CN119814786A - Edge node access method, device, equipment and medium - Google Patents
Edge node access method, device, equipment and medium Download PDFInfo
- Publication number
- CN119814786A CN119814786A CN202412000434.8A CN202412000434A CN119814786A CN 119814786 A CN119814786 A CN 119814786A CN 202412000434 A CN202412000434 A CN 202412000434A CN 119814786 A CN119814786 A CN 119814786A
- Authority
- CN
- China
- Prior art keywords
- edge node
- proxy
- hash value
- component
- proxy component
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Computer And Data Communications (AREA)
Abstract
本发明实施例提供了一种边缘节点接入方法、装置、设备及介质，包括：预先设置若干个代理组件；当接收若干个边缘节点的名称时，根据每个边缘节点的名称进行哈希值计算，得到第一哈希值；根据第一哈希值和预先获取的代理组件对应的第二哈希值向边缘节点分配对应的代理组件；通过代理组件接收对应的边缘节点的接入请求；针对接入请求接入边缘节点。本发明实施例中，通过哈希值计算和代理组件的分配，实现了边缘节点与代理组件的高效匹配和负载均衡，确保了边缘节点接入请求的快速响应和处理，只计算边缘节点名称哈希值，因此，资源消耗不会随着边缘节点增加而线性增加，通过设置代理组件从而支持边缘节点的动态接入。
The embodiment of the present invention provides an edge node access method, device, equipment and medium, including: pre-setting several proxy components; when receiving the names of several edge nodes, performing hash value calculation according to the name of each edge node to obtain a first hash value; allocating the corresponding proxy component to the edge node according to the first hash value and the second hash value corresponding to the pre-acquired proxy component; receiving the access request of the corresponding edge node through the proxy component; accessing the edge node in response to the access request. In the embodiment of the present invention, through the hash value calculation and the allocation of the proxy component, efficient matching and load balancing of the edge node and the proxy component are achieved, ensuring the rapid response and processing of the edge node access request, and only calculating the hash value of the edge node name, so that resource consumption will not increase linearly with the increase of edge nodes, and the dynamic access of the edge node is supported by setting the proxy component.
Description
Technical Field
The present invention relates to the field of cache technologies, and in particular, to an edge node access method, device, equipment, and medium.
Background
With the continued advancement of internet of things (IoT) technology, more and more devices, such as intelligent transportation systems, remote monitoring devices, etc., are deployed at the edge of the network for collecting and processing data. Compared with the nodes in the traditional data center, the edge nodes have the characteristics of distributed positions, complex network environment, poor connection stability and the like.
In order to effectively manage these edge nodes, container technology and its orchestration tools, particularly Kubernetes, have become industry standards. Kubernetes is an open-source container orchestration engine whose main functions include automating the deployment of container applications, expanding applications, and managing the lifecycle of applications. However, the specificity of the edge computing environment, such as the instability of the network, makes the traditional data center node management method difficult to deal with, in the related art, load balancers are deployed in the cloud, and load distribution is performed by algorithms such as random, load, connection number, and the like, however, because the network connection between the edge node and the cloud may be frequently disconnected, the load balancers are difficult to continuously and effectively work, resulting in the problems of uneven load distribution, delay of requests, failure of access, and the like.
Disclosure of Invention
The embodiment of the invention aims to provide an edge node access method, device, equipment and medium, and the specific technical scheme is as follows:
In a first aspect of the present invention, there is provided an edge node access method, the method including:
Presetting a plurality of agent components;
When names of a plurality of edge nodes are received, carrying out hash value calculation according to the names of the edge nodes to obtain a first hash value;
distributing corresponding proxy components to the edge nodes according to the first hash values and the second hash values corresponding to the proxy components, which are obtained in advance;
Receiving an access request of the corresponding edge node through the proxy components, wherein each proxy component corresponds to at least one edge node;
and accessing the edge node for the access request.
Optionally, after the step of accessing the edge node for the access request, the method comprises:
responding to a restarting instruction or a preset edge area network disconnection recovery event, and determining a re-access time interval according to the pre-acquired access time of each edge node and the corresponding current time of each edge node;
And receiving an access request of the edge node according to the re-access time interval.
Optionally, after the step of accessing the edge node for the access request, the method comprises:
after sending a target message to the edge node, persistence is carried out on the target message to generate a persistence rollback version;
said persisting the target message comprising:
Storing the target message into a preset shared memory, wherein the shared memory is used for storing the target message sent to each edge node by each agent component;
after the step of persisting the target message and generating a persisted rollback version, the method further comprises:
when any agent component corresponding to the edge node is in a fault state, the replacement agent component corresponding to the edge node is redetermined;
comparing, by the replacement proxy component, the latest persistent rollback version corresponding to the current persistent rollback version and the cluster;
if the first version number corresponding to the current lasting rollback version is consistent with the second version number corresponding to the latest lasting rollback version corresponding to the cluster, the resource information corresponding to the current lasting rollback version is sent to the edge node;
and if the first version number corresponding to the current lasting rollback version is inconsistent with the second version number corresponding to the latest lasting rollback version corresponding to the cluster, sending the resource information corresponding to the latest lasting rollback version to the edge node.
