CN112118278B

CN112118278B - Computing node access method, device, electronic equipment and machine-readable storage medium

Info

Publication number: CN112118278B
Application number: CN201910759229.6A
Authority: CN
Inventors: 徐立峰
Original assignee: Hangzhou Hikvision System Technology Co Ltd
Current assignee: Hangzhou Hikvision System Technology Co Ltd
Priority date: 2019-06-04
Filing date: 2019-08-16
Publication date: 2023-07-04
Anticipated expiration: 2039-08-16
Also published as: CN112118278A

Abstract

The embodiment of the invention provides a computing node access method, a device, electronic equipment and a machine-readable storage medium, wherein the computing node access method comprises the following steps: acquiring a first attribute feature group of a to-be-accessed computing node, a second attribute feature group of each access node accessed to the Internet of things and a third attribute feature group of each computing node accessed to the Internet of things, searching attribute features contained in the first attribute feature group from each second attribute feature group, determining the node type of the access node with the attribute features according to the searched attribute features and each third attribute feature group, and accessing the to-be-accessed computing node to the Internet of things according to a preset access mode corresponding to the node type. Through the scheme, the overall access and calculation efficiency of the Internet of things can be improved.

Description

Computing node access method, device, electronic equipment and machine-readable storage medium

Technical Field

The present invention relates to the field of internet of things, and in particular, to a method and apparatus for accessing a computing node, an electronic device, and a machine-readable storage medium.

Background

The internet of things is a network which is based on information carriers such as the Internet, a traditional telecommunication network and the like and enables all common physical objects which can be independently addressed to realize interconnection and intercommunication. The internet of things has the characteristics of common object equipment, autonomous terminal interconnection, universal service intellectualization and the like, so that the internet of things is widely applied in application scenes such as intelligent home, intelligent industry, intelligent enterprises and the like, and more household electrical appliances, vehicle-mounted equipment, processing equipment and the like have networking capability, so that the integrated management of the equipment is facilitated.

The Internet of things comprises two types of equipment nodes: access nodes and computing nodes. The access node is used for data access, the input of the access node is a data source, and preprocessing such as filtering and standardization is carried out on the data so that the processed data has consistency when the computing node carries out data operation and analysis processing, and the access node sends the preprocessed data to the accessed computing node for operation, analysis and other processing.

The computing nodes are connected to the Internet of things, normal data processing can be performed, and in general, each time one computing node is added, all the computing nodes are required to be connected to all access nodes in the Internet of things. However, when the computing node processes data, not all the data of the access nodes are valid data, and the network connection is invalid, so that the overall access and the computing efficiency of the internet of things are low.

Disclosure of Invention

The embodiment of the invention aims to provide a method, a device, electronic equipment and a machine-readable storage medium for accessing a computing node so as to improve the overall access and computing efficiency of the Internet of things. The specific technical scheme is as follows:

in a first aspect, an embodiment of the present invention provides a method for accessing a computing node, where the method includes:

Acquiring a first attribute feature group of a computing node to be accessed, a second attribute feature group of each access node accessed to the Internet of things and a third attribute feature group of each computing node accessed to the Internet of things, wherein the attribute feature groups comprise a plurality of attribute features;

searching attribute features contained in the first attribute feature group from each second attribute feature group;

determining the node type of the access node with the attribute characteristics according to the searched attribute characteristics and each third attribute characteristic group;

and accessing the computing node to be accessed into the Internet of things according to a preset access mode corresponding to the node type.

Optionally, the step of determining the node type of the access node having the attribute feature according to the searched attribute feature and each third attribute feature group includes:

judging whether the attribute features are contained in at least one third attribute feature group according to the searched attribute features and each third attribute feature group;

if the attribute features are not included in any third attribute feature set, determining that the node type of the access node with the attribute features is an outlier node type;

according to a preset access mode corresponding to the node type, accessing the computing node to be accessed to the Internet of things, which comprises the following steps:

Counting all access nodes with the node type being an outlier node type to form an outlier node set;

the method comprises the steps that a to-be-accessed computing node is accessed to each access node in an outlier node set, and computing capacity updated after the to-be-accessed computing node is accessed to each access node is obtained;

and when the computing capacity reaches a preset computing threshold, determining that the computing node to be accessed completes access.

Optionally, the step of accessing the to-be-accessed computing node to each access node in the outlier node set includes:

calculating distance parameters between each access node and a to-be-accessed computing node in the outlier node set;

and sequentially accessing the computing nodes to be accessed to each access node according to descending order of each distance parameter.

if the attribute features are contained in at least one third attribute feature group, acquiring the output-input efficiency ratio of the access node with the attribute features, and judging whether the output-input efficiency ratio reaches a preset efficiency ratio threshold;

If not, determining the node type of the access node with the attribute characteristic as an unsatisfactory node type;

counting all access nodes with unsatisfied node types to form an unsatisfied node set;

obtaining forwarding efficiency parameters of each access node and a to-be-accessed computing node in an unsatisfactory node set;

according to forwarding efficiency parameters of each access node and each to-be-accessed computing node, accessing each access node meeting preset efficiency improving conditions in the to-be-accessed computing node set, and acquiring updated computing capacity of each access node to be accessed computing node;

Optionally, after the step of determining whether the output-to-input efficiency ratio reaches the preset efficiency ratio threshold, the method further includes:

if so, determining the node type of the access node with the attribute characteristics as a satisfactory node type;

Counting all access nodes with the node types being satisfaction node types to form a satisfaction node set;

obtaining connection parameters of each access node and a computing node to be accessed in a satisfied node set;

according to the connection parameters of each access node and the to-be-accessed computing node, accessing each access node meeting the preset connection simplification condition in the to-be-accessed computing node set, and acquiring the updated computing capacity of the to-be-accessed computing node after accessing each access node;

after the computing nodes to be accessed are accessed, acquiring state information of each computing node of each access node in the accessed satisfied node set;

judging whether the state information of at least one computing node in each computing node is changed or not;

if yes, acquiring a third attribute feature group of the computing node with unchanged state information and a third attribute feature group of the computing node with changed state information, taking the third attribute feature group of the computing node with changed state information as a first attribute feature group of the computing node to be accessed, taking the third attribute feature group of the computing node with unchanged state information as a third attribute feature group of each computing node accessed into the Internet of things, and returning to execute the step of searching attribute features contained in the first attribute feature group from each second attribute feature group.

Optionally, after the step of taking the third attribute feature group of the computing node with changed state information as the first attribute feature group of the computing node to be accessed and taking the third attribute feature group of the computing node with unchanged state information as the third attribute feature group of each computing node accessed to the internet of things, the method further includes:

acquiring current refreshing strength, wherein the refreshing strength is preset initial refreshing strength aiming at the initially built internet of things;

according to a preset attenuation strategy, attenuating the refreshing strength;

judging whether the refreshing strength is smaller than a preset strength threshold value or not;

returning to the step of searching the attribute features contained in the first attribute feature group from the second attribute feature groups, the step comprising:

and if the refresh strength is not less than the preset strength threshold, returning to execute the step of searching the attribute features contained in the first attribute feature group from the second attribute feature groups.

In a second aspect, an embodiment of the present invention provides a computing node access device, including:

the system comprises an acquisition module, a storage module and a storage module, wherein the acquisition module is used for acquiring a first attribute feature group of a computing node to be accessed, a second attribute feature group of each access node accessed to the Internet of things and a third attribute feature group of each computing node accessed to the Internet of things, wherein the attribute feature groups comprise a plurality of attribute features;

The searching module is used for searching the attribute characteristics contained in the first attribute characteristic group from the second attribute characteristic groups;

the determining module is used for determining the node type of the access node with the attribute characteristics according to the searched attribute characteristics and each third attribute characteristic group;

and the access module is used for accessing the computing node to be accessed to the Internet of things according to a preset access mode corresponding to the node type.

