CN115002041A - Node balance scheduling method, device, equipment and storage medium - Google Patents

Abstract

The application discloses a node equilibrium scheduling method, device, equipment and storage medium, which are applied to a server cluster, relate to the technical field of communication and comprise the following steps: acquiring a connection request sent by a client; sending the connection request to each server node so that each server node processes the connection request according to the request sequence in the request queue of the server node and generates corresponding response information; and feeding back the response information of each server node to the client so that the client screens out the server node with the shortest feedback time as a target node, and sending the data to be processed to the target node so that the target node processes the data to be processed. By the method, the imbalance shown by the working load of each node is utilized, the real-time node scheduling is automatically carried out according to the dynamic change of the node load, and the self-adaptive balance is achieved among the nodes. And the nodes do not need to communicate with each other, so the node scheduling is simple to realize and has lower cost.

Description

A node balance scheduling method, device, equipment and storage medium

technical field

The present invention relates to the field of communication technologies, and in particular, to a node balanced scheduling method, device, equipment and storage medium.

Background technique

The TCP (Transmission Control Protocol, Transmission Control Protocol) protocol is a transmission protocol specially designed to provide a reliable end-to-end byte stream on an unreliable Internet. The formal definition of TCP is given by RFC793, September 1981. Over time, many improvements have been made to it, and various bugs and inconsistencies have been gradually fixed. The TCP protocol works in the transport layer of the seven-layer network model. It is a communication protocol for WAN. Its purpose is to provide a communication method with the following characteristics between two communication endpoints when communicating across multiple networks: Based on Streaming method; connection-oriented; reliable communication method; try to reduce the bandwidth overhead of the system due to retransmission when the network condition is not good; communication connection maintenance is oriented to the two endpoints of communication, regardless of intermediate network segments and nodes . In order to ensure the reliability of the TCP connection, when the client wants to establish a TCP connection with the server, it needs to go through the process of "three-way handshake": the first handshake: At the beginning, both the client and the server are in the CLOSED state. When the server actively listens to a port, it will be in the LISTEN state. The client generates a random client_isn (initialization sequence number), places the sequence number in the sequence number field of the TCP header, and sets the SYN flag to 1, indicating a SYN message. The first SYN message is sent to the server, indicating that a connection is initiated to the server. The message does not contain application layer data, and then the client is in the SYN-SENT state. The second handshake: After the server receives the SYN message from the client, the server also randomly initializes its own server_isn (serial number), fills the serial number into the serial number field of the TCP header, and then sets the confirmation response number of the TCP header. Fill in the field client_isn+1, and then set the SYN and ACK flags to 1. Finally, the message is sent to the client, and the message does not contain application layer data, and then the server is in the SYN-RCVD state. The third handshake: After the client receives the server message, it also responds to the server with the last response message. First, the ACK flag in the TCP header of the response message is set to 1, and secondly, the confirmation response number field is filled in server_isn +1, finally send the message to the server, this time the message can carry the data from the client to the server, and then the client is in the ESTABLISHED state. After receiving the response message from the client, the server also enters the ESTABLISHED state.

In the prior art, usually according to the content of the connection request and the type of service, the workload of each node is often unbalanced, and this unbalance is unavoidable and changes over time. When transmitting data with the TCP protocol, there are several commonly used distributed task scheduling algorithms: Round-robin method: Round-robin algorithm is the simplest and easiest to implement among all algorithms. Round-robin method simply rotates linearly in a series of nodes. , the balancer sends the new request to the next node in the node table. And so on. This algorithm is widely used in DNS domain name round robin. However, simply applying round-robin DNS translation may result in continuous access to the same node, which interferes with normal network load balancing and makes the network balancing system unable to work efficiently. The round-robin method is typically suitable for situations where the processing power and performance of all nodes in the cluster are the same. In practical applications, it is generally more effective to use it in combination with other simple methods. Weighted method: The weighted algorithm distributes the load according to the priority or weight of the nodes. Weights are assumptions or estimates based on the capabilities of each node. The weighted method intelligence is used in conjunction with other methods and is a good complement to them. Hash method: Hash method is also called Hash method. It sends network requests to cluster nodes according to certain rules through a single-shot irreversible Hash function. Similar to simple weighting. Least connection method: For TCP connections, in the least connection method, the management node records all current active connections, and sends the next new request to the node with the least number of connections. The disadvantage is that some application layer sessions consume more system resources. Although the number of connections in the cluster is balanced, the processing volume may vary greatly, and the number of connections cannot reflect the real application load. Minimum missing method: In the minimum missing method, the management node records the request status of each node for a long time, and sends the next request to the node that has processed the fewest requests in history. Unlike the least join method, the minimum missing records the number of past joins rather than the current number of joins. Fastest response method: The management node records the network response time from itself to each cluster node, and assigns the next incoming connection request to the node with the shortest response time. In most LAN based clusters, the fastest response algorithm does not work very well, most modern systems connected to Ethernet, with partial load, can respond in 1ms or less, which makes this method Pointless.