Optionally, the assigning the corresponding proxy component to the edge node according to the first hash value and the second hash value corresponding to the previously acquired proxy component includes:
Obtaining a first address and port information corresponding to each proxy component and performing hash value calculation to obtain a second hash value corresponding to each proxy component;
And searching a proxy component corresponding to the second hash value which is the same as the first hash value, and taking the proxy component as the proxy component of the edge node.
Optionally, after the step of obtaining the first address and the port information corresponding to each proxy component and performing hash value calculation to obtain the second hash value corresponding to each proxy component, the method includes:
and if the second hash value which is the same as the first hash value does not exist, determining a first-appearing proxy component in all the proxy components according to a preset sequence, and taking the first-appearing proxy component as the proxy component of the edge node.
Optionally, after the step of accessing the edge node for the access request, the method comprises:
When the number of the edge nodes accessed by each proxy component in any cluster is detected to be larger than a first preset threshold value, the corresponding number of the proxy components is increased;
when the resource utilization rate of each proxy component in any one of the clusters is detected to be larger than a second preset threshold value, increasing the corresponding quantity of the proxy components;
when detecting that a timeout error log exists in a pre-acquired server log, increasing the number corresponding to the cluster;
And when detecting that the resource utilization rate in any one of the clusters is greater than a third preset threshold value, increasing the number corresponding to the clusters.
Optionally, after the step of accessing the edge node for the access request, the method comprises:
And when detecting that the resource utilization rate of all the proxy components in any cluster is smaller than a fourth preset threshold, reducing the corresponding quantity of the proxy components, wherein the cluster is used for managing the edge nodes.
In a second aspect of the present invention, there is also provided an edge node access device, the device comprising:
The setting module is used for presetting a plurality of agent components;
The computing module is used for computing hash values of the names of the edge nodes when the names of the edge nodes are received, so that a first hash value is obtained;
The distribution module is used for distributing the corresponding proxy component to the edge node according to the first hash value and the second hash value which is obtained in advance and corresponds to the proxy component;
the receiving module is used for receiving the access request of the corresponding edge node through the proxy components, and each proxy component corresponds to at least one edge node;
And the access module is used for accessing the edge node aiming at the access request.
In a third aspect of the present invention, there is also provided a communication device comprising a transceiver, a memory, a processor and a program stored on the memory and executable on the processor;
The processor is configured to read a program in a memory to implement the edge node access method according to any one of the first aspect.
In a fourth aspect of the implementation of the present invention, there is also provided a readable storage medium, which when executed by a processor, implements an edge node access method according to any of the first aspects.
The edge node access method provided by the embodiment of the invention comprises the steps of presetting a plurality of proxy components, calculating a hash value of each name of the edge node to obtain a first hash value when the names of the plurality of edge nodes are received, distributing corresponding proxy components to the edge node according to the first hash value and a second hash value corresponding to the pre-obtained proxy components, receiving an access request of the corresponding edge node through the proxy components, wherein each proxy component corresponds to at least one edge node, and accessing the edge node according to the access request. In the embodiment of the invention, the efficient matching and load balancing of the edge node and the proxy component are realized through the hash value calculation and the proxy component distribution, the quick response and processing of the edge node access request are ensured, and only the edge node name hash value is calculated, so that the resource consumption cannot be linearly increased along with the increase of the edge node, the proxy component is arranged to support the dynamic access of the edge node, the capacity expansion and the capacity contraction can be carried out on the proxy component subsequently, and the high performance and the stability of the access service can be ensured.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below.
Fig. 1 is a flowchart of steps of an edge node access method according to an embodiment of the present invention;
fig. 2 is a device block diagram of an edge node access device according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a communication device according to an embodiment of the present invention;
Fig. 4 is a schematic diagram of an exemplary edge node access architecture provided by an embodiment of the present invention;
fig. 5 is a schematic diagram of an exemplary decision algorithm provided by an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the following detailed description of the embodiments of the present invention will be given with reference to the accompanying drawings. However, those of ordinary skill in the art will understand that in various embodiments of the present invention, numerous technical details have been set forth in order to provide a better understanding of the present invention. The claimed invention may be practiced without these specific details and with various changes and modifications based on the following embodiments. The following embodiments are divided for convenience of description, and should not be construed as limiting the specific implementation of the present invention, and the embodiments can be mutually combined and referred to without contradiction.
It should be noted that, in the embodiment of the present invention, the overall application scenario architecture, referring to fig. 4, is an edge node access method architecture, specifically, includes two parts, namely a cloud cluster and an edge node, where in the embodiment of the present invention, in the case that the number of edge nodes is massive, the edge nodes are managed to implement dynamic access of the edge nodes and the cloud.
The cloud end can be provided with at least one k8s cluster, each k8s cluster can comprise a cluster management component, an agent component, a decision component and a detection component, and a multi-cluster management component is deployed on the cloud end to manage and schedule a plurality of clusters.
Specifically, the multi-cluster management component is used for uniformly managing a plurality of k8s clusters and displaying all edge nodes, and can select to distribute the container application to a specific certain edge node.