Optionally, the determining module is specifically configured to:

the access module is specifically used for:

Optionally, the access module is specifically configured to, when configured to access the to-be-accessed computing node to each access node in the set of outlier nodes:

Optionally, the determining module is specifically configured to:

the access module is specifically used for:

Optionally, the determining module is further configured to:

if the output-input efficiency ratio reaches a preset efficiency ratio threshold, determining that the node type of the access node with the attribute characteristic is a satisfactory node type;

the access module is specifically used for:

Optionally, the access module is further configured to:

the access module is used for searching the attribute features contained in the first attribute feature group from each second attribute feature group when being used for returning execution, and is specifically used for:

In a third aspect, an embodiment of the present invention provides an electronic device, including a processor and a memory, where,

the memory is used for storing a computer program;

the processor is configured to implement the method provided in the first aspect of the embodiment of the present invention when executing the computer program stored in the memory.

In a fourth aspect, embodiments of the present invention provide a machine-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method provided by the first aspect of the embodiments of the present invention.

According to the method, the device, the electronic equipment and the machine-readable storage medium for accessing the computing nodes, the first attribute feature group of the computing nodes to be accessed, the second attribute feature groups of all access nodes accessed to the Internet of things and the third attribute feature groups of all computing nodes accessed to the Internet of things are obtained, the attribute features contained in the first attribute feature groups are searched from all the second attribute feature groups, the node types of the access nodes with the attribute features are determined according to the searched attribute features and all the third attribute feature groups, and the computing nodes to be accessed to the Internet of things are accessed according to the preset access modes corresponding to the node types. Based on the first attribute feature group of the computing node to be accessed and the second attribute feature group of each access node, determining the attribute features contained in the first attribute feature group in each second attribute feature group, classifying the access nodes according to the third attribute feature group of each accessed computing node, and accessing the computing node to be accessed in different access modes aiming at different types of access nodes. The attribute characteristics of the access node accessed by the computing node to be accessed are contained in the first attribute characteristic group of the computing node to be accessed, so that the data of the access node accessed by the computing node to be accessed is ensured to be effective data, and the different types of access nodes access the computing node to be accessed in different access modes, so that the effective access of the computing node to be accessed is ensured, invalid network connection is avoided, and the overall access and computing efficiency of the Internet of things are improved.

Drawings

In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of a method for computing node access according to an embodiment of the invention;

FIG. 2 is a flow chart of a method for computing node access according to another embodiment of the present invention;

FIG. 3 is a flowchart of a method for accessing a computing node according to another embodiment of the present invention;

fig. 4 is a schematic structural diagram of a computing node access device according to an embodiment of the present invention;

fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

In order to improve overall access and calculation efficiency of the Internet of things, the embodiment of the invention provides a method and a device for accessing a calculation node, electronic equipment and a machine-readable storage medium. The following first describes a method for accessing a computing node according to an embodiment of the present invention.

The execution main body of the computing node access method provided by the embodiment of the invention is an electronic device, which can be a server, a controller and the like in the Internet of things, and at least comprises a core processing chip with data processing capability. The manner of implementing the method for accessing the computing node provided by the embodiment of the invention can be at least one manner of software, hardware circuits and logic circuits arranged in an execution main body.

As shown in fig. 1, a method for accessing a computing node according to an embodiment of the present invention may include the following steps:

s101, acquiring a first attribute feature group of a computing node to be accessed, a second attribute feature group of each access node accessed to the Internet of things and a third attribute feature group of each computing node accessed to the Internet of things, wherein the attribute feature groups comprise a plurality of attribute features.

The computing node is a device node used for performing operations, analysis and other processes on data in the internet of things, such as a processor, a server, a personal computer and the like, and the access node is a device used for data access in the internet of things, such as a switch, a router, an access gateway and the like.

The computing node and the access node are correspondingly provided with attribute features, and the attribute features represent inherent attributes of network setting and type setting of data which the computing node needs to calculate or the access node needs to access, such as network attributes (such as IP address range, flow requirement, local area network limitation and the like) to which the data belongs, data type attributes (such as video type data, audio type data, picture type data and the like), data time attributes (such as data of 0:00-8:00) and the like. One attribute feature is a combination of several attributes, for example, data of a video type of an x.x.x.x network segment collected from 0:00 to 8:00 can be one attribute feature, and a plurality of attribute features can form an attribute feature group.

For the internet of things, a feature space V can be pre-established based on the topological structure of the internet of things, wherein the feature space V is a multidimensional feature internet of things space, the feature space V comprises a plurality of first subspaces, each first subspace is a third attribute feature group of one computing node accessed in the internet of things, and if the number of the computing nodes accessed in the internet of things is m, the corresponding first subspaces are m, and are V respectively _i (i∈[1,m])。

S102, searching the attribute features contained in the first attribute feature group from the second attribute feature groups.

If one attribute feature of the access node is included in the first attribute feature group of the computing node to be accessed, it is stated that the data input to the access node with the attribute feature can be operated by the computing node to be accessed, so that the attribute feature included in the first attribute feature group should be screened out of the second attribute feature groups first, that is, the input data need to be operated by the computing node to be accessed is screened out of the built internet of things.

If each attribute feature in the second attribute feature set specifically includes a data time attribute, the data time attribute needs to be refined, for example, each hour, and by such division, it can be determined which data specific to the access node needs to be input to the computing node to be accessed. For example, there are two access nodes, the second attribute feature set of the access node 1 is {0:00-8:00 video type data }, 14:00-17:00 video type data }, the second attribute feature set of the access node 2 is {6:00-8:00 collected voice type data, 15:00-16:00 collected document type data }, and the first attribute feature set of the computing node to be accessed is {6:00-10:00 video type data, 15:00-16:00 document type data }, then the attribute features included in the first attribute feature set can be found: 6:00-8:00 video type data and 15:00-16:00 document type data.

In a specific implementation manner, a subspace may be established for the to-be-accessed computing node, where the subspace is the first attribute feature set of the to-be-accessed computing node, for example, for the to-be-accessed computing node

Can establish subspace V _n . In this way, it is possible to directly determine which attribute features in each second attribute feature group are included in the subspace V _n 。

S103, determining the node type of the access node with the attribute characteristics according to the searched attribute characteristics and the third attribute characteristic groups.

According to the searched attribute characteristics and each third attribute characteristic group, the node type of the access node with the attribute characteristics can be determined, and the node types of the access node are mainly divided into three categories: an outlier node type, an unsatisfied node type, and a satisfied node type.

If the searched attribute characteristics are not contained in any third attribute characteristic group, the node type of the access node with the attribute characteristics is an outlier node type; if the searched attribute features are contained in at least one third attribute feature group and the output-input efficiency ratio of the access node with the attribute features does not reach the preset efficiency ratio threshold, the node type of the access node is an unsatisfactory node type; if the searched attribute features are contained in at least one third attribute feature group, and the output-input efficiency ratio of the access node with the attribute features reaches a preset efficiency ratio threshold, the node type of the access node is a satisfactory node type. The ratio of output and input efficiency of an access node refers to the ratio of the data volume output by the access node to the data volume input, and the smaller the output and input efficiency is, the less data is sent to a computing node on the access node for computation, and the more data is accumulated on the access node without operation. Therefore, in the embodiment of the invention, a preset efficiency ratio threshold is set, if the output/input efficiency ratio does not reach the preset efficiency ratio threshold, the accumulated data which is not operated on the access node is more, so the node type of the access node is set as unsatisfactory node type; if the ratio of output to input efficiency reaches the preset ratio of efficiency threshold, the data accumulated on the access node is less, so the node type of the access node is set as a satisfactory node type.

