CN114296914A

CN114296914A - Node distribution method, system, device and electronic equipment

Info

Publication number: CN114296914A
Application number: CN202111582731.8A
Authority: CN
Inventors: 赵浩霖
Original assignee: Hangzhou Ezviz Software Co Ltd
Current assignee: Hangzhou Ezviz Software Co Ltd
Priority date: 2021-12-22
Filing date: 2021-12-22
Publication date: 2022-04-08

Abstract

The application provides a node distribution method, a system, a device and electronic equipment. In this embodiment, according to the feature correlation between the gene feature data of the user side and the gene feature data of each computing node registered to the device (also referred to as a double-end matching method, a double end refers to a user side and an edge end composed of all computing nodes), candidate computing nodes for executing computing tasks are determined for the user side; and then, further selecting the current online target computing node from all the candidate computing nodes and distributing the current online target computing node to the user side, so that compared with the conventional mode of randomly distributing the computing nodes to the user side, the method realizes that the computing nodes are distributed to the user side to execute the computing task of the user side at least according to the gene data of the user side and the gene data of each computing node at the edge end.

Description

Node distribution method, system, device and electronic equipment

Technical Field

The application relates to the technology of internet of things, in particular to a node distribution method, a node distribution system, a node distribution device and electronic equipment.

Background

In the internet of things, in the face of an increasingly rich intelligent computing requirement (also referred to as a computing task), at present, it is most common that a cloud control center randomly allocates an edge computing node (computing node for short) to solve the computing requirement by using a random allocation mode.

Here, a compute node, which may be deployed on the side near the origin of an object or data, is an edge compute hub or edge compute device with task computing capabilities. In a specific implementation, a device or a cluster that generally meets the design scope requirements and has edge computing capability may be referred to as an edge computing node (computing node for short).

However, at present, the cloud control center randomly allocates an edge computing node (computing node for short) in a random allocation mode to solve the problem that the allocated computing node cannot complete the computing task at the user side.

Disclosure of Invention

The embodiment of the application provides a node distribution method, a system, a device and electronic equipment, which are used for distributing computing nodes for a user side to execute computing tasks of the user side at least according to gene data of the user side and gene data of each computing node at an edge side.

The embodiment of the application provides a first node allocation method, which is applied to node management equipment and comprises the following steps:

receiving a computing task of a user side;

determining candidate computing nodes for executing the computing task according to the obtained gene characteristic data of the user side and the obtained gene characteristic data of each computing node registered to the equipment; wherein, the gene characteristic data of the user terminal is used for representing the user terminal; the gene characteristic data of the computing nodes are used for representing the computing nodes;

selecting a target computing node which is online currently and used for executing the computing task from all candidate computing nodes;

and distributing the target computing node to the user side so that the user side sends a computing task to the target computing node to be executed.

The embodiment of the present application further provides a second node allocation method, where the method is applied to a node management device, and includes:

receiving a computing task of a user side;

determining candidate computing nodes for executing the computing task according to the obtained gene characteristic data of the user side and the obtained gene characteristic data of each computing node registered to the equipment; determining candidate storage nodes for storing the calculation data generated by the calculation task according to the obtained gene characteristic data of the user side and the obtained gene characteristic data of each storage node; the gene characteristic data of the user side is used for representing the user side; the gene characteristic data of the computing nodes are used for representing the computing nodes, and the gene characteristic data of the storage nodes are used for representing the storage nodes;

selecting a target computing node which is online currently and used for executing the computing task from all candidate computing nodes, and selecting a target storage node which is used for assisting the target computing node in executing the computing task from all candidate storage nodes;

and distributing the target computing node and the target storage node to the user side so that the user side sends a computing task to the target computing node to execute and obtains a computing result by interacting with the target computing node and the target storage node.

The embodiment of the application provides a node distribution system, which comprises node management equipment, a computing node and a user side;

the user side is used for sending a computing task to the node management equipment when a computing requirement exists, and reporting gene characteristic data of the user side to the node management equipment; the gene characteristic data of the user side is used for representing the user side;

the computing node is used for registering to the node management equipment and reporting the gene characteristic data of the computing node to the node management equipment, wherein the gene characteristic data of the computing node is used for representing the computing node; when the computing task is used as a target computing node of the user side, the computing task is operated to obtain a computing result when the computing task of the user side is received;

the node management apparatus performs according to the first method as described above.

A node distribution system, the system comprising: the system comprises node management equipment, computing nodes, storage nodes and a user side;

the computing node is used for registering to the node management equipment and reporting the gene characteristic data of the computing node to the node management equipment, wherein the gene characteristic data of the computing node is used for representing the computing node; when the target computing node is used as the target computing node of the user side, the computing result is obtained by interacting with the target storage node when the computing task of the user side is received;

the storage node is used for reporting the gene characteristic data of the storage node to the node management equipment, and the gene characteristic data of the storage node is used for representing the storage node; when the target storage node is used as the target storage node of the user side, the target computing node is assisted to complete the computing task received by the target computing node;

the node management apparatus performs according to the second method as described above.