The cluster management component is characterized in that the unified management of the edge nodes is realized by adopting a K8S-based edge computing framework, a K8S cluster is composed of management nodes and computing nodes, the deployment of the component nodes is called as management nodes, the edge nodes are computing nodes in the K8S cluster, the management component only needs to adopt K8S original components, and three copies are adopted in specific deployment, so that high availability is realized.
The process of accessing the proxy component, namely the original K8S or KubeEdge/OpenYurt computing node to the cluster is that all edge nodes register with the management component, and as each management component caches metadata of all edge nodes and container applications on the nodes, the metadata cannot be linearly expanded, and the architecture has performance bottlenecks when facing to large-scale edge nodes. Thus, horizontally extensible proxy components are designed and deployed in the cloud, through which edge nodes (compute nodes) access the K8S cluster.
And the decision component is designed and deployed at the cloud, and before the edge node accesses the proxy component, the edge node firstly inquires the decision component which proxy component is accessed into the cluster through, and accesses the cluster according to the returned proxy component access address.
The cloud end designs a detection module for detecting whether the current cloud end proxy component has a performance bottleneck, and if so, expansion is required, or capacity reduction is required along with the reduction of edge nodes.
That is, referring to fig. 4, it can be seen that by means of multi-stage distributed edge node access, dynamic access and management of mass edge nodes are achieved, and by means of horizontally expandable proxy components accessing edge nodes, automatic dynamic expansion can be achieved according to the number of edge nodes and access loads, and uniformity and high availability of edge node access are guaranteed by adopting a consistent hash algorithm.
Referring to fig. 1, a step flow chart of an edge node access method provided by an embodiment of the present invention is shown, where the method may include:
step 101, presetting a plurality of agent components;
102, when names of a plurality of edge nodes are received, carrying out hash value calculation according to the names of each edge node to obtain a first hash value;
It should be noted that, in the embodiment of the present invention, a decision component is designed and deployed at the cloud end, and before an edge node accesses an agent component, it firstly inquires the decision component which agent component accesses a cluster through, and accesses the cluster according to the returned access address of the agent component.
Therefore, first, the proxy component needs to be allocated to the edge node to be accessed, where the edge node performs hash value calculation according to the node name to obtain a first hash value, for example, as shown in fig. 2, where the hash value calculated by the edge node 1 is 80.
Step 103, distributing corresponding proxy components to the edge node according to the first hash value and a second hash value corresponding to the proxy components, which is obtained in advance;
Further, the assigning the corresponding proxy component to the edge node according to the first hash value and the second hash value corresponding to the previously acquired proxy component includes:
Obtaining a first address and port information corresponding to each proxy component and performing hash value calculation to obtain a second hash value corresponding to each proxy component;
And searching a proxy component corresponding to the second hash value which is the same as the first hash value, and taking the proxy component as the proxy component of the edge node.
Further, after the step of obtaining the first address and the port information corresponding to each proxy component to perform hash value calculation to obtain the second hash value corresponding to each proxy component, the method includes:
and if the second hash value which is the same as the first hash value does not exist, determining a first-appearing proxy component in all the proxy components according to a preset sequence, and taking the first-appearing proxy component as the proxy component of the edge node.
It should be noted that, in the embodiment of the present invention, the hash value calculation is performed by obtaining the first address and the Port information corresponding to each proxy component to obtain the second hash value, specifically, the hash value calculation is performed by combining ip+ports of the proxy components to obtain a second hash value, for example, a value from 0 to 1023 is obtained, the edge node hashes according to the node name and maps to a range from 0 to 1023, at this time, the decision component returns the proxy component equal to the hash value of the edge node name, if the first proxy component cannot be found, for example, as shown in fig. 5, the hash value is calculated by the edge node 1 to be 80, and if the proxy component equal to 80 cannot be found, the first proxy component, that is, the proxy component 2 with the hash value of 500 in the figure, can be found clockwise.
It should be noted that, in the process of computing the hash, when the edge node is accessed, the edge node name needs to be provided, and a decision component uniformly generates a random character string, for example edgenode-xxx, where xxx is a randomly generated character string, and by converting the edge node name into the random character string, the uniformity of the edge node falling into the consistent hash agent component instance can be improved.
Step 104, receiving an access request of the corresponding edge node through the proxy components, wherein each proxy component corresponds to at least one edge node;
It should be noted that, in the embodiment of the present invention, after determining a proxy component, an access request of the corresponding edge node is received through the proxy component.
Specifically, after a certain proxy component is selected, the proxy component is called to generate an edge node access command, and the token of the decision component is added to return to the edge node.
The token of the decision-making component can provide a unified token authentication management center in the decision-making component, so that the token is provided when an access address is returned to the return access node, and when the proxy component receives an edge node access request, the token provided by the edge node can be used for verifying validity in the token authentication management center of the decision-making component.
Step 105, accessing the edge node for the access request.
The proxy component is responsible for receiving an access request of an edge node and then sending the request to the cluster management component for registration of the edge node, while the proxy component records information of the edge node, including name, IP, session, etc.
When the management component needs to send an instruction to a certain edge node, the proxy component monitors the instruction and forwards the instruction to the corresponding edge node through the session.