Furthermore, if the access node includes a cache module, whether the node type of the access node is an unsatisfactory node type or a satisfactory node type should also be considered for the cache state of the cache module: if the output-input efficiency ratio of the access node does not reach the preset efficiency ratio threshold value or the buffering quantity of the buffering module does not reach the preset buffering threshold value, the node type of the access node is an unsatisfactory node type; if the output-input efficiency ratio of the access node reaches a preset efficiency ratio threshold value and the buffer capacity of the buffer module reaches a preset buffer threshold value, the node type of the access node is a satisfactory node type.

Optionally, S103 may specifically be:

judging whether the attribute features are contained in at least one third attribute feature group according to the searched attribute features and the third attribute feature groups; if the attribute is not included in any of the third attribute sets, determining that the node type of the access node having the attribute is an outlier node type.

The access node of the outlier node type refers to an access node which is not in topological connection with other node equipment in an internet of things topological structure in a certain specified statistical mode, and the output/input efficiency ratio of the access node of the outlier node type does not reach a preset efficiency ratio threshold value, namely the outlier node type can be understood as an unsatisfactory node type. Specifically, an access node of an outlier type may be expressed as:

Wherein I is _d An access node representing an outlier node type, T (x) representing an output-to-input efficiency ratio of the access node x, S representing a preset efficiency ratio threshold, V _n Representing a first attribute feature set (i.e., a second subspace) of a computing node to be accessed, x ε V _n The attribute characteristics representing access node x are included in a first set of attribute characteristics, V _i A third attribute feature set (i.e., a first subspace) representing an ith computing node of the accessed internet of things, m representing a total number of computing nodes of the accessed internet of things, F (x, V) _i ) The attribute characteristic representing access node x is included in a third set of attribute characteristics V _i Is a counting function of F (x, V) _i ) The concrete steps are as follows:

optionally, S103 may specifically be:

judging whether the attribute features are contained in at least one third attribute feature group according to the searched attribute features and each third attribute feature group; if the attribute features are contained in at least one third attribute feature group, acquiring the output-input efficiency ratio of the access node with the attribute features, and judging whether the output-input efficiency ratio reaches a preset efficiency ratio threshold; if not, determining the node type of the access node with the attribute characteristics as an unsatisfactory node type.

The unsatisfactory node type may be determined according to whether the output-to-input efficiency ratio reaches a preset efficiency ratio threshold, and if the output-to-input efficiency ratio does not reach the preset efficiency ratio threshold, the node type of the access node may be determined to be the unsatisfactory node type. Specifically, an access node of an unsatisfactory node type may be expressed as:

Wherein I is _u An access node representing an unsatisfactory node type.

For unsatisfactory node types, it may include: multi-objective unsatisfied node types and single-objective unsatisfied node types. Specifically, the method can be obtained by counting the number of the third attribute feature groups included in the attribute features of the access node, if the number of the third attribute feature groups included in the attribute features of the access node is greater than 1, determining that the node type of the access node is a multi-target unsatisfactory node type, and if the number of the third attribute feature groups included in the attribute features of the access node is equal to 1, determining that the node type of the access node is a single-target unsatisfactory node type.

The number of the third attribute feature groups included in the attribute features of the access node is more than 1, and the number of the third attribute feature groups included in the attribute features of the access node is only 1, so that the specific access node of the multi-objective unsatisfactory node type can be expressed as:

wherein I is _mu An access node representing a multi-target unsatisfactory node type.

An access node of a single-target unsatisfactory node type may be expressed as:

Wherein I is _su An access node representing a single target unsatisfactory node type.

Optionally, after the step of determining whether the output-input efficiency ratio reaches the preset efficiency ratio threshold, the method for accessing a computing node provided by the embodiment of the present invention may further include the following steps:

and if the output-input efficiency ratio reaches a preset efficiency ratio threshold, determining the node type of the access node with the attribute characteristic as a satisfactory node type.

The type of the satisfied node may be determined according to whether the output-input efficiency ratio reaches a preset efficiency ratio threshold, and if the output-input efficiency ratio reaches the preset efficiency ratio threshold, the node type of the access node may be determined to be the satisfied node type. Specifically, an access node of a satisfactory node type may be expressed as:

wherein I is _s An access node representing a satisfactory node type.

For a satisfactory node type, similar to an unsatisfactory node type, it may also include: multiple target satisfaction node types and single target satisfaction node types. Specifically, the access node of the multi-target satisfaction node type can be expressed as:

wherein I is _ms An access node representing a multi-target satisfaction node type.

The access node of the single target satisfaction node type may be expressed as:

Wherein I is _ss An access node representing a single target satisfaction node type.

And S104, accessing the computing node to be accessed into the Internet of things according to a preset access mode corresponding to the node type.

As described above, the node types of the access node mainly include an outlier node type, an unsatisfied node type and a satisfied node type, and for different node types, corresponding access modes are preset, and the computing node to be accessed is accessed according to different access modes, so that effective access of the computing node to be accessed can be ensured, and invalid network connection is avoided.

Optionally, for the access node of the outlier node type, the access process of the computing node to be accessed mainly includes:

counting all access nodes with the node type being an outlier node type to form an outlier node set; the method comprises the steps that a to-be-accessed computing node is accessed to each access node in an outlier node set, and computing capacity updated after the to-be-accessed computing node is accessed to each access node is obtained; and when the computing capacity reaches a preset computing threshold, determining that the computing node to be accessed completes access.

Because the access node of the outlier node type refers to an access node which is not in topological connection with other node equipment in the topological structure of the internet of things in a certain appointed statistical mode, when the computing node to be accessed is accessed, any processing is not needed in advance, and the computing node to be accessed can be directly accessed to each access node. And after the computing nodes to be accessed are accessed, the computing nodes to be accessed can operate the data accessed by each access node, the computing capacity of the computing nodes to be accessed is obtained in real time, if the computing capacity reaches a preset computing threshold value, the current computing capacity of the computing nodes to be accessed is saturated, and the computing nodes to be accessed can be determined to finish the access.

Optionally, for the access node of the unsatisfactory node type, the access process of the computing node to be accessed mainly includes:

counting all access nodes with unsatisfied node types to form an unsatisfied node set; obtaining forwarding efficiency parameters of each access node and a to-be-accessed computing node in an unsatisfactory node set; according to forwarding efficiency parameters of each access node and each to-be-accessed computing node, accessing each access node meeting preset efficiency improving conditions in the to-be-accessed computing node set, and acquiring updated computing capacity of each access node to be accessed computing node; and when the computing capacity reaches a preset computing threshold, determining that the computing node to be accessed completes access.

Since the output/input efficiency of the access node of the unsatisfied node type is lower, more computing nodes can be accessed again, therefore, before accessing the computing nodes to be accessed, forwarding efficiency parameters (such as bandwidth between the access node and the computing nodes to be accessed, data quantity to be forwarded in the access node, etc.) of each access node and the computing nodes to be accessed in the unsatisfied node set can be obtained first, whether the data forwarding efficiency is improved if each access node is accessed to the computing nodes to be accessed is judged, for example, if the bandwidth between the access node and the computing nodes to be accessed is 80M, and the data quantity to be forwarded in a plurality of access nodes exceeds 80M, if the plurality of access nodes access the computing nodes to be accessed, the plurality of access nodes are made to approach to a satisfactory state, and the access node with larger data quantity is selected to access the computing nodes to be accessed to the satisfactory state, therefore, the preset efficiency improving condition can be that the data quantity exceeds the bandwidth and the access is performed in order from large to small. After the to-be-accessed computing nodes are accessed, the to-be-accessed computing nodes operate the accessed data of each access node, the computing capacity of the to-be-accessed computing nodes is obtained in real time, if the computing capacity reaches a preset computing threshold value, the current computing capacity of the to-be-accessed computing nodes is saturated, and the to-be-accessed computing nodes can be determined to finish the access.