An embodiment of the present application provides a node allocation apparatus, where the apparatus is applied to a node management device, and the apparatus includes:

the first receiving unit is used for receiving a computing task of a user side;

a first determining unit, configured to determine candidate computing nodes for executing the computing task according to the obtained genetic feature data of the user side and the obtained genetic feature data of each computing node registered to the device; wherein, the gene characteristic data of the user terminal is used for representing the user terminal; the gene characteristic data of the computing nodes are used for representing the computing nodes;

the first target unit is used for selecting a target computing node which is online currently and used for executing the computing task from all candidate computing nodes;

the first processing unit is used for distributing the target computing node to the user side so that the user side can send a computing task to the target computing node to execute the computing task.

the second receiving unit is used for receiving the computing task of the user side;

a second determining unit, configured to determine candidate computing nodes for executing the computing task according to the obtained genetic feature data of the user side and the obtained genetic feature data of each computing node registered to the device; determining candidate storage nodes for storing the calculation data generated by the calculation task according to the obtained gene characteristic data of the user side and the obtained gene characteristic data of each storage node; the gene characteristic data of the user side is used for representing the user side; the gene characteristic data of the computing nodes are used for representing the computing nodes, and the gene characteristic data of the storage nodes are used for representing the storage nodes;

the second target unit is used for selecting a target computing node which is online currently and is used for executing the computing task from all the candidate computing nodes and selecting a target storage node which is used for assisting the target computing node in executing the computing task from all the candidate storage nodes;

the second processing unit is used for distributing the target computing node and the target storage node to the user side so that the user side sends a computing task to the target computing node to execute the computing task and obtains a computing result through interaction with the target computing node and the target storage node.

The embodiment of the application also provides the electronic equipment. The electronic device includes: a processor and a machine-readable storage medium;

the machine-readable storage medium stores machine-executable instructions executable by the processor;

the processor is configured to execute machine-executable instructions to implement the steps of the above-disclosed method.

As can be seen from the above technical solutions, in this embodiment, according to the feature correlation between the gene feature data of the user side and the gene feature data of each computing node registered to the device (also referred to as a double-end matching manner, a double end refers to the user side and an edge end composed of all computing nodes), candidate computing nodes for executing computing tasks are determined for the user side first; then, further selecting a current online target computing node from all candidate computing nodes and distributing the current online target computing node to the user side, compared with the existing mode of randomly distributing the computing nodes to the user side, the method realizes that the computing nodes are distributed to the user side to execute the computing task of the user side at least according to the gene data of the user side and the gene data of each computing node at the edge end;

further, in this embodiment, when the computing nodes are allocated to the user side, the feature correlation between the gene feature data of the user side and the gene feature data of each computing node registered to the device is used in the above-mentioned dual-end matching manner, and the feature correlation can ensure that the finally allocated target computing node can efficiently complete the computing task of the user side.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure.

FIG. 1 is a flow chart of a method provided by an embodiment of the present application;

fig. 2 is a flowchart of registration provided in an embodiment of the present application;

fig. 3 is a schematic diagram of reporting gene characteristic data by a computing node according to an embodiment of the present application;

FIG. 4 is a flow chart of an online detection provided by an embodiment of the present application;

fig. 5 is a flow chart of logout provided by an embodiment of the present application;

FIG. 6 is a flowchart of an implementation of step 102 provided by an embodiment of the present application;

FIG. 7 is a flow chart of another method provided by an embodiment of the present application;

FIG. 8 is a block diagram of a system provided in an embodiment of the present application;

FIG. 9 is a block diagram of another system according to an embodiment of the present application

FIG. 10 is a block diagram of an apparatus according to an embodiment of the present disclosure;

FIG. 11 is a block diagram of another apparatus according to an embodiment of the present disclosure;

fig. 12 is a block diagram of an electronic device according to an embodiment of the present application.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

In order to make the technical solutions provided in the embodiments of the present application better understood and make the above objects, features and advantages of the embodiments of the present application more comprehensible, the technical solutions in the embodiments of the present application are described in further detail below with reference to the accompanying drawings.

Referring to fig. 1, fig. 1 is a flowchart of a method provided in an embodiment of the present application. The method is applied to the node management equipment. Optionally, as an embodiment, the node management device may be a cloud control center. In this embodiment, the cloud control center may be a cloud service platform of the internet or the internet of things.

As shown in fig. 1, the process may include the following steps:

step 101, receiving a computing task of a user side.

Optionally, in this embodiment, the user terminal refers to a device or an application having real computing requirements. The user side can provide direct user-oriented services to realize user-service interaction.

When the user side has a computing requirement, the user side sends a computing task to the node management equipment. Based on this, the node management device will eventually receive the computing task from the user side.

And 102, determining candidate computing nodes for executing computing tasks according to the acquired gene characteristic data of the user side and the acquired gene characteristic data of each computing node registered to the equipment.

In this embodiment, the genetic characteristic data of the user terminal is used to characterize the user terminal, and when implemented, the genetic characteristic data may include, for example: (1) device hardware data such as hardware feature codes, etc.; (2) the local area special data such as the local country, province (or region), main city, etc.; (3) network characteristic data such as network identity (typically characterized by public network IP address and Port), network service operator, network upstream and downstream bandwidth, etc.