In summary, how to use the unbalanced workload of such nodes to automatically schedule nodes according to the dynamic changes of the load, achieve an adaptive balance between nodes, and avoid mutual communication between nodes, which is simple and expensive to achieve Smaller node scheduling is an open problem in the art.

SUMMARY OF THE INVENTION

In view of this, the purpose of the present invention is to provide a node balanced scheduling method, device, equipment, and storage medium, which can make use of the unbalanced workload of such nodes to automatically perform node scheduling according to the dynamic changes of the load. An adaptive balance is achieved between nodes, and mutual communication between nodes is avoided to achieve simple node scheduling with less overhead. Its specific plan is as follows:

In a first aspect, the present application discloses a node balanced scheduling method, applied to a server cluster, including:

Get the connection request sent by the client;

Sending the connection request to each server node, so that each server node processes the connection request according to the request sequence in its own request queue, and generates corresponding response information;

Feeding back the response information of each server node to the client, so that the client selects the server node with the shortest feedback time as the target node, and sends the data to be processed to the target node, so that The target node processes the data to be processed.

Optionally, the obtaining the connection request sent by the client includes:

Obtain a connection request sent by the client including the public Internet Protocol address bound to the preset network card in the server cluster.

Optionally, the sending the connection request to each server node includes:

The connection request is sent to the homogeneous network card of each server node through the front-side bus.

Optionally, each of the server nodes respectively processes the connection request according to the request sequence in its own request queue, and generates corresponding response information, including:

Delay processing is performed on the connection request based on the load situation of each server node and the request sequence, and corresponding response information is generated.

Optionally, after feeding back the response information of each server node to the client, so that the client selects the server node with the shortest feedback time as the target node, the method further includes:

Obtain the notification message fed back by the client after the target node is screened;

Delete the response information fed back by other server nodes except the target node among the server nodes.

Optionally, the response information of each server node is fed back to the client, so that the client selects the server node with the shortest feedback time as the target node, and sends the data to be processed to the client. After the target node, it also includes:

The target node compares the sequence number of the data to be processed with the sequence number in the response information, and if the comparison results are consistent, the data to be processed is processed.

Optionally, after the target node compares the sequence number of the data to be processed with the sequence number in the response information, the method further includes:

If the comparison results are inconsistent, a cancellation data packet is sent to the transmission control protocol stack to cancel the connection with the client.

In a second aspect, the present application discloses a node balancing scheduling device, which is applied to a server cluster, including:

The request acquisition module is used to acquire the connection request sent by the client;

a request processing module, configured to send the connection request to each server node, so that each server node processes the connection request according to the request sequence in its own request queue, and generates corresponding response information;

A node scheduling module is used to feed back the response information of each server node to the client, so that the client selects the server node with the shortest feedback time as the target node, and sends the data to be processed to the target node, so that the target node processes the data to be processed.

In a third aspect, the present application discloses an electronic device, comprising:

memory for storing computer programs;

The processor is configured to execute the computer program to implement the steps of the node balance scheduling method disclosed above.

In a fourth aspect, the present application discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the steps of the aforementioned method for balanced scheduling of nodes are implemented.

It can be seen that the present application discloses a node balanced scheduling method, which is applied to a server cluster, including: obtaining a connection request sent by a client; sending the connection request to each server node, so that each server node can queue according to its own request. The connection request is processed in the request sequence in the server node, and corresponding response information is generated; The server node acts as a target node, and sends the data to be processed to the target node, so that the target node can process the data to be processed. It can be seen that, by utilizing the unbalanced workload of each node, after receiving the connection request sent by the client, according to the load of each node, a certain time will be generated when the response information is fed back. Delay, among which, the larger the load, the longer the time delay, and the longer the time to feedback the response information, so the real-time node scheduling is automatically performed according to the dynamic change of the node load, and an adaptive balance is achieved among the nodes. And it does not need to communicate with each other between nodes, so the implementation is simple and the node scheduling with less overhead.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only It is an embodiment of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to the provided drawings without creative work.