Optionally, for the access node of the satisfied node type, the access process of the computing node to be accessed mainly includes:

counting all access nodes with the node types being satisfaction node types to form a satisfaction node set; obtaining connection parameters of each access node and a computing node to be accessed in a satisfied node set; according to the connection parameters of each access node and the to-be-accessed computing node, accessing each access node meeting the preset connection simplification condition in the to-be-accessed computing node set, and acquiring the updated computing capacity of the to-be-accessed computing node after accessing each access node; after the computing nodes to be accessed are accessed, acquiring state information of each computing node of each access node in the accessed satisfied node set; judging whether the state information of at least one computing node in each computing node is changed or not; if yes, acquiring a third attribute feature group of the computing node with unchanged state information and a third attribute feature group of the computing node with changed state information, taking the third attribute feature group of the computing node with changed state information as a first attribute feature group of the computing node to be accessed, taking the third attribute feature group of the computing node with unchanged state information as a third attribute feature group of each computing node accessed to the Internet of things, and returning to execute S102.

For the access node with the satisfied node type, more computing nodes can be accessed again, before the access to the computing node to be accessed, the connection parameters (such as the computing capacity of the computing node to be accessed, the data quantity to be forwarded in the access node and the like) of each access node and the computing node to be accessed in the satisfied node set can be acquired first, the computing capacity of the computing node to be accessed can be judged to the greatest extent, for example, if the computing capacity of the computing node to be accessed is 100M and the sum of the data quantity to be forwarded in a plurality of access nodes is less than 100M, the data of the access nodes can be sent to the computing node to be accessed for computing, the connection number of the access nodes and the computing node can be reduced, and the topology network structure is simplified.

In addition, since the output and input efficiency of the access node of the satisfied node type is higher, the new access computing node may affect the original computing node (such that the original computing node is disconnected from the topology connection and the connection flow is reduced, etc.), after the access operation of the computing node to be accessed is completed, the accessed computing node in the internet of things includes the computing node to be accessed which is just accessed and each computing node which is originally accessed, the state information of the computing nodes is obtained, and whether the state information of at least one computing node is changed in the computing nodes is judged, wherein the state information may be the state of the topology connection, the connection flow strength, etc., when the computing node has the phenomena of disconnection of the topology connection and the connection flow is reduced, the state information of the computing node is described to be changed, at this time, the topology structure of the computing node which is accessed to the satisfied node set needs to be refreshed, the third attribute feature group of the computing node which is not changed in the state information is obtained, the third attribute feature group of the computing node which is changed in the internet of things is obtained, the state information is taken as the first attribute feature group of the computing node which is changed in the state information, and the state information of the computing node which is usually called the third attribute group of the computing node which is changed in the state information of the internet of the computing node is refreshed.

For example, if the original computing node includes 1 to M, the computing node to be accessed is N, and the access of the computing node to be accessed affects the original computing node 3, the original computing nodes 1, 2, 4 to M and the computing node N are taken as the accessed computing nodes, the third attribute feature sets thereof are obtained, the original computing node 3 is taken as the computing node to be accessed, the first attribute feature set thereof is obtained, and the access operation is performed again based on the updated third attribute feature sets and the first attribute feature set.

Optionally, after executing the step of taking the third attribute feature set of the computing node with changed state information as the first attribute feature set of the computing node to be accessed and taking the third attribute feature set of the computing node with unchanged state information as the third attribute feature set of each computing node accessed to the internet of things, the method provided by the embodiment of the invention may further execute:

acquiring current refreshing strength, wherein the refreshing strength is preset initial refreshing strength aiming at the initially built internet of things; according to a preset attenuation strategy, attenuating the refreshing strength; and judging whether the refreshing strength is smaller than a preset strength threshold value.

And when the refresh strength is less than the preset strength threshold, returning to S102.

Since the computing nodes have influence on each other, for example, the computing node 3 affects the computing nodes 1 and 2, the computing nodes 1 and 2 affect the computing nodes 4, 5 and 6, and the computing nodes 4, 5 and 6 affect the computing node 3, which causes jitter in the internet of things. In order to avoid such jitter, the influence of access of one computing node on the topology network is limited, the initial refresh strength can be set according to the number of existing computing nodes in the internet of things, for example, 1000 computing nodes in the internet of things can be set, the initial refresh strength can be set to be 50, and when the topology structure of the internet of things is updated once, the refresh strength is reduced by 1 until the update is reduced to 0, and the update is stopped. Here, the refresh strength may be set to a quantized value, which may be set by a technician according to an actual state and experience of the internet of things, and is not further listed here.

Optionally, the step of accessing the to-be-accessed computing node to each access node in the outlier node set may specifically be:

calculating distance parameters between each access node and a to-be-accessed computing node in the outlier node set; and sequentially accessing the computing nodes to be accessed to each access node according to descending order of each distance parameter.

Optionally, the step of accessing the to-be-accessed computing node to each access node meeting the preset efficiency improvement condition in the unsatisfied node set may specifically be:

calculating distance parameters between each access node and a to-be-accessed computing node in the unsatisfied node set, wherein the distance parameters meet preset efficiency lifting conditions; and sequentially accessing the computing nodes to be accessed to each access node according to descending order of each distance parameter.

Optionally, the step of accessing the to-be-accessed computing node to each access node meeting the preset connection simplification condition in the satisfied node set may specifically be:

calculating distance parameters between each access node and a node to be accessed in the satisfied node set, wherein the distance parameters meet preset connection simplification conditions; and sequentially accessing the computing nodes to be accessed to each access node according to descending order of each distance parameter.

For the step of accessing the computing node to be accessed to each access node, as the types of the access nodes are different, the same type of access node is divided into the same set, as described above, the different types of access nodes are respectively divided into an outlier node set, an unsatisfied node set and a satisfied node set, when the computing node is accessed to the access nodes in each set, the computing node can be accessed unordered to the access nodes in the same set, or can be accessed according to a certain sequence, for example, the access nodes and the computing node to be accessed are arranged in descending order according to the distance parameters of the access nodes; or the access nodes and the calculation nodes to be accessed can be sequentially accessed according to the ascending sequence of the distance parameters of the access nodes and the calculation nodes to be accessed. Wherein the distance parameter is a numerical parameter, such as a vector norm, that characterizes the distance between two nodes. The greater the distance parameter is, the greater the possibility of improving access greedy and topology simplification is, so that it is preferable to sequentially access the computing nodes to be accessed to each access node according to the descending order of the distance parameter.

By applying the embodiment, the first attribute feature group of the to-be-accessed computing node, the second attribute feature groups of all access nodes accessed to the internet of things and the third attribute feature groups of all computing nodes accessed to the internet of things are obtained, the attribute features contained in the first attribute feature group are searched from all the second attribute feature groups, the node type of the access node with the attribute features is determined according to the searched attribute features and all the third attribute feature groups, and the to-be-accessed computing node is accessed to the internet of things according to a preset access mode corresponding to the node type. Based on the first attribute feature group of the computing node to be accessed and the second attribute feature group of each access node, determining the attribute features contained in the first attribute feature group in each second attribute feature group, classifying the access nodes according to the third attribute feature group of each accessed computing node, and accessing the computing node to be accessed in different access modes aiming at different types of access nodes. The attribute characteristics of the access node accessed by the computing node to be accessed are contained in the first attribute characteristic group of the computing node to be accessed, so that the data of the access node accessed by the computing node to be accessed is ensured to be effective data, and the different types of access nodes access the computing node to be accessed in different access modes, so that the effective access of the computing node to be accessed is ensured, invalid network connection is avoided, and the overall access and computing efficiency of the Internet of things are improved.