At the beginning, after the user end is on line, the gene characteristic data of the user end is reported if the gene characteristic data of the user end is not reported at present. If the user side finds that the gene characteristic data of the user side is reported before after the user side is online, if at least one current gene characteristic data changes compared with the corresponding gene characteristic data reported before, the changed gene characteristic data can be reported to the node management equipment currently. And then, once at least one gene characteristic data is found to be changed, the changed gene characteristic data can be reported to the node management equipment in time. Therefore, the node management equipment can always obtain the real-time latest gene characteristic data of the user side.

Similarly, in this embodiment, the genetic characteristic data of the computing node is used to characterize the computing node, which in practical implementation may include, for example: (1) device hardware data such as hardware feature codes, etc.; (2) the local area special data such as the local country, province (or region), main city, etc.; (3) network characteristic data such as network identification (generally represented by a public network IP address and a monitoring Port, where the monitoring Port is an interface for connecting the user terminal), a network service operator, network uplink and downlink bandwidth, and the like; (4) calculating load data, including the number of tasks which are simultaneously calculated in parallel, the number of tasks which are currently calculated, the surplus of the number of tasks which can be calculated and the like; (5) other temporarily specified feature data that may affect computing power, etc. The present embodiment is not particularly limited.

Initially, after the computing node is on line, if the computing node finds that the gene characteristic data of the computing node is not reported yet, the gene characteristic data of the computing node is reported. If the computing node finds that the gene characteristic data of the computing node is reported before after the computing node is on line, if at least one current gene characteristic data changes compared with the corresponding gene characteristic data reported before, the changed gene characteristic data is reported to the node management equipment currently. And then, once at least one gene characteristic data is found to be changed, the changed gene characteristic data can be reported to the node management equipment in time. So as to ensure that the node management equipment can always obtain the real-time latest gene characteristic data of the computing node.

Based on the above description, in the present embodiment, in the step 102, the candidate computing nodes are determined according to the obtained real-time latest genetic feature data of the user end and the obtained real-time latest genetic feature data of each computing node registered to the device. As to how to determine candidate computing nodes for performing the computing task according to the obtained real-time latest gene feature data of the user side and the obtained real-time latest gene feature data of each computing node registered to the device, the following description will be given by way of example, and details are not repeated here.

And 103, selecting a target computing node which is online currently and is used for executing the computing task from all the candidate computing nodes.

Alternatively, as an embodiment, the step 103 may select a target computing node that is currently online and used for executing the computing task from all candidate computing nodes depending on the filtering condition. Such as the filtering conditions include at least: the physical distance between the ue and the ue is closest, and/or the load is lower than the set load, and the present embodiment is not limited in particular.

Taking the example that the filtering condition includes the physical distance between the filtering condition and the user side being the closest, the finally selected target computing node may be the computing node which is currently online and has the closest physical distance between the filtering condition and the user side. The target computing node is further described below, and will not be described in detail here.

And 104, distributing the target computing node to the user side so that the user side sends the computing task to the target computing node to be executed.

Optionally, in this embodiment, the target computing node is allocated to the user side, and then the user side can establish a connection with the target computing node and complete mutual authentication. And then, the user side can send the calculation task to the target calculation node through the connection between the user side and the target calculation node, so that the target calculation node performs corresponding intelligent calculation according to the calculation task to obtain a calculation result. And then, the target computing node returns the computing result to the user side, and further, the target computing node returns the computing task and the computing result to the node management equipment, so that the node management equipment records the association relationship among the computing task, the user side and the computing result. Here, the node management device records the association relationship, so as to subsequently further allocate candidate computing nodes to the user side, which will be described below and will not be described herein for the sake of brevity.

Thus, the flow shown in fig. 1 is completed.

As can be seen from fig. 1, in this embodiment, according to the characteristic correlation between the gene feature data of the user end and the gene feature data of each computing node registered to the device (also referred to as a double-end matching method, a double end refers to the user end and an edge end composed of all computing nodes), candidate computing nodes for executing computing tasks are determined for the user end first; then, further selecting a current online target computing node from all candidate computing nodes and distributing the current online target computing node to the user side, compared with the existing mode of randomly distributing the computing nodes to the user side, the method realizes that the computing nodes are distributed to the user side to execute the computing task of the user side at least according to the gene data of the user side and the gene data of each computing node at the edge end;

In the above step 102, each computing node registered to the node management device is mentioned, and how the computing node is registered to the node management device is described below:

referring to fig. 2, fig. 2 is a registration flowchart provided in an embodiment of the present application. As shown in fig. 2, the process may include the following steps:

in step 201, the node management device receives a registration parameter sent by the computing node during registration.

Optionally, in this embodiment, for any computing node, it needs to complete registration through the HTTP interface of the node management device. When the computing node registers, it sends registration parameters to the HTTP interface.

As an embodiment, the registration parameters herein include at least one parameter shown in table 1.

TABLE 1

Step 202, the node management device performs node registration logic processing on the registration parameters, and returns a registration result to the computing node sending the registration parameters.