It should be noted that, in the embodiment of the present invention, each proxy component is only responsible for part of the edge node access and management, so the metadata maintained by each proxy component is also within a relatively predictable value, so the consumption resources of a single proxy component will not increase with the increase of the number of the whole edge nodes, and thus horizontal expansion can be performed. A certain proxy component can only belong to one K8S cluster, a single K8S cluster can be provided with a plurality of proxy components, only one cluster and one proxy component can be deployed during initialization, and the number of the proxy components can be expanded along with the increase of edge nodes.
In addition, in the embodiment of the invention, the proxy component exposes an edge node access address on a network accessible by the edge node, wherein the address comprises information such as IP, port number, access token and the like, and the token is authentication information when accessed. Meanwhile, an interface for generating the access command needs to be provided for the decision component to call.
In one embodiment, after the step of accessing the edge node for the access request, the method comprises:
responding to a restarting instruction or a preset edge area network disconnection recovery event, and determining a re-access time interval according to the pre-acquired access time of each edge node and the corresponding current time of each edge node;
And receiving an access request of the edge node according to the re-access time interval.
It should be noted that, in the embodiment of the present invention, when the management node of the cloud is restarted (in response to a restart instruction) or the edge area network disconnection is restored (in response to a preset edge area network disconnection restoration event), all or a large number of edge nodes will be simultaneously re-connected to the cloud, so that an instant bottleneck is generated, and therefore, the proxy component uses a preset backoff algorithm to re-connect through different time window lengths.
Specifically, the preset backoff algorithm is to take the number of seconds, which is different from the current time of the edge node, of the cloud end access time by the edge node, modulo 100, and take the remainder as the interval time of re-access, so that different edge nodes can randomly re-access the cloud end within 100 seconds.
In another embodiment, after the step of accessing the edge node for the access request, the method comprises:
after sending a target message to the edge node, persistence is carried out on the target message to generate a persistence rollback version;
said persisting the target message comprising:
Storing the target message into a preset shared memory, wherein the shared memory is used for storing the target message sent to each edge node by each agent component;
it should be noted that, in the embodiment of the present invention, the latest message that the proxy component has sent to the edge node is persisted, and the sharing storage is adopted between different proxy components to realize data sharing, where the message is resource change information in K8S, such as Pod, service, etc.
After the step of persisting the target message and generating a persisted rollback version, the method further comprises:
when any agent component corresponding to the edge node is in a fault state, the replacement agent component corresponding to the edge node is redetermined;
comparing, by the replacement proxy component, the latest persistent rollback version corresponding to the current persistent rollback version and the cluster;
if the first version number corresponding to the current lasting rollback version is consistent with the second version number corresponding to the latest lasting rollback version corresponding to the cluster, the resource information corresponding to the current lasting rollback version is sent to the edge node;
and if the first version number corresponding to the current lasting rollback version is inconsistent with the second version number corresponding to the latest lasting rollback version corresponding to the cluster, sending the resource information corresponding to the latest lasting rollback version to the edge node.
It should be noted that, in the embodiment of the present invention, the detection component may detect the health status of the proxy components, and if some proxy components fail, it needs to be removed from the consistent hash ring, so as to avoid the edge node from being connected to the failed proxy component.
The specific detection method is carried out in a mode of detecting for many times, if faults are judged for many times within a set time, the faults are finally judged, and the judging mode is that an edge node access command can be obtained to serve as a health identifier.
Thus, when a certain proxy component fails, the take-over proxy component continues to send messages after reassignment by the edge node of the decision component.
Since the resource changes in the K8S correspond to the specific version number revision, when the new proxy component resends, it is only necessary to compare whether the current persisted reversion and the latest revision of the K8S are consistent.
Specifically, if the first version number corresponding to the current persistent rollback version is consistent with the second version number corresponding to the latest persistent rollback version corresponding to the cluster, the transmission can be continued.
And if the first version number corresponding to the current persistence rollback version is inconsistent with the second version number corresponding to the latest persistence rollback version corresponding to the cluster, namely, the transmitted version is older, transmitting the latest version resource information.
In addition, in the embodiment of the present invention, if only a short restart occurs, the memory cache message is lost, and it is also necessary to compare the sent and non-sent messages, instead of resending all the messages again.
In another embodiment, after the step of accessing the edge node for the access request, the method comprises:
When the number of the edge nodes accessed by each proxy component in any cluster is detected to be larger than a first preset threshold value, the corresponding number of the proxy components is increased;
when the resource utilization rate of each proxy component in any one of the clusters is detected to be larger than a second preset threshold value, increasing the corresponding quantity of the proxy components;
when detecting that a timeout error log exists in a pre-acquired server log, increasing the number corresponding to the cluster;
And when detecting that the resource utilization rate in any one of the clusters is greater than a third preset threshold value, increasing the number corresponding to the clusters.
It should be noted that, in the embodiment of the present invention, by setting the proxy component, the performance of the proxy component may be detected, and the proxy component and the cloud deployment cluster may be processed correspondingly for the current performance, for example, capacity expansion or capacity shrinkage.