Because the access nodes of the outlier node type are not accessed to the computing nodes and the output and input efficiency of the access nodes of the unsatisfied node type are lower, the two types of access nodes can access more computing nodes, so that the requirement of access priority can be met; the input and output efficiency of the access node with the satisfied node type is higher, and the new access computing node possibly affects the original computing node, so that the topology optimization of the Internet of things can be optimized first, and the requirement of the priority of the topology optimization can be met. For general internet of things, the default access is preferential, and for some special internet of things, the user may require the topology optimization to be preferential, so that the computing node to be accessed can be selected to be preferentially accessed to an access node of a certain type according to actual conditions.

In the following, taking access priority as an example, a detailed description of a computing node access method provided by the embodiment of the present invention, as shown in fig. 2, may include the following steps:

s201, acquiring a first attribute feature group of a computing node to be accessed, a second attribute feature group of each access node accessed to the Internet of things and a third attribute feature group of each computing node accessed to the Internet of things, wherein the attribute feature groups comprise a plurality of attribute features.

S202, searching the attribute features contained in the first attribute feature group from the second attribute feature groups.

S203, determining the node type of the access node with the attribute characteristics according to the searched attribute characteristics and the third attribute characteristic groups, wherein the node type comprises an outlier node type, an unsatisfactory node type and a satisfactory node type.

Specifically, the manner of determining the node type of each access node is the same as that of the embodiment shown in fig. 1, and will not be described herein.

S204, counting all access nodes with the node types being outlier node types to form an outlier node set.

S205, the to-be-accessed computing node is accessed to each access node in the outlier node set, and the computing capacity of the to-be-accessed computing node updated after being accessed to each access node is obtained.

And S206, determining that the computing node to be accessed completes access when the computing capability reaches a preset computing threshold.

After the computing node to be accessed is accessed to the computing nodes in the outlier node set, the computing capacity of the computing node to be accessed is refreshed in real time, if the computing capacity reaches a preset computing threshold value, the computing node is fully utilized, for example, the CPU utilization rate of the computing node reaches 95%, the completion of the access of the computing node to be accessed can be determined, and the computing node to be accessed has insufficient capacity to process more data, so that more access nodes do not need to be accessed. As shown in the embodiment of fig. 1, the access mode of the computing node to be accessed may be an unordered access mode of accessing each access node in the outlier node set, or may be an access mode of arranging according to a descending order or an ascending order of distance parameters, which is not limited specifically herein. The computing capability may not have access to all access nodes in the set of outlier nodes to the computing node to be accessed when the computing capability reaches a preset computing threshold, and thus, immediately stop continuing access when the computing capability reaches the preset computing threshold.

S207, if the computing capacity of the computing node to be accessed does not reach the preset computing threshold after all the access nodes in the outlier node set are accessed, counting all the access nodes with the node type being the unsatisfied node type to form an unsatisfied node set.

If the to-be-accessed computing node has accessed all the access nodes in the outlier node set, but the computing capacity still does not reach the preset computing threshold, more access nodes can be continuously accessed, and no access nodes which can be accessed in the outlier node set exist, so that the access nodes in the unsatisfied node set can be continuously accessed until the computing capacity reaches the preset computing threshold.

S208, acquiring and according to forwarding efficiency parameters of each access node and each to-be-accessed computing node in the dissatisfied node set, accessing each access node meeting preset efficiency improvement conditions in the dissatisfied node set to the to-be-accessed computing node, and acquiring updated computing capacity of each access node to be accessed computing node.

And S209, determining that the computing node to be accessed completes access when the computing capability reaches a preset computing threshold.

As shown in the embodiment of fig. 1, the access mode of the computing node to be accessed may be an unordered access dissatisfied node set, or may be an access mode in which the access nodes are arranged in descending order or ascending order of distance parameters, which is not limited specifically herein.

S210, if the computing capacity of the computing node to be accessed does not reach the preset computing threshold after all the access nodes in the unsatisfied node set are accessed, counting all the access nodes with the node type being the satisfied node type to form the satisfied node set.

If the to-be-accessed computing node has accessed all the access nodes in the outlier node set and the dissatisfied node set, but the computing capacity still does not reach the preset computing threshold, more access nodes can be further accessed, and no access nodes which can be accessed yet exist in the outlier node set and the dissatisfied node set, so that the access nodes in the satisfied node set can be further accessed until the computing capacity reaches the preset computing threshold.

S211, according to the connection parameters of each access node and the to-be-accessed computing node in the satisfied node set, accessing the to-be-accessed computing node into each access node meeting the preset connection simplification condition in the satisfied node set, and acquiring the updated computing capacity of the to-be-accessed computing node after accessing each access node.

S212, judging whether the access of the to-be-accessed computing node affects the computing node of each access node in the accessed satisfied node set, if so, executing S213, and if not, executing S214.

S213, obtaining a third attribute feature set of the unaffected computing nodes and the third attribute feature set of the affected computing nodes, taking the third attribute feature set of the affected computing nodes as a first attribute feature set of the computing nodes to be accessed, taking the third attribute feature set of the unaffected computing nodes as the third attribute feature set of each computing node accessed to the Internet of things, and returning to S202.

And S214, determining that the computing node to be accessed completes access when the computing capability reaches a preset computing threshold.

As shown in the embodiment of fig. 1, the access mode of the computing node to be accessed may be an unordered access mode of accessing each access node in the satisfied node set, or may be an access mode of arranging according to a descending order or an ascending order of distance parameters, which is not limited specifically herein. If the computing capacity of the computing node to be accessed does not reach the preset computing threshold after all the access nodes in the satisfied node set are accessed, the fact that all the access nodes in the Internet of things are traversed is indicated, no redundant access nodes can access the computing node to be accessed, and the access is stopped.

For the embodiment of the topology optimization priority, after receiving the instruction of the topology optimization priority input by the user, the to-be-accessed computing node is accessed to the access node in the satisfied node set, if the computing capacity does not reach the preset computing threshold, then the access node in the outlier node set is accessed, and if the computing capacity does not reach the preset computing threshold, then the access node in the unsatisfied node set is accessed. The detailed procedure is similar to the access preferred embodiment described above and will not be described in detail here.

According to the embodiment, based on the first attribute feature group of the computing node to be accessed and the second attribute feature groups of all the access nodes, the attribute features contained in the first attribute feature group in all the second attribute feature groups are determined, and according to the third attribute feature groups of all the accessed computing nodes, the access nodes can be classified, and the computing nodes to be accessed are accessed in different access modes aiming at different types of access nodes. The attribute characteristics of the access node accessed by the computing node to be accessed are contained in the first attribute characteristic group of the computing node to be accessed, so that the data of the access node accessed by the computing node to be accessed is ensured to be effective data, and the different types of access nodes access the computing node to be accessed in different access modes, so that the effective access of the computing node to be accessed is ensured, invalid network connection is avoided, and the overall access and computing efficiency of the Internet of things are improved.

And according to different types of access nodes, based on the requirement of common access priority, the access node of the to-be-accessed computing node type is preferentially accessed to the access node of the outlier node type, and under the condition that the computing capacity does not reach the preset computing threshold, the to-be-accessed computing node is accessed to the access node of the unsatisfied node type, and if the computing capacity does not reach the preset computing threshold, the to-be-accessed computing node is accessed to the access node of the satisfied node type, so that the effective access of the computing node is ensured.