Optionally, in this embodiment, after receiving the registration parameter, the node management device performs node registration logic processing, such as authentication, on the registration parameter, and finally obtains a registration result.

Alternatively, in this embodiment, the node management apparatus may return the registration result in the form of an HTTP Code. The registration result may include one of the result identifiers shown in table 2:

TABLE 2

In table 2, there may be many identifiers that cause the fourth result, such as a server-side error in the node management device, a network error between the node management device and the computing node, and the like. Once the fourth result identifier appears, the computing node sending the registration parameter needs to retry for multiple times, the suggested time frequency of each retry is sequentially increased, and a fixed time frequency is not used for performing the registration retry, so that the problem that the node management device is attacked by DDos attacks due to the registration retry performed by the batch computing node is mainly solved.

Thus, the flow shown in fig. 2 is completed.

How the computing node is registered to the node management device is realized through the flow shown in fig. 2, so that the node management device subsequently allocates the computing node registered to the device to the user side.

As described above with respect to reporting gene signature data by a computing node, the following is illustrated by way of example in fig. 3:

referring to fig. 3, fig. 3 is a schematic diagram of reporting gene feature data by a computing node according to an embodiment of the present disclosure. As shown in fig. 3, the process may include the following steps:

step 301, the node management device receives gene characteristic data reported by the computing node.

As described above, if the computing node finds that at least one piece of gene feature data changes after going online, or the gene feature data is not reported yet, the computing node sends the gene feature data to the HTTP interface of the node management device to complete the reporting of the gene feature data.

As described above, the computing node may report the gene feature data as follows:

the node identification ID of the computing node and the protocol version number of the computing node are reported as shown in table 3, and then the specific gene characteristic data of the computing node is reported as shown in table 4. Here, the specific gene feature data refers to gene feature data other than the node identification ID, the protocol version number at which the computing node operates. In this embodiment, the specific gene feature data is a combination of reported data pairs (each data pair includes a signaling ID and a data packet data) in a key-value (KV) pair manner, which is specifically shown in table 4.

TABLE 3

TABLE 4

And step 302, the node management equipment performs logic processing on the gene characteristic data and returns a report result to the computing node sending the gene characteristic data.

Optionally, in this embodiment, after receiving the gene feature data reported by the computing node, the node management device performs logic processing on the gene feature data, such as calling a function to store the gene feature data, and finally obtains a reporting result.

Optionally, in this embodiment, the node management device may return the report result in the form of an HTTP Code. The reported result may include one of the result identifiers shown in table 5:

TABLE 5

In table 5, there may be many identifiers, such as a server-side error in the node management device, a network error between the node management device and the computing node, and the like. Once the fifth result identifier appears, the computing node sending the gene feature data needs to retry for multiple times, the suggested time frequency of each retry is sequentially increased, and reporting retry is not needed to be performed by using a fixed time frequency, so that the DDos attack to the node management device caused by the retry of reporting the gene feature data by the batch computing nodes is mainly prevented.

The flow shown in fig. 3 is completed.

How the computing node reports the gene characteristic data to the node management device is realized through the process shown in fig. 3, so that the node management device subsequently allocates the computing nodes registered to the device to the user side.

It should be noted that, in step 103, the finally selected target computing node must be online. As to how to detect that the target computing node is online, in this embodiment, each computing node calls an HTTP heartbeat interface of the node management device to complete heartbeat keep-alive, so that the node management device detects a network state of each computing node, such as online or offline. Fig. 4 illustrates an example of how the node management device detects the network status of each computing node, such as online or offline:

referring to fig. 4, fig. 4 is a flowchart of online detection provided in the embodiment of the present application. As shown in fig. 4, the process may include the following steps:

step 401, a node management device receives a heartbeat keep-alive message sent by a computing node.

Optionally, in this embodiment, the heartbeat keep-alive message at least carries: a computing node Identification (ID), a protocol version number the computing node runs. The computing node Identification (ID), the protocol version number the computing node runs on may be as shown in table 3.

Step 402, the node management device performs heartbeat logic processing on the heartbeat keep-alive message to obtain a response result, and returns the response result to the calculation result.

Optionally, the response result includes: the heartbeat keep-alive message is successfully sent, or the heartbeat keep-alive message needs to be sent repeatedly at least once again, or the node management equipment is abnormal, or the version of the computing node needs to be upgraded, or the computing node is logged off, or the computing node is pulled into a blacklist. And when the response result comprises that the heartbeat keep-alive message is successfully sent, determining that the computing node sending the heartbeat keep-alive message is on line. Table 6 illustrates the response results, which include how one result identifies:

TABLE 6

In table 6, there may be many errors, such as a server-side error in the node management device, a network error between the node management device and the computing node, and the like. Once the eleventh result identifier appears, the computing node sending the heartbeat keep-alive message needs to retry for multiple times, the suggested time frequency of each retry is sequentially increased, and a fixed time frequency is not used for reporting and retrying, so that the DDos attack to the node management device caused by the fact that the batch computing node retries to send the heartbeat keep-alive message is mainly prevented.

The flow shown in fig. 4 is completed.