Specifically, the edge node accesses the cluster through the proxy component, and then the cluster manages the accessed edge node. The capacity expansion of the access proxy component can be realized by setting the highest edge node number of proxy access, expanding the proxy component examples when most of examples in the proxy component reach the maximum number percentage, such as 80%, and expanding the proxy component examples as long as one resource reaches the bottleneck, such as the CPU exceeds 80% of the total amount, so that the capacity expansion is needed. And the two methods can be combined, and when the number of the accessed edge nodes exceeds a certain value, the performance bottleneck detection of the resource pool is performed, so that the resource consumption caused by the utilization rate of the resource performance is avoided when the number of the nodes is too small.
The management components of a single cluster are mainly etcd and apiserver, which have the bottleneck of the upper limit of performance, and the constraint conditions are CPU, memory and network bandwidth, so that the agent component can not be infinitely expanded to realize mass edge node access on the premise of the single cluster, a plurality of clusters need to be horizontally expanded, and the agent component is added in the newly added clusters. The detection method comprises the steps of firstly detecting apiserver logs at fixed time and adding the cluster when finding out the overtime error logs, and secondly detecting whether the CPU, the memory and the network bandwidth performance of the cluster reach the bottleneck, if the utilization rate exceeds the percentage, for example 80%. The invention aims to manage massive edge nodes not through massive clusters, and further expands the scale of the edge nodes managed by a single cluster, so that the number of the clusters is controlled to be smaller in order of magnitude, and firstly, resources are saved, and secondly, the complexity of management is reduced.
In addition, in the embodiment of the invention, after the performance bottleneck is detected, a plurality of execution strategies are supported, namely 1) the most conservative strategy, namely alarming, prompting the capacity expansion agent component or the capacity expansion cluster, and manually expanding capacity, 2) the medium strategy, namely informing the decision-making module to reject access of the edge node and guaranteeing the stability of the existing system, and 3) the automatic strategy, namely triggering an automatic expansion mechanism and needing to access the automatic expansion system.
In another embodiment, after the step of accessing the edge node for the access request, the method comprises:
And when detecting that the resource utilization rate of all the proxy components in any cluster is smaller than a fourth preset threshold, reducing the corresponding quantity of the proxy components, wherein the cluster is used for managing the edge nodes.
It should be noted that, in the embodiment of the present invention, besides the capacity expansion described above, it may also be determined whether the capacity can be reduced based on the edge node and the proxy component resource utilization, so as to dynamically reduce the overall system loss.
Specifically, the capacity shrinking time is that the CPU, the memory and the network bandwidth of each proxy component are polled, and if each index is below a certain threshold value, such as 30%, the proxy components are reduced one by one, so that the impact of simultaneous access of a large number of edge nodes caused by the proxy components is avoided. Until a certain threshold, e.g., 60%, is reached, the agent component stops being reduced. Since container application cross-cluster migration occurs in a reduced-capacity cluster, and the service continuity of edge applications is affected, cluster reduction is only performed when no edge node exists in a certain cluster.
In addition, in the embodiment of the invention, after the edge nodes are reduced, the proxy components can be correspondingly reduced, the proxy components can be directly removed from the hash ring, the corresponding component service is closed, the edge nodes originally proxy for the reduced components can be re-accessed, and new nodes are allocated according to a new consistent hash algorithm during access.
In addition, referring to fig. 4, it can be seen that, in the embodiment of the present invention, the scale of the single cluster management edge node is further enlarged, and simultaneously, ultra-large scale unified management of the edge node can be realized, and meanwhile, through the elastic access of automatic sensing of the load, the dynamic access and push-out of the edge node are satisfied, so that the complexity of operation and management is reduced as a whole, and the resource consumption level is also reduced. The performance of the system is monitored in real time through the performance detection component, and the self-adaptive expansion of the system is realized by combining an automatic expansion mechanism, so that the stability and the response speed of the system are ensured.
The edge node access method provided by the embodiment of the invention comprises the steps of presetting a plurality of proxy components, calculating a hash value according to the name of each edge node to obtain a first hash value when the names of the plurality of edge nodes are received, distributing corresponding proxy components to the edge nodes according to the first hash value and a second hash value corresponding to the proxy components, receiving access requests of the corresponding edge nodes through the proxy components, wherein each proxy component corresponds to at least one edge node, and accessing the edge nodes according to the access requests. According to the embodiment of the invention, through the calculation of the hash value and the distribution of the proxy component, the efficient matching and the load balancing of the edge node and the cloud proxy component are realized, the quick response and the processing of the access request of the edge node are ensured, and the cloud only calculates the hash value of the name of the edge node, so that the resource consumption cannot be linearly increased along with the increase of the edge node, the proxy component is arranged to support the dynamic access of the edge node, the capacity expansion and the capacity contraction can be carried out on the proxy component subsequently, and the high performance and the stability of the cloud access service can be ensured.
