storage and sharing module for distributed storage and sharing of sensor data; a resource utilization status table establishment module for establishing a resource utilization status table in the main virtual server node; and an integration and migration module of virtual server nodes for integrating virtual server nodes and judging the resource utilization status of each virtual server node according to the resource utilization status table, and dynamically migrating different virtual server resources. In the embodiments of the present invention, the hyper-converged server integrates traditional computing resources and storage management functions, directly distributes data to storage for computing, reduces the data transmission volume and bandwidth between sensor devices and the data center, and at the same time integrates a variety of virtual technologies to achieve resource load balancing among multiple services, with practicability and ease of use.

[0118] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those of ordinary skill in the art can understand that the above devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code is provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuits of programmable hardware devices such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, or can be implemented by a combination of the above hardware circuits and software, such as firmware.

[0119] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention by those skilled in the art within the technical scope disclosed by the present invention shall be covered by the protection scope of the present invention.

Claims

1. A hyper-converged server control system, characterized in that, The hyper-converged server control system includes: A hyper-converged server deployment module for deploying hyper-converged servers and server nodes in the first area and the second area; A data storage and sharing module for distributed storage and sharing of sensor data; A resource utilization status table establishment module for establishing a resource utilization status table in the main virtual server node; An integration and migration module for virtual server nodes for integrating virtual server nodes and dynamically migrating different virtual server resources according to the resource utilization status of each virtual server node; The construction method of the hyper-converged server control system includes: Step 1: Deploy a hyper-converged server within the range of the first area of the sensor and create multiple virtual server nodes using virtual technology; Step 2: Perform data storage and sharing; determine the running status of the virtual server nodes and establish a resource utilization status table for the virtual server nodes; Step 3: Perform integration and migration of the virtual server nodes; In Step 1, the deployment of the hyper-converged server within the range of the first area of the sensor and the creation of multiple virtual server nodes using virtual technology include: (1) Construct a hyper-converged server under the ARM architecture and determine the network topology of the ubiquitous Internet of Things; (2) According to the locations of the edge sensor nodes in the perception layer, delimit the range of the first area and deploy a hyper-converged server within the first area. The hyper-converged server stores and calculates data for each edge sensor node within the range of the first area; (3) Use virtual technology to create multiple virtual server nodes, delimit the range of the second area, and deploy one or more basic services within the range of the second area on each virtual server node; The range of the second area is a subset of the range of the first area. The virtual server nodes serve as storage and computing devices for the sensor nodes within the range of the second area, and the total of the services running on the virtual server nodes should correspond to the total of all services within the range of the first area.

2. The hyper-converged server control system according to claim 1, wherein The data storage and sharing module includes: A local directory for storing data returned by the edge sensors; A data sharing directory set on the main virtual server node for storing shareable service data using the distributed Ceph architecture.

3. The hyper-converged server control system according to claim 1, wherein The two attributes of the resource utilization status table are the virtual server node name and the current status; among them, the current status includes: overloaded status, high-load status, and low-load status.

4. The hyper-converged server control system according to claim 1, wherein In Step 2, the performance of data storage and sharing includes: Create a main virtual server node in the hyper-converged server, set readable permissions for the specified virtual server node according to service requirements; set a data sharing file directory locally on the main virtual server node using the distributed Ceph architecture to store data returned by the edge sensors; specify a mount point locally on the specified virtual server node to be accessed by other virtual server nodes with permissions; In Step 2, the determination of the running status of the virtual server nodes and the establishment of a resource utilization status table for the virtual server nodes include: A main virtual server node is established to monitor the running status of other virtual server nodes. A resource utilization status table is established in the main virtual server node, and load status thresholds are set respectively.

5. The hyper-converged server control system according to claim 4, wherein The load status thresholds include: The threshold for the overloaded state is the critical value of the hyper-converged server resource utilization rate that causes the virtual server node to crash. The threshold for the high-load state is the critical value of the hyper-converged server resource utilization rate that causes the performance of the virtual server node to seriously decline. The threshold for the low-load state is the critical value of the hyper-converged server resource utilization rate when the energy efficiency output of the hyper-converged server is relatively low.

6. The hyper-converged server control system according to claim 1, wherein In step three, the integration and migration of the virtual server nodes include: 1) Create a separate main virtual server node on the hyper-converged server to monitor the running status of other virtual server nodes. The main virtual server node periodically judges the current status of other different virtual server nodes. 2) Set a sliding time window and periodically update the resource utilization status table in the main virtual server node; sort in descending order according to the resource utilization status of the virtual server nodes, and reserve a certain amount of CPU and memory resources for each virtual server node in the hyper-converged server. 3) Place the virtual server nodes into the hyper-converged server according to the strategy of giving priority to the largest resource occupancy, and allocate resources to each virtual server node; integrate the virtual server nodes, and perform dynamic migration on different virtual server resources according to the resource utilization status of each virtual server node. The main virtual server node periodically judging the current status of other different virtual server nodes includes: Through the main virtual server node, obtain the relevant resource dimensions occupied by the virtual server node and its resource utilization situation, calculate the current hyper-converged server resource utilization rate, compare it with the preset load status threshold, and determine the current status of the virtual server node. The discriminant dimensions of the resource utilization situation include CPU resources, memory resources, and network IO resources.

7. A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor performs the following steps: Step one, deploy a hyper-converged server within the first area range of the sensor, and use virtual technology to create multiple virtual server nodes. Step 2, perform data storage and sharing; Determine the running status of the virtual server nodes, and establish a resource utilization status table of the virtual server nodes. Step three, perform integration and migration of the virtual server nodes. In step one, the deploying a hyper-converged server within the first area range of the sensor and using virtual technology to create multiple virtual server nodes includes: (1) Build a hyper-converged server under the ARM architecture and determine the network topology of the ubiquitous Internet of Things. (2) According to the locations of the edge sensor nodes in the perception layer, delimit the first area range, and deploy a hyper-converged server within the first area. The hyper-converged server stores and calculates data for each edge sensor node within the first area range. (3) Use virtual technology to create multiple virtual server nodes, delimit the second area range, and deploy one or more basic services within the second area range on each virtual server node. The range of the second region is a subset of the range of the first region. The virtual server node serves as the storage and computing device for the sensor nodes within the range of the second region, and the total sum of the services running on the virtual server node should correspond to the total sum of all services within the range of the first region.

8. An information data processing terminal, characterized in that, The information data processing terminal is used to implement the hyper-converged server control system described in any one of claims 1-3.

Images