How the node management device detects the network status of each computing node, such as online or offline, is realized through the flow shown in fig. 4, so that when a computing node registered to the node management device is allocated to a user side, the computing node is ensured to be online first.

It should be noted that, in this embodiment, the computing node may further invoke an HTTP interface of the node management device based on the logout event, so as to complete the logout operation. Here, there are many logout events, such as the resources of the computing node are full, and the embodiment is not limited in particular.

Referring to fig. 5, fig. 5 is a flowchart of logout provided in the embodiment of the present application. As shown in fig. 5, the process may include the following steps:

step 501, a node management device receives a logout parameter sent by a computing node when logging out.

The logging-off parameter here may include at least a computing node identification ID, a protocol version number run by the computing node as shown in table 3.

Step 502, the node management device performs node logout logic processing on the logout parameters, and returns a processing result to the computing node.

In this embodiment, the processing result may include: the deregistration is successful or at least one more re-deregistration is attempted. Table 7 illustrates the processing results:

TABLE 7

The flow shown in fig. 5 is completed.

How the node management device manages the logout of each computing node is realized by the flow shown in fig. 5.

How to determine candidate computing nodes for executing the computing task according to the obtained genetic characteristic data of the user side and the obtained genetic characteristic data of each computing node registered to the device in the above step 102 is as follows:

referring to fig. 6, fig. 6 is a flowchart of step 102 implementation provided in an embodiment of the present application. As shown in fig. 6, the process may include the following steps:

step 601, identifying whether the computing task is the first computing task sent by the user side, if so, executing step 602, and if not, executing step 603.

Optionally, in this embodiment, there are many implementations of identifying whether the computing task is the first computing task sent by the user side, for example, when receiving each calculation task sent by the user end, if the calculation task is received before, the recorded time (recorded as the calculation task receiving time) of the calculation task received by the user end before is updated to the current time, otherwise, the current time is taken as the calculation task receiving time of the user end, and based on the time, in this step 601, it can be identified whether the computing task is the first computing task sent by the user side based on whether the computing task receiving time of the user side currently exists, that is, if there is no computation task receiving time of the user side, it is determined whether the computation task is the first computation task sent by the user side, otherwise, it is determined that the computation task is not the first computation task sent by the user side.

For another example, as described below, in this embodiment, a corresponding computing node allocation set is generated for the user side in advance (specifically, as described in step 603). Based on this, in this step 601, whether the computing task is the first computing task sent by the user side may be identified based on whether the computing node allocation set corresponding to the user side exists currently, that is, if the computing node allocation set of the user side does not exist, it is determined whether the computing task is the first computing task sent by the user side, otherwise, it is determined that the computing task is not the first computing task sent by the user side.

Step 602, determining candidate computing nodes for executing computing tasks from the computing nodes registered to the device according to the obtained gene feature data of the user side and the obtained gene feature data of the computing nodes registered to the device.

This step 602 is executed on the premise that the computing task is the first computing task sent by the user side.

Alternatively, in this embodiment, a feature correlation (which may have a confidence expression) between the gene feature data of the user terminal (the latest gene feature data by the current user terminal) and the gene feature data of each computing node registered to the apparatus (the latest gene feature data by the current computing node) may be calculated by using a trained gene affinity algorithm, and the computing node whose confidence of the feature correlation with the user terminal satisfies a condition (for example, is greater than or equal to a preset confidence) is taken as the candidate computing node.

Step 603, determining all the computing nodes in the computing node distribution set generated for the user side as candidate computing nodes for executing computing tasks; the computing node distribution set is determined according to the obtained gene characteristic data of the user side, the obtained gene characteristic data of each computing node registered to the device, and historical target computing nodes which are distributed by the user side before and used for executing historical computing tasks.

This step 603 is performed on the premise that the computing task is not the first computing task sent by the user side.

In this embodiment, when detecting that any of the genetic characteristic data of the user side changes and/or any of the genetic characteristic data of the computing nodes changes, the computing node allocation set of the user side is adjusted in time.

In this embodiment, the computing node allocation set is determined according to the genetic feature data of the user end, the genetic feature data of each computing node, and the history target computing node that has been previously allocated by the user end for performing the history computing task. In a specific implementation, a feature correlation (which may have a confidence expression) between gene feature data of a user terminal (latest gene feature data by a current user terminal) and gene feature data of each computing node registered to the device (latest gene feature data by a current computing node) may be calculated by using a trained gene affinity algorithm, and a computing node allocation set may be determined based on a computing node whose confidence in the feature correlation with the user terminal satisfies a condition (e.g., is greater than or equal to a preset confidence) and a historical target computing node to which the user terminal has been previously allocated for performing a historical computing task.

Based on the distribution set of computing nodes, in this step 602, all computing nodes in the distribution set of computing nodes may be directly determined as candidate computing nodes for executing the computing task.

As for the gene affinity algorithm, the embodiment is not particularly limited as long as it can ensure the characteristic correlation between the gene feature data of the determined user end (the latest gene feature data of the current user end), and the gene feature data of each computing node registered in the device (the latest gene feature data of the current computing node).

The flow shown in fig. 6 is completed.