Referring to fig. 2, a schematic structural diagram of an edge node access device according to an embodiment of the present invention is shown, where the device includes:
a setting module 201, configured to preset a plurality of proxy components;
A calculation module 202, configured to, when receiving names of a plurality of edge nodes, perform hash value calculation on the name of each edge node to obtain a first hash value;
an allocation module 203, configured to allocate a corresponding proxy component to the edge node according to the first hash value and a second hash value that is obtained in advance and corresponds to the proxy component;
a receiving module 204, configured to receive, by the proxy components, an access request of a corresponding edge node, where each proxy component corresponds to at least one edge node;
an access module 205, configured to access the edge node for the access request.
The edge node access device provided by the embodiment of the invention is used for carrying out hash value calculation on the name of each edge node to obtain a first hash value when receiving the names of a plurality of edge nodes by presetting a plurality of agent components, distributing corresponding agent components to the edge nodes according to the first hash value and a second hash value corresponding to the agent components, receiving access requests of the corresponding edge nodes by the agent components, wherein each agent component corresponds to at least one edge node, and accessing the edge nodes according to the access requests. In the embodiment of the invention, the efficient matching and load balancing of the edge node and the proxy component are realized through the hash value calculation and the proxy component distribution, the quick response and processing of the edge node access request are ensured, and only the edge node name hash value is calculated, so that the resource consumption cannot be linearly increased along with the increase of the edge node, the proxy component is arranged to support the dynamic access of the edge node, the capacity expansion and the capacity contraction can be carried out on the proxy component subsequently, and the high performance and the stability of the access service can be ensured.
The embodiment of the invention also provides a communication device, as shown in fig. 3, comprising a processor 301, a communication interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communication interface 302 and the memory 303 complete communication with each other through the communication bus 304,
A memory 303 for storing a computer program;
The processor 301, when executing the program stored in the memory 303, may implement the following steps:
Presetting a plurality of agent components;
When names of a plurality of edge nodes are received, carrying out hash value calculation according to the names of the edge nodes to obtain a first hash value;
distributing corresponding proxy components to the edge nodes according to the first hash values and the second hash values corresponding to the proxy components, which are obtained in advance;
Receiving an access request of the corresponding edge node through the proxy components, wherein each proxy component corresponds to at least one edge node;
and accessing the edge node for the access request.
Where the memory and the processor are connected by a bus, the bus may comprise any number of interconnected buses and bridges, the buses connecting the various circuits of the one or more processors and the memory together. The bus may also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver may be one element or may be a plurality of elements, such as a plurality of receivers and transmitters, providing a means for communicating with various other apparatus over a transmission medium. The data processed by the processor may be transmitted over a wired medium or through an antenna on a wireless medium, and the antenna further receives and transmits data to the processor. The processor is responsible for managing the bus and general processing and may also provide various functions including timing, peripheral interfaces, voltage regulation, power management, and other control functions. And memory may be used to store data used by the processor in performing operations.
The communication bus mentioned by the above terminal may be a peripheral component interconnect standard (PERIPHERAL COMPONENT INTERCONNECT, abbreviated as PCI) bus or an extended industry standard architecture (Extended Industry Standard Architecture, abbreviated as EISA) bus, etc. The communication bus may be classified as an address bus, a data bus, a control bus, or the like. For ease of illustration, the figures are shown with only one bold line, but not with only one bus or one type of bus.
The communication interface is used for communication between the terminal and other devices.
The memory may include random access memory (Random Access Memory, RAM) or may include non-volatile memory (non-volatile memory), such as at least one disk memory. Optionally, the memory may also be at least one memory device located remotely from the aforementioned processor.
The processor may be a general-purpose processor, including a central Processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), a digital signal processor (DIGITAL SIGNAL Processing, DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable gate array (FPGA) or other Programmable logic device, discrete gate or transistor logic device, or discrete hardware components.
In yet another embodiment of the present invention, a computer readable storage medium is provided, in which instructions are stored, which when run on a computer, cause the computer to perform the edge node access method according to any of the above embodiments.
In a further embodiment of the present invention, a computer program product comprising instructions which, when run on a computer, causes the computer to perform the edge node access method according to any of the above embodiments is also provided.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present invention, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another, for example, by wired (e.g., coaxial cable, optical fiber, digital Subscriber Line (DSL)), or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk Solid STATE DISK (SSD)), etc.
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 an 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 system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.
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 (10)
1. An edge node access method, the method comprising:
Presetting a plurality of agent components;
When names of a plurality of edge nodes are received, carrying out hash value calculation according to the names of the edge nodes to obtain a first hash value;
distributing corresponding proxy components to the edge nodes according to the first hash values and the second hash values corresponding to the proxy components, which are obtained in advance;
Receiving an access request of the corresponding edge node through the proxy components, wherein each proxy component corresponds to at least one edge node;
and accessing the edge node for the access request.
2. The method according to claim 1, characterized in that after the step of accessing the edge node for the access request, the method comprises:
responding to a restarting instruction or a preset edge area network disconnection recovery event, and determining a re-access time interval according to the pre-acquired access time of each edge node and the corresponding current time of each edge node;
And receiving an access request of the edge node according to the re-access time interval.