FIG. 1 is a flowchart of a node balanced scheduling method disclosed in the present application;

2 is a schematic diagram of a process of a three-way handshake based on the TCP protocol disclosed in the application;

FIG. 3 is a schematic diagram of a node scheduling principle of a server group disclosed in the present application;

4 is a flowchart of a specific node balancing scheduling method disclosed in the present application;

FIG. 5 is a schematic structural diagram of a node balance scheduling apparatus disclosed in the present application;

FIG. 6 is a structural diagram of an electronic device disclosed in this application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

In the prior art, usually according to the content of the connection request and the type of service, the workload of each node is often unbalanced, and this unbalance is unavoidable and changes over time. When transmitting data with the TCP protocol, there are several commonly used distributed task scheduling algorithms: Round-robin method: Round-robin algorithm is the simplest and easiest to implement among all algorithms. Round-robin method simply rotates linearly in a series of nodes. , the balancer sends the new request to the next node in the node table. And so on. This algorithm is widely used in DNS domain name round robin. However, simply applying round-robin DNS translation may result in continuous access to the same node, which interferes with normal network load balancing and makes the network balancing system unable to work efficiently. The round-robin method is typically suitable for situations where the processing power and performance of all nodes in the cluster are the same. In practical applications, it is generally more effective to use it in combination with other simple methods. Weighted method: The weighted algorithm distributes the load according to the priority or weight of the nodes. Weights are assumptions or estimates based on the capabilities of each node. The weighted method intelligence is used in conjunction with other methods and is a good complement to them. Hash method: Hash method is also called Hash method. It sends network requests to cluster nodes according to certain rules through a single-shot irreversible Hash function. Similar to simple weighting. Least connection method: For TCP connections, in the least connection method, the management node records all current active connections, and sends the next new request to the node with the least number of connections. The disadvantage is that some application layer sessions consume more system resources. Although the number of connections in the cluster is balanced, the processing volume may vary greatly, and the number of connections cannot reflect the real application load. Minimum missing method: In the minimum missing method, the management node records the request status of each node for a long time, and sends the next request to the node that has processed the fewest requests in history. Unlike the least join method, the minimum missing records the number of past joins rather than the current number of joins. Fastest response method: The management node records the network response time from itself to each cluster node, and assigns the next incoming connection request to the node with the shortest response time. In most LAN based clusters, the fastest response algorithm does not work very well, most modern systems connected to Ethernet, with partial load, can respond in 1ms or less, which makes this method Pointless.

Therefore, the present application discloses a node balancing scheduling scheme, which can make use of the unbalanced workload of such nodes to automatically perform node scheduling according to the dynamic changes of the load, and achieve an adaptive balance among the nodes. And avoid mutual communication between nodes, to achieve simple node scheduling with less overhead.

Referring to FIG. 1, an embodiment of the present invention discloses a node balanced scheduling method, which is applied to a server cluster, including:

Step S11: Obtain the connection request sent by the client.

In this embodiment, the connection request sent by the client and including the public internet protocol address bound to the preset network card in the server cluster is obtained. It is understandable that in the initial stage of connection request establishment, the second handshake signal of TCP connection is delayed for a certain period according to the load status of each Servcanzher node, so that the node with the lightest load is always the first to respond to the connection request of the client. Its working principle is described as follows: At t1, the client accesses the cluster server, sends a SYN request to the public IP address, and prepares to establish a connection; as shown in Figure 2, when the client and the server prepare to communicate through the TCP protocol, first the client Both the client and the server are in the closed state. When the server actively monitors a certain port, the client generates a random initialization sequence number, and places the initialization sequence number in the corresponding field of the TCP header, and sets the SYN flag to 1. , indicating a SYN packet. The SYN message is then sent to the server to initiate a connection to the server. It should be noted that at t1, the SYN message sent by the client does not contain application layer data, but is just a connection request.