How to determine candidate computing nodes for executing the computing task according to the obtained genetic characteristic data of the user side and the obtained genetic characteristic data of each computing node registered to the device in step 102 is realized through the flow shown in fig. 6.

It should be noted that, in this embodiment, after the target computing node is allocated to the user end in step 104, in this case, the target computing node may also be referred to as a history target computing node to which the user end has been allocated for performing the history computing task. Based on this, the embodiment may further update the computing node allocation set according to the genetic feature data of the user end, the genetic feature data of each computing node, and the historical target computing node that has been previously allocated by the user end for performing the historical computing task.

In addition, as described above, each candidate computing node has a corresponding confidence with the feature relevance of the user side; the confidence level is used to indicate a likelihood that the candidate compute node is assigned to the user side. Based on this, in the present embodiment, the target computing node is a computing node that satisfies the filtering condition and is selected from all candidate computing nodes, and the confidence of the feature correlation between the target computing node and the user side is greater than the confidence of the feature correlation between other candidate computing nodes that satisfy the filtering condition and the user side.

The above is that the station distributes the computing nodes for the user terminal by matching the computing nodes of the double end, i.e. the user terminal and the edge terminal. The following stations distribute computing nodes for the user side in a way that three ends, namely the computing nodes of the user side and the edge side are matched with the storage nodes of the edge side:

here, the storage node may generally provide that the computing node may need to use some intermediate data or resources in order to complete the computing task, i.e., the storage node acts as a data cache or a node providing important resources.

Referring to fig. 7, fig. 7 is a flowchart of another method provided by the embodiments of the present application. As shown in fig. 7, the process may include the following steps:

step 701, receiving a computing task of a user side;

step 702, determining candidate computing nodes for executing the computing task according to the obtained gene characteristic data of the user side and the obtained gene characteristic data of each computing node registered to the device; and determining candidate storage nodes for storing the calculation data generated by the calculation task according to the acquired gene characteristic data of the user side and the acquired gene characteristic data of each storage node.

In this step 702, the gene feature data of the storage node is used to characterize the storage node, which may include, for example: (1) hardware data of equipment such as hardware feature codes, etc., (2) regional specific data such as country, province (or region), main city, etc., (3) network characteristic data such as network identification (generally characterized by public network IP address and Port), network service operator, network uplink and downlink bandwidth, etc., (4) storage resources. The present embodiment is not particularly limited.

How to determine candidate storage nodes for storing the computing data generated by the computing task according to the obtained genetic characteristic data of the user side and the obtained genetic characteristic data of each storage node, which is similar to the above-mentioned manner for determining candidate computing nodes, for example, if the computing task is the first computing task sent by the user side, determining candidate storage nodes for storing the computing data generated by the computing task according to the obtained genetic characteristic data of the user side and the obtained genetic characteristic data of each storage node; if the computing task is not the first computing task sent by the user side, determining all computing nodes in a storage node distribution set generated for the user side as candidate storage nodes for storing computing data generated by the computing task; the storage node distribution set is determined according to the obtained gene characteristic data of the user side, the obtained gene characteristic data of each storage node and the history storage nodes distributed by the user side before. This embodiment is not described in detail.

Step 703, selecting a target computing node which is currently online and used for executing the computing task from all the candidate computing nodes, and selecting a target storage node used for assisting the target computing node in executing the computing task from all the candidate storage nodes.

In this embodiment, there are many ways to select a target storage node from all candidate storage nodes for assisting the target computing node in executing the computing task, but it is necessary to ensure that the finally selected target storage node must meet the requirement of assisting the target computing node in executing the computing task.

Step 704, distributing the target computing node and the target storage node to the user side, so that the user side sends a computing task to the target computing node to execute and obtains a computing result by interacting with the target computing node and the target storage node.

The flow shown in fig. 7 is completed.

As can be seen from the flow shown in fig. 7, in this embodiment, according to the gene feature data of the user end, the gene feature data of each computing node that has been registered to the device, and the feature correlation between the gene feature data of each storage node (also referred to as a three-terminal matching method, where three terminals refer to the user end, the edge computing node, and the edge storage node), candidate computing nodes and candidate storage nodes are determined for the user end first; then further selecting a target computing node from all candidate computing nodes and selecting a target storage node from all candidate storage nodes and distributing the target storage node to the user side, compared with the existing mode of randomly distributing computing nodes to the user side, the method realizes that computing nodes are distributed to the user side to execute computing tasks of the user side at least according to gene data of the user side and gene data of each computing node at the edge side;

further, in this embodiment, when the computing nodes are allocated to the user side, the three-end matching method is used to obtain the feature correlation between the gene feature data of the user side, the gene feature data of each computing node registered in the device, and the gene feature data of each storage node, and the feature correlation can ensure that the finally allocated target computing node can efficiently complete the computing task of the user side.

The method provided by the embodiment of the application is described above. The following describes a system and an apparatus provided in an embodiment of the present application:

referring to fig. 8, fig. 8 is a system structure diagram provided in the embodiment of the present application. As shown in fig. 8, the system includes a node management device, a computing node, and a user side.

In this embodiment, the user side is configured to send a computation task to the node management device when there is a computation demand, and report gene feature data of the user side to the node management device; the gene characteristic data of the user side is used for representing the user side;

the computing node is used for registering to the node management equipment and reporting the gene characteristic data of the computing node to the node management equipment, wherein the gene characteristic data of the computing node is used for representing the computing node; and when the target computing node is used as the target computing node of the user side, running the computing task to obtain a computing result when the computing task of the user side is received.

The node management device is configured to execute the process shown in fig. 1.

The embodiment of the application also provides another system structure diagram. Referring to fig. 9, fig. 9 is a block diagram of another system provided in the embodiment of the present application. As shown in fig. 9, the system may include: the system comprises node management equipment, computing nodes, storage nodes and a user side.

the node management apparatus is configured to execute the process shown in fig. 7.

The system provided by the embodiment of the application is described above. The following describes the apparatus provided in the embodiments of the present application:

referring to fig. 10, fig. 10 is a diagram illustrating a structure of an apparatus according to an embodiment of the present disclosure. The device is applied to node management equipment and comprises:

the first receiving unit is used for receiving a computing task of a user side;

Optionally, the first receiving unit further receives a registration parameter sent by the computing node at the time of registration; the registration parameter is used for representing a parameter required to be subjected to registration authentication when the computing node needs to be registered to the node management equipment; the first processing unit further performs node registration logic processing on the registration parameters and returns registration results to the computing nodes sending the registration parameters;

wherein the registration result comprises: a first result identification, a second result identification, a third result identification, or a fourth result identification;

the first result identification is used for indicating that the computing node is successfully registered to the node management equipment;

the second result identification is used for indicating that the computing node is pulled into a blacklist;

the third result identification is used for indicating that the version of the computing node needs to be upgraded;

the fourth result identifier indicates that the computing node is to re-attempt at least one more re-registration.

Optionally, the registration parameters include at least:

the method comprises the following steps that a node identification ID of a computing node, a computing type supported by the computing node, a protocol version number operated by the computing node, whether the computing node supports Unicom detection or not and a public network IP address of the computing node are obtained;

the connectivity detection is used to test the quality of network connectivity between the node management device and the compute node.

Optionally, the determining, by the first determining unit, the candidate computing nodes for performing the computing task according to the obtained genetic feature data of the user side and the obtained genetic feature data of each computing node registered to the device includes:

if the computing task is the first computing task sent by the user side, determining candidate computing nodes for executing the computing task from the computing nodes registered to the equipment according to the obtained gene characteristic data of the user side and the obtained gene characteristic data of the computing nodes registered to the equipment;

if the computing task is not the first computing task sent by the user side, determining all computing nodes in a computing node distribution set generated for the user side as candidate computing nodes for executing the computing task; the computing node distribution set is determined according to the obtained gene characteristic data of the user side, the obtained gene characteristic data of each computing node registered to the device, and historical target computing nodes which are distributed by the user side before and used for executing historical computing tasks.

Optionally, each candidate computing node has a corresponding confidence with the feature correlation of the user side; the confidence level is used for indicating the possibility that the candidate computing node is allocated to the user side;

the target computing node is a computing node which is selected from all candidate computing nodes and meets the filtering condition, and the confidence degree of the feature correlation between the target computing node and the user side is greater than the confidence degrees of the feature correlations between other candidate computing nodes which meet the filtering condition and the user side;

the filtration conditions include at least: the physical distance between the user terminal and the user terminal is the nearest, and/or the load is lower than the set load.

Optionally, the gene feature data of the computing node at least comprises:

the node identification ID of the computing node, the protocol version number of the computing node operation and the data pair reported by adopting the way of combining the keyword-value KV pair are combined, and the data pair comprises: the method comprises the steps of signaling ID and data packet data, wherein when the signaling ID is a monitoring address combination identifier ADDR, the data packet data represents monitoring information consisting of a public network IP address of a computing node and a monitoring Port of the computing node; and when the signaling ID is the STATUS, the data packet data represents the STATUS identifier of the computing node, and the STATUS identifier represents idle or working.

Optionally, the first receiving unit further receives a logout parameter sent by the computing node when logging out, where the logout parameter at least carries: calculating node identification ID and protocol version number of the operation of the calculating node; the first processing unit further carries out node logout logic processing on the logout parameter and returns a processing result to the computing node; the processing result comprises: successful logoff or at least one re-logoff is required to be tried again; and/or the presence of a gas in the gas,

the first receiving unit further receives a heartbeat keep-alive message sent by the computing node, wherein the heartbeat keep-alive message at least carries: calculating node identification ID and protocol version number of the operation of the calculating node; the first processing unit further returns a response result to the calculation result; the response result includes: the heartbeat keep-alive message is successfully sent, or the heartbeat keep-alive message needs to be repeatedly sent at least once again, or the node management equipment is abnormal, or the version of the computing node needs to be upgraded, or the computing node is logged off, or the computing node is pulled into a blacklist;

and when the response result comprises that the heartbeat keep-alive message is successfully sent, determining that the computing node sending the heartbeat keep-alive message is online.