The dissatisfied node types include a multi-target dissatisfied node type and a single-target dissatisfied node type, and the satisfied node types include a multi-target satisfied node type and a single-target satisfied node type. When the to-be-accessed computing node is accessed to all the access nodes in the dissatisfied node set, the access can be performed according to the sequence of firstly accessing the access nodes of the multi-target dissatisfied node type and then accessing the access nodes of the single-target dissatisfied node type, or the access can be performed according to the sequence of firstly accessing the access nodes of the single-target dissatisfied node type and then accessing the access nodes of the multi-target dissatisfied node type, the access sequence does not need to be specific, and all the access nodes of the dissatisfied node type are accessed in the disordered sequence; similarly, when the to-be-accessed computing node is accessed to each access node in the satisfaction node set, the access sequence can be according to the sequence of accessing the access node of the multi-target satisfaction node type and then accessing the access node of the single-target satisfaction node type, or the sequence of accessing the access node of the single-target satisfaction node type and then accessing the access node of the multi-target satisfaction node type, or the access sequence does not need to be specific, and the access nodes of all the satisfaction node types are accessed in disorder sequence.

The following are in access order: the outlier node type, the multi-target unsatisfied node type, the single-target unsatisfied node type, the multi-target satisfied node type and the single-target satisfied node are taken as examples, the method for accessing the computing node provided by the embodiment of the invention is described in detail, and other access sequences are similar to the embodiment and are not repeated here. As shown in fig. 3, a computing node access method may include the steps of:

the first step, the computing node to be accessed adds the space of the Internet of things, and the computing capacity of the computing node to be accessed is C0.

For the Internet of things, a multidimensional feature Internet of things space V is established in advance; m computing nodes exist in the internet of things, and an independent subspace V is built for each existing computing node _i (i∈[1,m]) The established independent subspace is the third attribute feature set in the embodiment; establishing a subspace V for a computing node to be accessed _n The subspace is the first attribute feature set in the above embodiment. According to the second attribute characteristic group and subspace V of each access node in the Internet of things _n And each subspace V _i The node type of each access node may be determined as an outlier node type, a multi-target unsatisfied node type, a single-target unsatisfied node type, a multi-target satisfied node type, or a single-target satisfied node type. The specific node type determination manner is shown in the embodiment of fig. 1, and will not be described herein.

And secondly, accessing the to-be-accessed computing node into each access node of the outlier node type, and updating the computing capacity of the to-be-accessed computing node to be C1.

Specifically, the manner of accessing the computing node to be accessed to the access node of the outlier node type may be that the access nodes of the outlier node type are counted to form an outlier node set, vector norms of each access node and the computing node to be accessed in the outlier node set are sequentially calculated, the vector norms are arranged in descending order, and the computing node to be accessed to the corresponding computing node is sequentially accessed to the computing node to be accessed, which is more detailed in the embodiment shown in fig. 1 and 2 and will not be repeated herein.

And thirdly, judging whether the computing capacity reaches a preset computing threshold, if so, executing a fifteenth step, otherwise, executing a fourth step.

And fourthly, accessing the to-be-accessed computing node into each access node of the multi-target unsatisfactory node type, and updating the computing capacity of the to-be-accessed computing node to be C2.

Specifically, the manner of accessing the to-be-accessed computing node to the access node of the multi-objective unsatisfactory node type may be to count the access nodes of the multi-objective unsatisfactory node type to form a multi-objective unsatisfactory node set, sequentially calculate distance parameters between each access node in the multi-objective unsatisfactory node set and the to-be-accessed computing node, and sequentially arrange the distance parameters in descending order, so that the to-be-accessed computing node is accessed to the corresponding computing node, which is more detailed with reference to the embodiments shown in fig. 1 and 2 and will not be repeated herein.

And fifthly, judging whether the computing capacity reaches a preset computing threshold, if so, executing a fifteenth step, otherwise, executing a sixth step.

And sixthly, accessing the to-be-accessed computing node into each access node of the single-target unsatisfied node type, and updating the computing capability of the to-be-accessed computing node to be C3.

Specifically, the manner of accessing the to-be-accessed computing node to the access node of the single-target unsatisfied node type may be that the access node of the single-target unsatisfied node type is counted to form a single-target unsatisfied node set, the distance parameters between each access node in the single-target unsatisfied node set and the to-be-accessed computing node are sequentially calculated, the distance parameters are arranged in descending order, and the to-be-accessed computing node is sequentially accessed to the corresponding computing node, which is more detailed, refer to the embodiment shown in fig. 1 and fig. 2, and will not be repeated herein.

And seventhly, judging whether the computing capacity reaches a preset computing threshold, if so, executing the fifteenth step, and otherwise, executing the eighth step.

And eighth step, judging whether an updatable access node with multi-objective satisfied node type exists, if so, executing the ninth step, otherwise, executing the twelfth step.

An access node is considered to be an updatable access node of the multi-target satisfaction node type if access of the computing node to be accessed affects the computing nodes of each access node of the multi-target satisfaction node type.

And ninth, accessing the to-be-accessed computing node into each access node of the multi-target satisfactory node type, and updating the computing capability of the to-be-accessed computing node to be C4.

Specifically, the manner of accessing the to-be-accessed computing node to the access node of the multi-target satisfaction node type may be that the access node of the multi-target satisfaction node type is counted to form a multi-target satisfaction node set, distance parameters between each access node and the to-be-accessed computing node in the multi-target satisfaction node set are sequentially calculated, the distance parameters are arranged in a descending order, and the to-be-accessed computing node is sequentially accessed to the corresponding computing node, which is more detailed with reference to the embodiment shown in fig. 1 and 2, and details are not repeated here.

And tenth, updating the computing nodes of the access nodes which are accessed to the multi-target satisfied node type, and returning to execute the second step.

The method comprises the steps of updating the computing nodes of all access nodes which are accessed into the multi-target satisfaction node type, namely updating all first subspaces and second subspaces after the computing nodes to be accessed are accessed into the access nodes of the satisfaction node type and influence the original computing nodes, and the updating is called refreshing pulse. Further, as an optional operation, the intensity of the refresh pulse can be quantified, and an attenuation condition of the intensity can be defined, so that the influence of one computing node access on the topology structure can be limited, if the intensity is reduced by 1 once the refresh pulse is defined, and if the intensity is reduced to 0, the refresh is stopped.

And eleventh, judging whether the computing capacity reaches a preset computing threshold, if so, executing fifteenth, otherwise, executing twelfth.

And a twelfth step of judging whether an updatable single-target satisfactory node type access node exists, if so, executing the thirteenth step, otherwise, executing the fifteenth step.

An access node is considered to be an updatable access node of a single-target satisfaction node type if access of the computing node to be accessed affects the computing nodes of each access node of the single-target satisfaction node type.

And thirteenth, accessing the to-be-accessed computing node into each access node of the single-target satisfactory node type, and updating the computing capacity of the to-be-accessed computing node to be C5.

Specifically, the manner of accessing the to-be-accessed computing node to the access node of the single-target satisfaction node type may be that the access node of the single-target satisfaction node type is counted to form a single-target satisfaction node set, vector norms of each access node and the to-be-accessed computing node in the single-target satisfaction node set are sequentially calculated, the vector norms are arranged in descending order, and the to-be-accessed computing node is sequentially accessed to the corresponding computing node, which is more detailed with reference to the embodiment shown in fig. 1 and 2, and details are not repeated here.

And fourteenth step, updating the computing nodes of each access node which has accessed the single-target satisfaction node type, and returning to execute the second step.

And fifteenth, determining that the access is completed.

According to the scheme, according to the relation among the second attribute feature group of the access node, the subspace of the newly added computing node and the subspaces of other computing nodes, the access nodes are classified, the access or refreshing of the topological structure of the Internet of things is carried out layer by layer, and the topological structure is gradually approximated to the optimal topological structure; and the access of one computing node can transfer the topological structure influence brought by the computing node to the whole Internet of things through the access node in the subspace of the computing node, so that the problems of unbalanced resources, low resource utilization rate and the like caused by local generation of computing caused by the access of the node can be effectively avoided, and meanwhile, the attenuation strategy of refresh pulses can be set according to the network complexity degree, so that the high-resource utilization and high-efficiency balance are achieved. And because the access node accessed by the computing node to be accessed belongs to the subspace corresponding to the computing node to be accessed, the data of the access node accessed by the computing node to be accessed is ensured to be effective data, and the different types of access nodes are accessed to the computing node to be accessed in different access modes, so that the effective access of the computing node to be accessed is ensured, invalid network connection is avoided, and therefore, the overall access and computing efficiency of the Internet of things are improved.

Corresponding to the above method embodiment, the embodiment of the present invention further provides a computing node access device, as shown in fig. 4, where the computing node access device may include:

the obtaining module 410 is configured to obtain a first attribute feature set of a computing node to be accessed, a second attribute feature set of each access node that has accessed the internet of things, and a third attribute feature set of each computing node that has accessed the internet of things, where the attribute feature sets include a plurality of attribute features.

The searching module 420 is configured to search for the attribute features included in the first attribute feature set from each second attribute feature set.

A determining module 430, configured to determine a node type of the access node having the attribute feature according to the found attribute feature and each third attribute feature set.

The access module 440 is configured to access the computing node to be accessed to the internet of things according to a preset access mode corresponding to the node type.

Optionally, the determining module 430 may specifically be configured to: judging whether the attribute features are contained in at least one third attribute feature group according to the searched attribute features and each third attribute feature group; if the attribute is not included in any of the third attribute sets, determining that the node type of the access node having the attribute is an outlier node type.

The access module 440 may specifically be configured to: counting all access nodes with the node type being an outlier node type to form an outlier node set; the method comprises the steps that a to-be-accessed computing node is accessed to each access node in an outlier node set, and computing capacity updated after the to-be-accessed computing node is accessed to each access node is obtained; and when the computing capacity reaches a preset computing threshold, determining that the computing node to be accessed completes access.

Optionally, the access module 440, when configured to access the to-be-accessed computing node to each access node in the set of outlier nodes, may specifically be configured to: calculating distance parameters between each access node and a to-be-accessed computing node in the outlier node set; and sequentially accessing the computing nodes to be accessed to each access node according to descending order of each distance parameter.

Optionally, the determining module 430 may specifically be configured to: judging whether the attribute features are contained in at least one third attribute feature group according to the searched attribute features and each third attribute feature group; if the attribute features are contained in at least one third attribute feature group, acquiring the output-input efficiency ratio of the access node with the attribute features, and judging whether the output-input efficiency ratio reaches a preset efficiency ratio threshold; if not, determining the node type of the access node with the attribute characteristics as an unsatisfactory node type.

The access module 440 is specifically configured to: counting all access nodes with unsatisfied node types to form an unsatisfied node set; obtaining forwarding efficiency parameters of each access node and a to-be-accessed computing node in an unsatisfactory node set; according to forwarding efficiency parameters of each access node and each to-be-accessed computing node, accessing each access node meeting preset efficiency improving conditions in the to-be-accessed computing node set, and acquiring updated computing capacity of each access node to be accessed computing node; and when the computing capacity reaches a preset computing threshold, determining that the computing node to be accessed completes access.

Optionally, the determining module 430 may be further configured to: and if the output-input efficiency ratio reaches a preset efficiency ratio threshold, determining the node type of the access node with the attribute characteristic as a satisfactory node type.

The access module 440 may specifically be configured to: counting all access nodes with the node types being satisfaction node types to form a satisfaction node set; obtaining connection parameters of each access node and a computing node to be accessed in a satisfied node set; according to the connection parameters of each access node and the to-be-accessed computing node, accessing each access node meeting the preset connection simplification condition in the to-be-accessed computing node set, and acquiring the updated computing capacity of the to-be-accessed computing node after accessing each access node; after the computing nodes to be accessed are accessed, acquiring state information of each computing node of each access node in the accessed satisfied node set; judging whether the state information of at least one computing node in each computing node is changed or not; if yes, acquiring a third attribute feature group of the computing node with unchanged state information and a third attribute feature group of the computing node with changed state information, taking the third attribute feature group of the computing node with changed state information as a first attribute feature group of the computing node to be accessed, taking the third attribute feature group of the computing node with unchanged state information as a third attribute feature group of each computing node accessed into the Internet of things, and returning to execute the step of searching attribute features contained in the first attribute feature group from each second attribute feature group.

Optionally, the access module 440 may be further configured to: acquiring current refreshing strength, wherein the refreshing strength is preset initial refreshing strength aiming at the initially built internet of things; according to a preset attenuation strategy, attenuating the refreshing strength; and judging whether the refreshing strength is smaller than a preset strength threshold value.

The access module 440, when configured to perform the searching for the attribute features included in the first attribute feature set from the second attribute feature sets, may specifically be configured to: and if the refresh strength is not less than the preset strength threshold, returning to execute the step of searching the attribute features contained in the first attribute feature group from the second attribute feature groups.

The embodiment of the invention also provides an electronic device, as shown in fig. 5, comprising a processor 501 and a memory 502, wherein,

a memory 502 for storing a computer program;

the processor 501 is configured to implement all the steps of the method for accessing a computing node according to the embodiment of the present invention when executing the computer program stored in the memory 502.

The Memory may include RAM (Random Access Memory ) or NVM (Non-Volatile Memory), such as at least one magnetic disk Memory. Optionally, the memory may be at least one memory device located remotely from the processor.

The processor may be a general-purpose processor, including a CPU (Central Processing Unit ), NP (Network Processor, network processor), etc.; but also DSP (Digital Signal Processing, digital signal processor), ASIC (Application Specific Integrated Circuit ), FPGA (Field-Programmable Gate Array, field programmable gate array) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.

The memory 502 and the processor 501 may be in data transmission through a wired connection or a wireless connection, and the electronic device may communicate with other devices through a wired communication interface or a wireless communication interface. The example of data transmission between the processor 501 and the memory 502 through the bus is shown in fig. 5, and is not limited to a specific connection manner.

In this embodiment, the processor of the electronic device can implement by reading the computer program stored in the memory and by running the computer program: the method comprises the steps of obtaining a first attribute feature group of a to-be-accessed computing node, second attribute feature groups of all access nodes accessed to the Internet of things and third attribute feature groups of all computing nodes accessed to the Internet of things, searching attribute features contained in the first attribute feature groups from all the second attribute feature groups, determining node types of the access nodes with the attribute features according to the searched attribute features and all the third attribute feature groups, and accessing the to-be-accessed computing node to the Internet of things according to a preset access mode corresponding to the node types. Based on the first attribute feature group of the computing node to be accessed and the second attribute feature group of each access node, determining the attribute features contained in the first attribute feature group in each second attribute feature group, classifying the access nodes according to the third attribute feature group of each accessed computing node, and accessing the computing node to be accessed in different access modes aiming at different types of access nodes. The attribute characteristics of the access node accessed by the computing node to be accessed are contained in the first attribute characteristic group of the computing node to be accessed, so that the data of the access node accessed by the computing node to be accessed is ensured to be effective data, and the different types of access nodes access the computing node to be accessed in different access modes, so that the effective access of the computing node to be accessed is ensured, invalid network connection is avoided, and the overall access and computing efficiency of the Internet of things are improved.