Step S12: Send the connection request to each server node, so that each server node processes the connection request according to the request sequence in its own request queue, and generates corresponding response information.

In this embodiment, after the SYN packet reaches the front-side bus, the connection request is sent to the homogeneous network card of each server node through the front-side bus. It can be understood that the front-side bus is sent to the homogeneous network card of each node in the form of broadcast , that is to say, a SYN packet is propagated on the front-side bus. Since the front-side bus connects each server node, the same SYN packet can be sent to each server node through the front-side bus, that is, a one-to-many data propagation method is realized. After the request is sent to each server node, then each server node delays the request according to its own load situation given by the load collection module. When the server is running normally, due to its own CPU capability, business data currently being processed, etc., or external factors, each server node may have inconsistent CPU resource occupancy and remaining conditions at the same time of each node. When a new service request is added, the service request is placed in the own request queue of its own server node in advance, so that each server node processes the service requests in its own request queue in turn according to the request sequence, and performs the processing for the connection request allocated by the front bus. Process and generate corresponding response information.

In this embodiment, delay processing is performed on the connection request based on the load situation of each of the server nodes and the request sequence, and corresponding response information is generated. It can be understood that, referring to Figure 2, when the server receives the SYN packet from the client, the server also randomly initializes its own serial number, and fills the serial number into the field of the TCP header, and then confirms the The response information is filled in the field of the response information in the TCP header, and then the SYN and ACK flags are set to 1. Finally, the message of the response information is sent to the client, that is, the server node sends the second handshake signal, which contains SYN and ACK packets. It should be noted that the response message does not contain application layer data. .

Step S13: Feed back the response information of each server node to the client, so that the client selects the server node with the shortest feedback time as the target node, and sends the data to be processed to the target node, so that the target node processes the data to be processed.

In this embodiment, the response information of each server node is fed back to the client, so that the client selects the server node with the shortest feedback time as the target node. It can be understood that at time t2, The response information of the server node with less load reaches the client, and the server node whose response information reaches the client at time t2 is used as the target node, and then from all the server nodes, the target node to perform this task is determined, so the client The terminal sends the third handshake signal to the public IP address, the third handshake signal reaches the target node through the front-side bus again, and sends the data to be processed in this handshake signal to the target node, so that the target node can The data to be processed is processed.

In this embodiment, as shown in FIG. 3 , the front-side bus is used as a single entry for receiving data packets and sending data packets of the server cluster. With the help of the shared medium feature of the hub, all data packets accessing the server cluster can be processed by the internal servers. The node's homogeneous network card receives. All server nodes in the server cluster are equipped with two network cards. The upper network card is bound to an external public IP address to realize a single IP image of the server cluster; the lower network card is equipped with an internal 1P address, which is responsible for interacting with the server cluster management console. Send the packet to the external gateway. The console node is responsible for managing and monitoring the working status of each service node. It can be understood that when the client sends a connection request, it is sent to the front-side bus on the public IP address, so that the front-side bus receives all the data packets and distributes them to all the corresponding nodes in the server cluster according to the internal IP address, In order for the server nodes to feed back the corresponding response information for the connection request, and return the response information to the client through the front-end bus, it should be noted that there is also a console node in the server cluster, which uses It is used to monitor and manage the working status of each server node, so as to detect the abnormal situation of the server node in time, and generate the corresponding abnormal log and report it to the upper management layer in time, so that the operation and maintenance personnel can deal with the abnormal service node according to the abnormal log.

It can be seen that the present application discloses a node balanced scheduling method, which is applied to a server cluster, including: obtaining a connection request sent by a client; sending the connection request to each server node, so that each server node can queue according to its own request. The connection request is processed in the request sequence in the server node, and corresponding response information is generated; The server node acts as a target node, and sends the data to be processed to the target node, so that the target node can process the data to be processed. It can be seen that, by utilizing the unbalanced workload of each node, after receiving the connection request sent by the client, according to the load of each node, a certain time will be generated when the response information is fed back. Delay, among which, the larger the load, the longer the time delay, and the longer the time to feedback the response information, so the real-time node scheduling is automatically performed according to the dynamic change of the node load, and an adaptive balance is achieved among the nodes. And it does not need to communicate with each other between nodes, so the implementation is simple and the node scheduling with less overhead.

Referring to FIG. 4 , an embodiment of the present invention discloses a specific node balancing scheduling method. Compared with the previous embodiment, this embodiment further describes and optimizes the technical solution. specific:

Step S21: Obtain the connection request sent by the client.

Step S22: Send the connection request to each server node, so that each server node processes the connection request according to the request sequence in its own request queue, and generates corresponding response information.

Step S23: Feed back the response information of each server node to the client, so that the client selects the server node with the shortest feedback time as the target node.

Wherein, for more detailed processing procedures in steps S21, S22, and S23, please refer to the previously disclosed embodiments, which will not be repeated here.

Step S24: Acquire the notification message fed back by the client after screening the target node; delete the response information fed back by the remaining server nodes except the target node among the server nodes.

In this embodiment, when the response information of the node with less load at time t2 reaches the client, the client sends a third handshake signal to the public IP address, and at time t3, the response information of the node with larger load reaches the client Since t3>t2, the client has received the response information of server node A at time t2, and the corresponding TCP protocol stack will automatically discard the response information of server node B at time t3. In this way, according to the load of each server node and the request time for processing the connection request, the present embodiment can simply filter out the server node that is currently suitable for processing the new task, so as to achieve an adaptive balance among the nodes, avoiding the need for Additional communication is also required between nodes, and the implementation process is simple and convenient.

Step S25: Send the data to be processed to the target node, so that the target node can process the data to be processed.

In this embodiment, the response information of each server node is fed back to the client, so that the client selects the server node with the shortest feedback time as the target node, and sends the data to be processed After reaching the target node, it also includes: comparing the serial number of the data to be processed with the serial number in the response information through the target node, and if the comparison results are consistent, comparing the data to be processed to be processed. If the comparison results are inconsistent, a cancellation data packet is sent to the transmission control protocol stack to cancel the connection with the client. It can be understood that when the third handshake signal sent by the client reaches each server node via the front-end bus, the packet filtering module checks whether the ACK sequence number of the response signal is consistent with its own initial sequence number, and if it is equal, it is allowed to pass; Otherwise, send an RST packet to the upper TCP protocol stack to cancel the connection.

It can be seen that in this embodiment, the server node with the shortest feedback response time is selected as the target node for processing the data to be processed at the current moment by using the length of time for the response information generated by each server node to the connection request sent by the server to reach the client, and Before processing the data to be processed, it is also necessary to compare whether the serial numbers are consistent, so as to avoid the situation of error processing, and realize the distributed scheduling of the real-time load of the nodes conveniently and quickly.

Referring to FIG. 5 , an embodiment of the present invention further discloses a node balancing scheduling device, which is applied to a server cluster, including:

A request obtaining module 11, for obtaining the connection request sent by the client;

The request processing module 12 is configured to send the connection request to each server node, so that each server node processes the connection request according to the request sequence in its own request queue, and generates corresponding response information;

The node scheduling module 13 is used to feed back the response information of each server node to the client, so that the client selects the server node with the shortest feedback time as the target node, and sends the data to be processed to the target node, so that the target node can process the data to be processed.

In the request processing module 12, the front-side bus is used as a single entry for the server cluster to receive data packets and send data packets. With the help of the shared medium feature of the hub, all the data packets accessing the server cluster can be homogeneous by the internal server nodes. NIC receives. All server nodes in the server cluster are equipped with two network cards. The upper network card is bound to an external public IP address to realize a single IP image of the server cluster; the lower network card is equipped with an internal 1P address, which is responsible for interacting with the server cluster management console. Send the packet to the external gateway. The console node is responsible for managing and monitoring the working status of each service node. It can be understood that when the client sends a connection request, it is sent to the front-side bus on the public IP address, so that the front-side bus receives all the data packets and distributes them to all the corresponding nodes in the server cluster according to the internal IP address, In order for the server nodes to feed back the corresponding response information for the connection request, and return the response information to the client through the front-end bus, it should be noted that there is also a console node in the server cluster, which uses It is used to monitor and manage the working status of each server node, so as to detect the abnormal situation of the server node in time, and generate the corresponding abnormal log and report it to the upper management layer in time, so that the operation and maintenance personnel can deal with the abnormal service node according to the abnormal log.

It can be seen that the present application discloses a node balanced scheduling method, which is applied to a server cluster, including: obtaining a connection request sent by a client; sending the connection request to each server node, so that each server node can queue according to its own request. The connection request is processed in the request sequence in the server node, and corresponding response information is generated; The server node acts as a target node, and sends the data to be processed to the target node, so that the target node can process the data to be processed. It can be seen that, by utilizing the unbalanced workload of each node, after receiving the connection request sent by the client, according to the load of each node, a certain time will be generated when the response information is fed back. Delay, among which, the larger the load, the longer the time delay, and the longer the time to feedback the response information, so the real-time node scheduling is automatically performed according to the dynamic change of the node load, and an adaptive balance is achieved among the nodes. And it does not need to communicate with each other between nodes, so the implementation is simple and the node scheduling with less overhead.

In some specific embodiments, the request obtaining module 11 may specifically include:

The first request obtaining unit is configured to obtain a connection request sent by a client and including a public Internet protocol address bound to a preset network card in the server cluster.

In some specific embodiments, the request processing module 12 may specifically include:

The request sending unit is used for sending the connection request to the homogeneous network card of each server node through the front side bus.

In some specific embodiments, the request processing module 12 may specifically include:

The response information acquisition unit is configured to perform delay processing on the connection request based on the load situation of each of the server nodes and the request sequence, and generate corresponding response information.

In some specific embodiments, the node scheduling module 13 may specifically include:

An information deletion unit, configured to acquire the notification message fed back by the client after screening the target node; and delete the response information fed back by the remaining server nodes except the target node among the server nodes.

In some specific embodiments, the node scheduling module 13 may specifically include:

The sequence number comparison sub-module is used to compare the sequence number of the data to be processed with the sequence number in the response information through the target node, and if the comparison results are consistent, the data to be processed is compared. deal with.

In some specific embodiments, the sequence number comparison sub-module may specifically include:

The connection revocation unit is used for sending revocation data packets to the transmission control protocol stack if the comparison results are inconsistent, and revoking the connection with the client.

Further, an embodiment of the present application also discloses an electronic device. FIG. 6 is a structural diagram of an electronic device 20 according to an exemplary embodiment, and the content in the figure should not be considered as any limitation on the scope of application of the present application.

FIG. 6 is a schematic structural diagram of an electronic device 20 according to an embodiment of the present application. The electronic device 20 may specifically include: at least one processor 21 , at least one memory 22 , a power supply 23 , a communication interface 24 , an input and output interface 25 and a communication bus 26 . Wherein, the memory 22 is used for storing a computer program, and the computer program is loaded and executed by the processor 21 to implement the relevant steps in the node balance scheduling method disclosed in any of the foregoing embodiments. In addition, the electronic device 20 in this embodiment may specifically be an electronic computer.

In this embodiment, the power supply 23 is used to provide working voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows is applicable Any communication protocol in the technical solution of the present application is not specifically limited here; the input and output interface 25 is used to obtain external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, here No specific limitation is made.

The processor 21 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and the like. The processor 21 may use at least one hardware form among DSP (Digital Signal Processing, digital signal processing), FPGA (Field-Programmable Gate Array, field programmable gate array), PLA (Programmable Logic Array, programmable logic array) accomplish. The processor 21 may also include a main processor and a coprocessor. The main processor is a processor used to process data in the wake-up state, also called a CPU (Central Processing Unit, central processing unit); A low-power processor for processing data in a standby state. In some embodiments, the processor 21 may be integrated with a GPU (Graphics Processing Unit, image processor), and the GPU is used for rendering and drawing the content that needs to be displayed on the display screen. In some embodiments, the processor 21 may further include an AI (Artificial Intelligence, artificial intelligence) processor, where the AI processor is used to process computing operations related to machine learning.

In addition, as a carrier for resource storage, the memory 22 can be a read-only memory, a random access memory, a magnetic disk or an optical disk, etc. The resources stored on it can include an operating system 221, a computer program 222, etc., and the storage method can be short-term storage or permanent storage. .

The operating system 221 is used to manage and control each hardware device and computer program 222 on the electronic device 20, so as to realize the operation and processing of the massive data 223 in the memory 22 by the processor 21, which can be Windows Server, Netware, Unix, Linux etc. The computer program 222 may further include a computer program that can be used to complete other specific tasks in addition to the computer program that can be used to complete the node balance scheduling method executed by the electronic device 20 disclosed in any of the foregoing embodiments. The data 223 may not only include the data received by the electronic device and transmitted from the external device, but also may include the data collected by the own input/output interface 25 and the like.

Further, the present application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the aforementioned method for balanced scheduling of nodes is implemented. For the specific steps of the method, reference may be made to the corresponding content disclosed in the foregoing embodiments, which will not be repeated here.

The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments may be referred to each other. As for the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant part can be referred to the description of the method.

Professionals may further realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two, in order to clearly illustrate the possibilities of hardware and software. Interchangeability, the above description has generally described the components and steps of each example in terms of functionality. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of this application. The steps of a method or algorithm described in conjunction with the embodiments disclosed herein may be directly implemented in hardware, a software module executed by a processor, or a combination of the two. A software module can be placed in random access memory (RAM), internal memory, read only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other in the technical field. in any other known form of storage medium.

Finally, it should also be noted that in this document, relational terms such as first and second are used only to distinguish one entity or operation from another, and do not necessarily require or imply these entities or that there is any such actual relationship or sequence between operations. Moreover, the terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device that includes a list of elements includes not only those elements, but also includes not explicitly listed or other elements inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

A node balance scheduling method, device, device, and storage medium provided by the present invention are described above in detail, and specific examples are used in this paper to illustrate the principles and implementations of the present invention. The descriptions of the above embodiments are only used for Help to understand the method of the present invention and its core idea; at the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation and application scope. In summary, the content of this specification It should not be construed as a limitation of the present invention.

Claims (10)

1. A node balanced scheduling method, characterized in that, applied to a server cluster, comprising: Get the connection request sent by the client; Sending the connection request to each server node, so that each server node processes the connection request according to the request sequence in its own request queue, and generates corresponding response information; Feeding back the response information of each server node to the client, so that the client selects the server node with the shortest feedback time as the target node, and sends the data to be processed to the target node, so that The target node processes the data to be processed. 2. The node balance scheduling method according to claim 1, wherein the acquiring the connection request sent by the client comprises: Obtain a connection request sent by the client including the public Internet Protocol address bound to the preset network card in the server cluster. 3. The node balanced scheduling method according to claim 1, wherein the sending the connection request to each server node comprises: The connection request is sent to the homogeneous network card of each server node through the front-side bus. 4 . The node balanced scheduling method according to claim 1 , wherein each of the server nodes processes the connection requests according to the request sequence in their own request queues, and generates corresponding response information, including: 5 . : Delay processing is performed on the connection request based on the load situation of each server node and the request sequence, and corresponding response information is generated. 5 . The node balance scheduling method according to claim 1 , wherein the response information of each server node is fed back to the client, so that the client can filter out all the nodes with the shortest feedback time. 6 . After the above server node is used as the target node, it also includes: Obtain the notification message fed back by the client after the target node is screened; Delete the response information fed back by other server nodes except the target node among the server nodes. 6. The node balanced scheduling method according to any one of claims 1 to 5, wherein the response information of each server node is fed back to the client, so that the client can filter out The server node with the shortest feedback time is used as the target node, and after sending the data to be processed to the target node, the method further includes: The target node compares the sequence number of the data to be processed with the sequence number in the response information, and if the comparison results are consistent, the data to be processed is processed. 7 . The node balance scheduling method according to claim 6 , wherein after the target node compares the sequence number of the data to be processed with the sequence number in the response information, 8 . ,Also includes: If the comparison results are inconsistent, a cancellation data packet is sent to the transmission control protocol stack to cancel the connection with the client. 8. A node balance scheduling device, characterized in that, applied to a server cluster, comprising: The request acquisition module is used to acquire the connection request sent by the client; a request processing module, configured to send the connection request to each server node, so that each server node processes the connection request according to the request sequence in its own request queue, and generates corresponding response information; A node scheduling module is used to feed back the response information of each server node to the client, so that the client selects the server node with the shortest feedback time as the target node, and sends the data to be processed to the target node, so that the target node processes the data to be processed. 9. An electronic device, characterized in that, comprising: memory for storing computer programs; The processor is configured to execute the computer program to implement the steps of the node balanced scheduling method according to any one of claims 1 to 7. 10. A computer-readable storage medium, characterized in that it is used for storing a computer program; wherein, when the computer program is executed by a processor, the steps of the node balanced scheduling method according to any one of claims 1 to 7 are implemented .

Images