3. The method according to claim 1, characterized in that after the step of accessing the edge node for the access request, the method comprises:
after sending a target message to the edge node, persistence is carried out on the target message to generate a persistence rollback version;
said persisting the target message comprising:
Storing the target message into a preset shared memory, wherein the shared memory is used for storing the target message sent to each edge node by each agent component;
after the step of persisting the target message and generating a persisted rollback version, the method further comprises:
when any agent component corresponding to the edge node is in a fault state, the replacement agent component corresponding to the edge node is redetermined;
comparing, by the replacement proxy component, the latest persistent rollback version corresponding to the current persistent rollback version and the cluster;
if the first version number corresponding to the current lasting rollback version is consistent with the second version number corresponding to the latest lasting rollback version corresponding to the cluster, the resource information corresponding to the current lasting rollback version is sent to the edge node;
and if the first version number corresponding to the current lasting rollback version is inconsistent with the second version number corresponding to the latest lasting rollback version corresponding to the cluster, sending the resource information corresponding to the latest lasting rollback version to the edge node.
4. The method of claim 1, wherein the assigning the corresponding proxy component to the edge node based on the first hash value and a pre-obtained second hash value corresponding to the proxy component comprises:
Obtaining a first address and port information corresponding to each proxy component and performing hash value calculation to obtain a second hash value corresponding to each proxy component;
And searching a proxy component corresponding to the second hash value which is the same as the first hash value, and taking the proxy component as the proxy component of the edge node.
5. The method according to claim 1, wherein after the step of obtaining the first address and port information corresponding to each proxy component and performing hash value calculation to obtain the second hash value corresponding to each proxy component, the method includes:
and if the second hash value which is the same as the first hash value does not exist, determining a first-appearing proxy component in all the proxy components according to a preset sequence, and taking the first-appearing proxy component as the proxy component of the edge node.
6. The method according to claim 1, characterized in that after the step of accessing the edge node for the access request, the method comprises:
When the number of the edge nodes accessed by each proxy component in any cluster is detected to be larger than a first preset threshold value, the corresponding number of the proxy components is increased;
when the resource utilization rate of each proxy component in any one of the clusters is detected to be larger than a second preset threshold value, increasing the corresponding quantity of the proxy components;
when detecting that a timeout error log exists in a pre-acquired server log, increasing the number corresponding to the cluster;
And when detecting that the resource utilization rate in any one of the clusters is greater than a third preset threshold value, increasing the number corresponding to the clusters.
7. The method according to claim 1, characterized in that after the step of accessing the edge node for the access request, the method comprises:
And when detecting that the resource utilization rate of all the proxy components in any cluster is smaller than a fourth preset threshold, reducing the corresponding quantity of the proxy components, wherein the cluster is used for managing the edge nodes.
8. An edge node access device, the device comprising:
The setting module is used for presetting a plurality of agent components;
The computing module is used for computing hash values according to the names of the edge nodes when the names of the edge nodes are received, so as to obtain a first hash value;
The distribution module is used for distributing the corresponding proxy component to the edge node according to the first hash value and the second hash value which is obtained in advance and corresponds to the proxy component;
the receiving module is used for receiving the access request of the corresponding edge node through the proxy components, and each proxy component corresponds to at least one edge node;
And the access module is used for accessing the edge node aiming at the access request.
9. A communication device comprising a transceiver, a memory, a processor, and a program stored on the memory and executable on the processor;
The processor being configured to read a program in a memory to implement the edge node access method according to any of claims 1-7.
10. A readable storage medium storing a program, wherein the program when executed by a processor implements an edge node access method according to any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202412000434.8A CN119814786B (en) | 2024-12-31 | 2024-12-31 | Edge node access method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202412000434.8A CN119814786B (en) | 2024-12-31 | 2024-12-31 | Edge node access method, device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119814786A true CN119814786A (en) | 2025-04-11 |
| CN119814786B CN119814786B (en) | 2025-12-09 |
Family
ID=95269083
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202412000434.8A Active CN119814786B (en) | 2024-12-31 | 2024-12-31 | Edge node access method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119814786B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120675991A (en) * | 2025-08-15 | 2025-09-19 | 济南浪潮数据技术有限公司 | Method, device, equipment and medium for detecting node states in k8s cluster |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040088412A1 (en) * | 2002-07-24 | 2004-05-06 | Ranjit John | System and method for highly-scalable real-time and time-based data delivery using server clusters |