Thus, the description of the structure of the apparatus shown in fig. 10 is completed.

Referring to fig. 11, fig. 11 is a structural diagram of another apparatus according to an embodiment of the present disclosure. The device is applied to node management equipment and comprises:

Optionally, the determining, by the second determining unit, a candidate storage node for storing the calculation data generated by the calculation task according to the obtained genetic characteristic data of the user side and the obtained genetic characteristic data of each storage node includes:

if the computing task is the first computing task sent by the user side, determining candidate storage nodes for storing computing data generated by the computing task according to the obtained gene characteristic data of the user side and the obtained gene characteristic data of each storage node;

if the computing task is not the first computing task sent by the user side, determining all computing nodes in a storage node distribution set generated for the user side as candidate storage nodes for storing computing data generated by the computing task; the storage node distribution set is determined according to the obtained gene characteristic data of the user side, the obtained gene characteristic data of each storage node and the history storage nodes distributed by the user side before.

Thus, the structure tracing of the apparatus shown in FIG. 11 is completed.

The embodiment of the application also provides a hardware structure of the device shown in fig. 10 or fig. 11. Referring to fig. 12, fig. 12 is a block diagram of an electronic device according to an embodiment of the present disclosure. As shown in fig. 12, the hardware structure may include: a processor and a machine-readable storage medium having stored thereon machine-executable instructions executable by the processor; the processor is configured to execute machine-executable instructions to implement the methods disclosed in the above examples of the present application.

Based on the same application concept as the method, embodiments of the present application further provide a machine-readable storage medium, where several computer instructions are stored, and when the computer instructions are executed by a processor, the method disclosed in the above example of the present application can be implemented.

The machine-readable storage medium may be, for example, any electronic, magnetic, optical, or other physical storage device that can contain or store information such as executable instructions, data, and the like. For example, the machine-readable storage medium may be: a RAM (random Access Memory), a volatile Memory, a non-volatile Memory, a flash Memory, a storage drive (e.g., a hard drive), a solid state drive, any type of storage disk (e.g., an optical disk, a dvd, etc.), or similar storage medium, or a combination thereof.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. A typical implementation device is a computer, which may take the form of a personal computer, laptop computer, cellular telephone, camera phone, smart phone, personal digital assistant, media player, navigation device, email messaging device, game console, tablet computer, wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the units may be implemented in one or more software and/or hardware when implementing the present application.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

Furthermore, these computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A node distribution method is applied to a node management device and comprises the following steps:

receiving a computing task of a user side;

2. The method of claim 1, further comprising, prior to the method:

receiving registration parameters sent by a computing node during registration; the registration parameter is used for representing a parameter required to be subjected to registration authentication when the computing node needs to be registered to the node management equipment; carrying out node registration logic processing on the registration parameters, and returning registration results to the computing nodes sending the registration parameters;

3. The method of claim 2, wherein the registration parameters comprise at least:

4. The method of claim 1, wherein the determining candidate computing nodes for performing the computing task according to the obtained genetic characteristic data of the user side and the obtained genetic characteristic data of each computing node registered to the device comprises:

5. The method of claim 1, wherein each candidate compute node has a corresponding confidence level with respect to the feature correlation of the user side; the confidence level is used for indicating the possibility that the candidate computing node is allocated to the user side;

6. The method of claim 1, wherein the genetic signature data of the computing nodes comprises at least:

7. The method of claim 1, further comprising:

receiving a logout parameter sent by a computing node during logout, wherein the logout parameter at least carries: calculating node identification ID and protocol version number of the operation of the calculating node; carrying out node logout logic processing on the logout parameters, and returning a processing result to the computing node; the processing result comprises: successful logoff or at least one re-logoff is required to be tried again; and/or the presence of a gas in the gas,

receiving a heartbeat keep-alive message sent by a computing node, wherein the heartbeat keep-alive message at least carries: calculating node identification ID and protocol version number of the operation of the calculating node; returning a response result to the calculation result; the response result includes: the heartbeat keep-alive message is successfully sent, or the heartbeat keep-alive message needs to be repeatedly sent at least once again, or the node management equipment is abnormal, or the version of the computing node needs to be upgraded, or the computing node is logged off, or the computing node is pulled into a blacklist;

8. A node distribution method is applied to a node management device and comprises the following steps:

receiving a computing task of a user side;

9. The method of claim 8, wherein determining candidate storage nodes for storing the computation data generated by the computation task according to the obtained genetic characteristic data of the user side and the obtained genetic characteristic data of each storage node comprises:

10. A node distribution system is characterized by comprising node management equipment, a computing node and a user side;

the node management apparatus performing according to the method of any of claims 1 to 7.

11. A node distribution system, the system comprising: the system comprises node management equipment, computing nodes, storage nodes and a user side;

the node management apparatus performing according to any of the methods of claims 8 to 9.

12. A node allocation device, applied to a node management device, includes:

the first receiving unit is used for receiving a computing task of a user side;

13. A node allocation device, applied to a node management device, includes:

14. An electronic device, comprising: a processor and a machine-readable storage medium;

the processor is configured to execute machine executable instructions to perform the method steps of any of claims 1-9.