In addition, the embodiment of the invention also provides a machine-readable storage medium, and a computer program is stored in the machine-readable storage medium, and when the computer program is executed by a processor, all the steps of the computing node access method provided by the embodiment of the invention are realized.

In this embodiment, the machine-readable storage medium stores a computer program for executing the computing node access method provided by the embodiment of the present invention at runtime, so that it can implement: the method comprises the steps of obtaining a first attribute feature group of a to-be-accessed computing node, second attribute feature groups of all access nodes accessed to the Internet of things and third attribute feature groups of all computing nodes accessed to the Internet of things, searching attribute features contained in the first attribute feature groups from all the second attribute feature groups, determining node types of the access nodes with the attribute features according to the searched attribute features and all the third attribute feature groups, and accessing the to-be-accessed computing node to the Internet of things according to a preset access mode corresponding to the node types. Based on the first attribute feature group of the computing node to be accessed and the second attribute feature group of each access node, determining the attribute features contained in the first attribute feature group in each second attribute feature group, classifying the access nodes according to the third attribute feature group of each accessed computing node, and accessing the computing node to be accessed in different access modes aiming at different types of access nodes. The attribute characteristics of the access node accessed by the computing node to be accessed are contained in the first attribute characteristic group of the computing node to be accessed, so that the data of the access node accessed by the computing node to be accessed is ensured to be effective data, and the different types of access nodes access the computing node to be accessed in different access modes, so that the effective access of the computing node to be accessed is ensured, invalid network connection is avoided, and the overall access and computing efficiency of the Internet of things are improved.

For the electronic device and the machine-readable storage medium embodiments, the description is relatively simple, and reference should be made to part of the description of the method embodiments for the relevant matters, since the method content involved is basically similar to the method embodiments described above.

It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

In this specification, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for apparatus, electronic devices, and machine-readable storage medium embodiments, the description is relatively simple as it is substantially similar to method embodiments, with reference to the section descriptions of method embodiments being merely illustrative.

The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A method of computing node access, the method comprising:

acquiring a first attribute feature group of a computing node to be accessed, a second attribute feature group of each access node accessed to the Internet of things and a third attribute feature group of each computing node accessed to the Internet of things, wherein the attribute feature groups comprise a plurality of attribute features used for representing attributes of data which are required to be computed by the computing node or attributes of the data which are required to be accessed by the access node;

determining the node type of the access node with the attribute characteristics according to the attribute characteristics and the third attribute characteristic groups;

and accessing the computing node to be accessed to the Internet of things according to a preset access mode corresponding to the node type.

2. The method of claim 1, wherein determining the node type of the access node having the attribute feature based on the attribute feature and each third attribute feature set comprises:

judging whether the attribute features are contained in at least one third attribute feature group according to the attribute features and the third attribute feature groups;

the step of accessing the to-be-accessed computing node to the internet of things according to the preset access mode corresponding to the node type comprises the following steps:

the computing node to be accessed is accessed to each access node in the outlier node set, and the updated computing capacity of the computing node to be accessed after the computing node to be accessed is accessed to each access node is obtained;

3. The method of claim 2, wherein said accessing the computing node to be accessed to each access node in the set of outlier nodes comprises:

calculating the distance parameters between each access node in the outlier node set and the to-be-accessed computing node;

and sequentially accessing the to-be-accessed computing nodes into each access node according to descending order of each distance parameter.

4. The method of claim 1, wherein determining the node type of the access node having the attribute feature based on the attribute feature and each third attribute feature set comprises:

if not, determining the node type of the access node with the attribute characteristics as an unsatisfactory node type;

obtaining forwarding efficiency parameters of each access node and the to-be-accessed computing node in the dissatisfied node set;

according to the forwarding efficiency parameters of the access nodes and the to-be-accessed computing nodes, the to-be-accessed computing nodes are accessed to the access nodes meeting the preset efficiency improvement condition in the dissatisfied node set, and the updated computing capacity of the to-be-accessed computing nodes after being accessed to the access nodes is obtained;

5. The method of claim 4, wherein after said determining whether said output to input efficiency ratio reaches a preset efficiency ratio threshold, said method further comprises:

obtaining connection parameters of each access node and the to-be-accessed computing node in the satisfied node set;

according to the connection parameters of the access nodes and the to-be-accessed computing nodes, the to-be-accessed computing nodes are accessed to the access nodes meeting the preset connection simplification conditions in the satisfied node set, and the updated computing capacity of the to-be-accessed computing nodes after being accessed to the access nodes is obtained;

after the computing nodes to be accessed are accessed, acquiring state information of each computing node which is accessed to each access node in the satisfied node set;

judging whether the state information of at least one computing node in the computing nodes changes or not;

if yes, acquiring a third attribute feature group of the computing node with unchanged state information and a third attribute feature group of the computing node with changed state information, taking the third attribute feature group of the computing node with changed state information as a first attribute feature group of the computing node to be accessed, taking the third attribute feature group of the computing node with unchanged state information as a third attribute feature group of each computing node accessed to the Internet of things, returning to execute the second attribute feature groups, and searching attribute features contained in the first attribute feature group.

6. The method according to claim 5, wherein after taking the third attribute feature set of the computing node whose state information has changed as the first attribute feature set of the computing node to be accessed, and taking the third attribute feature set of the computing node whose state information has not changed as the third attribute feature set of each computing node that has accessed the internet of things, the method further comprises:

attenuating the refresh intensity according to a preset attenuation strategy;

and returning to execute the searching of the attribute features contained in the first attribute feature group from the second attribute feature groups, wherein the searching comprises the following steps:

and if the refresh intensity is not smaller than the preset intensity threshold, returning to execute the attribute features contained in the first attribute feature group from the second attribute feature groups.

7. A computing node access apparatus, the apparatus comprising:

the system comprises an acquisition module, a storage module and a storage module, wherein the acquisition module is used for acquiring a first attribute feature group of a computing node to be accessed, a second attribute feature group of each access node accessed to the Internet of things and a third attribute feature group of each computing node accessed to the Internet of things, the attribute feature groups comprise a plurality of attribute features, and the attribute features are used for representing the attribute of data which the computing node needs to calculate or the attribute of the data which the access node needs to access;

The searching module is used for searching the attribute characteristics contained in the first attribute characteristic group from each second attribute characteristic group;

the determining module is used for determining the node type of the access node with the attribute characteristics according to the attribute characteristics and the third attribute characteristic groups;

8. The apparatus of claim 7, wherein the determining module is specifically configured to:

the access module is specifically configured to:

9. The apparatus according to claim 8, wherein the access module, when configured to access each access node in the set of outlier nodes to the computing node to be accessed, is specifically configured to:

10. The apparatus of claim 7, wherein the determining module is specifically configured to:

The access module is specifically configured to:

11. The apparatus of claim 10, wherein the determining module is further configured to:

if the output-input efficiency ratio reaches the preset efficiency ratio threshold, determining that the node type of the access node with the attribute characteristics is a satisfactory node type;

the access module is specifically configured to:

12. The apparatus of claim 11, wherein the access module is further configured to:

attenuating the refresh intensity according to a preset attenuation strategy;

the access module is specifically configured to, when being configured to perform the returning to perform the searching for the attribute features included in the first attribute feature set from the second attribute feature sets:

13. An electronic device comprising a processor and a memory, wherein,

the memory is used for storing a computer program;

the processor being adapted to implement the method of any of claims 1-6 when executing a computer program stored on the memory.

14. A machine-readable storage medium, characterized in that it has stored therein a computer program which, when executed by a processor, implements the method of any of claims 1-6.