| US20090234845A1 (en) * | 2006-02-22 | 2009-09-17 | Desantis Raffaele | Lawful access; stored data handover enhanced architecture |
| CN103441906A (en) * | 2013-09-25 | 2013-12-11 | 哈尔滨工业大学 | System for detecting abnormity of proxy cache cluster based on automatic computing |
| CN114553604A (en) * | 2022-04-26 | 2022-05-27 | 南京邮电大学 | Internet of things terminal node access control method |
| CN114625527A (en) * | 2021-10-11 | 2022-06-14 | 亚信科技（中国）有限公司 | A method, apparatus, device and storage medium for processing connection allocation |
| CN114844865A (en) * | 2021-01-30 | 2022-08-02 | 华为技术有限公司 | Service identifier distribution method and communication device in computational power awareness network |
| US20220329608A1 (en) * | 2021-04-02 | 2022-10-13 | Sift Science, Inc. | Systems and methods for intelligently constructing a backbone network graph and identifying and mitigating digital threats based thereon in a machine learning task-oriented digital threat mitigation platform |
| CN119172250A (en) * | 2024-10-21 | 2024-12-20 | 上海哔哩哔哩科技有限公司 | Cluster node scheduling method and device |
-
2024
- 2024-12-31 CN CN202412000434.8A patent/CN119814786B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040088412A1 (en) * | 2002-07-24 | 2004-05-06 | Ranjit John | System and method for highly-scalable real-time and time-based data delivery using server clusters |
| US20090234845A1 (en) * | 2006-02-22 | 2009-09-17 | Desantis Raffaele | Lawful access; stored data handover enhanced architecture |
| CN103441906A (en) * | 2013-09-25 | 2013-12-11 | 哈尔滨工业大学 | System for detecting abnormity of proxy cache cluster based on automatic computing |
| CN114844865A (en) * | 2021-01-30 | 2022-08-02 | 华为技术有限公司 | Service identifier distribution method and communication device in computational power awareness network |
| US20220329608A1 (en) * | 2021-04-02 | 2022-10-13 | Sift Science, Inc. | Systems and methods for intelligently constructing a backbone network graph and identifying and mitigating digital threats based thereon in a machine learning task-oriented digital threat mitigation platform |
| CN114625527A (en) * | 2021-10-11 | 2022-06-14 | 亚信科技（中国）有限公司 | A method, apparatus, device and storage medium for processing connection allocation |
| CN114553604A (en) * | 2022-04-26 | 2022-05-27 | 南京邮电大学 | Internet of things terminal node access control method |
| CN119172250A (en) * | 2024-10-21 | 2024-12-20 | 上海哔哩哔哩科技有限公司 | Cluster node scheduling method and device |
Non-Patent Citations (4)
| Title |
|---|
| HATIM LOKHANDWALA: "Min-O-Mee：A Proximity Based Network application leveraging the AllJoyn framework", 《2015 INTERNATIONAL CONFERENCE ON COMPUTING AND NETWORK COMMUNICATIONS》, 18 February 2016 (2016-02-18) * |
| M.SGARNUKA: "Community Detection in Social Networks based on Local Edge Centrality and Adaptive Thresholding", 《2020 4TH INTERNATIONAL CONFERENCE ON TRENDS IN ELCTRONICS AND INFROMATICS》, 17 July 2020 (2020-07-17) * |
| 林远卓: "无线接入网缓存部署策略研究", 《中国优秀硕士学位论文全文数据库》, 15 June 2022 (2022-06-15) * |
| 潘文彬: "基于无线充能无人机的边缘计算卸载方法研究", 《中国优秀硕士学位论文全文数据库》, 15 June 2024 (2024-06-15) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120675991A (en) * | 2025-08-15 | 2025-09-19 | 济南浪潮数据技术有限公司 | Method, device, equipment and medium for detecting node states in k8s cluster |
| CN120675991B (en) * | 2025-08-15 | 2026-01-27 | 济南浪潮数据技术有限公司 | Method, device, equipment and medium for detecting node states in k8s cluster |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119814786B (en) | 2025-12-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210337046A1 (en) | Intelligent scheduling method, terminal device, edge node cluster and intelligent scheduling system | |
| CN108696581B (en) | Distributed information caching method and device, computer equipment and storage medium | |
| US11922059B2 (en) | Method and device for distributed data storage | |
| CN112948120A (en) | Load balancing method, system, device and storage medium | |
| CN110830283B (en) | Fault detection method, apparatus, equipment and system | |
| CN111901421B (en) | A data processing method and related equipment | |
| US9390156B2 (en) | Distributed directory environment using clustered LDAP servers | |
| CN119814786B (en) | Edge node access method, device, equipment and medium | |
| US9760370B2 (en) | Load balancing using predictable state partitioning | |
| CN109995839A (en) | A kind of load-balancing method, system and load balancer | |
| CN106790610B (en) | A cloud system message distribution method, device and system | |
| CN114925052A (en) | Method, device, equipment and computer readable medium for restarting database | |
| CN117319507A (en) | Route connection method, device, electronic equipment and storage medium | |
| CN108366102A (en) | A kind of service discovery method, device and electronic equipment based on Consul | |
| CN109347743B (en) | Special line communication method, computer readable storage medium and terminal equipment | |
| CN110221916B (en) | Memory capacity expansion method and device, configuration center system and electronic equipment | |
| WO2019119269A1 (en) | Network fault detection method and control center device | |
| CN112631756A (en) | Distributed regulation and control method and device applied to space flight measurement and control software | |
| US20160261476A1 (en) | Message system for avoiding processing-performance decline | |
| CN116455963A (en) | Cluster node registration method, medium, device and computing equipment | |
| CN116455753B (en) | Data smoothing method and device | |
| CN116208559B (en) | Method and system for block synchronization based on load balancing | |
| CN113765686B (en) | Device management method, device, service acquisition device and storage medium | |
| CN110784479A (en) | Data verification method and device, electronic equipment and storage medium | |
| CN117424907B (en) | File acquisition method, device, node device and computer-readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |