CN116962528A - Network computing power arranging method, device, equipment and computer storage medium - Google Patents
Network computing power arranging method, device, equipment and computer storage medium Download PDFInfo
- Publication number
- CN116962528A CN116962528A CN202211520184.5A CN202211520184A CN116962528A CN 116962528 A CN116962528 A CN 116962528A CN 202211520184 A CN202211520184 A CN 202211520184A CN 116962528 A CN116962528 A CN 116962528A
- Authority
- CN
- China
- Prior art keywords
- computing power
- computing
- calculation
- calculation force
- nodes
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000004364 calculation method Methods 0.000 claims abstract description 183
- 230000008520 organization Effects 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 13
- 238000005265 energy consumption Methods 0.000 claims description 10
- 238000010586 diagram Methods 0.000 description 18
- 230000033001 locomotion Effects 0.000 description 9
- 238000004891 communication Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000011161 development Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 4
- RYAUSSKQMZRMAI-YESZJQIVSA-N (S)-fenpropimorph Chemical compound C([C@@H](C)CC=1C=CC(=CC=1)C(C)(C)C)N1C[C@H](C)O[C@H](C)C1 RYAUSSKQMZRMAI-YESZJQIVSA-N 0.000 description 3
- 230000009471 action Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000004927 fusion Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 229910052799 carbon Inorganic materials 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000002071 nanotube Substances 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 238000003032 molecular docking Methods 0.000 description 1
- 238000006386 neutralization reaction Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000031068 symbiosis, encompassing mutualism through parasitism Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W36/00—Hand-off or reselection arrangements
- H04W36/0005—Control or signalling for completing the hand-off
- H04W36/0055—Transmission or use of information for re-establishing the radio link
- H04W36/0061—Transmission or use of information for re-establishing the radio link of neighbour cell information
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a network computing power arranging method, a device, equipment and a computer storage medium, wherein scene information of user equipment accessing to a computing power network is obtained; acquiring computing power resource information required to be called by the service in progress of the user equipment according to the scene information, wherein the computing power resource information comprises computing power resource demand types and computing power key indexes; based on the calculation force load and network load information of calculation force nodes at each level in the calculation force network system, a task receiving unit which meets the calculation force resource demand type and calculation force key index is matched in the calculation force network system, the task receiving unit comprises multi-level calculation force nodes with different calculation force demand types, the multi-level calculation force nodes with different calculation force demand types are arranged to the deployment position of the service, the arranged calculation force nodes can meet the service demand type and calculation force key index, and the performance characteristic demand and the utilization rate of calculation force resources are improved.
Description
Technical Field
The application belongs to the field of network computing power arrangement, and particularly relates to an arrangement method, device and equipment of network computing cases and a computer storage medium.
Background
The computing power network is an original technical idea provided by China first, and refers to a novel information infrastructure which integrates multi-level computing power resources such as network, cloud, number, intelligence, security, edge, end, chain and the like and provides integrated services such as data sensing, transmission, storage, operation and the like by means of high-speed, mobile, safe and ubiquitous network connection.
The innovative development of the computational network is helpful for improving the efficiency and benefit of the social application of computational power and promoting the fusion innovation of information technology. With the continuous expansion of the network scale, as the computational power requirement of new application is continuously increased, the equipment scale of the computational power network requiring nano tubes and the number of users to be supported are rapidly increased, the centralized management cannot meet the requirement, and the organization arrangement mode of the computational power network is required to be adjusted.
Because the computing power has no standard and uniform layout mode, but is distributed at each position node such as the end, the side, the cloud and the like of the network, the existing computing power arranging method divides a computing power task receiving unit according to the physical address of the computing power facility, and computing power resources, network resources and other resources are distributed in the task receiving unit.
However, the existing calculation power arrangement is based on the calculation power resource arrangement mode of the physical position, so that the calculation power resource of each physical position has large difference, and the calculation power resource of a task receiving unit formed after arrangement is seriously uneven; because the computing power resources of the same physical location generally have the same computing performance characteristics, a single computing performance characteristic cannot cope with increasingly complex computing performance characteristic requirements of current and future applications; moreover, the existing arrangement method has low utilization efficiency of computational resources.
Disclosure of Invention
The embodiment of the application provides a network computing power arranging method, device, equipment and computer storage medium, which solve the problems that the computing power of a task receiving unit formed after arranging is uneven, the computing performance characteristic requirement is not met, and the computing power resource utilization rate is low.
In a first aspect, a method for arranging network computing power is provided, the method comprising:
acquiring scene information of user equipment accessing to an algorithm network;
acquiring computing power resource information required to be called by the service in progress of the user equipment according to the scene information, wherein the computing power resource information comprises computing power resource demand types and computing power key indexes;
based on the calculation force load and network load information of calculation force nodes at all levels in the calculation force network system, matching a task receiving unit meeting calculation force resource demand types and calculation force key indexes from the calculation force network system, wherein the task receiving unit comprises multi-level calculation force nodes with different calculation force demand types;
multiple levels of computing power nodes of different computing power demand types in the task receiving unit are orchestrated into a deployment location of the business.
In one possible implementation, obtaining scene information of the user equipment accessing the computing power network includes:
calculating the times of cell switching of the user equipment in unit time;
and determining the scene information corresponding to the times based on the corresponding relation between the times of cell switching and the scene information.
In one possible implementation, the computing power resource demand type includes at least one of a type of computing power node and a device type; the computing power key index comprises at least one of average time delay from the computing power node to the organization arrangement module, computing power accounting, storage accounting, energy consumption and algorithm type supported by the computing power node.
In one possible implementation, before matching the task receiving units that satisfy the computing power resource requirement type and the computing power key index from the computing power network system based on the computing power load and the network load information of each level of computing power nodes in the computing power network system, the method further includes:
acquiring characteristic data and calculation force types of all calculation force nodes in a calculation force network system, wherein the characteristic data comprises calculation force key indexes, and the calculation force types are used for matching calculation force resource demand types;
clustering the characteristic data of each computing node to obtain a characteristic data set of computing nodes of a plurality of clustered clusters;
calculating the mean value and standard deviation of the characteristic data of the computing force nodes in each cluster;
calculating a target value based on the mean and standard deviation;
and performing calculation force node connection according to the calculation force type of the calculation force nodes in each cluster and the target value of the calculation force nodes to obtain a plurality of task receiving units, wherein the task receiving units comprise calculation force nodes of which the layers are divided according to the calculation force type, and the calculation force nodes of different layers are connected through IP routing.
In a second aspect, an embodiment of the present application provides a network computing power arrangement apparatus, including:
the acquisition unit is used for acquiring scene information of the user equipment accessing the power calculation network;
the acquisition unit is also used for acquiring the computing power resource information required to be called by the service in progress of the user equipment according to the scene information, wherein the computing power resource information comprises computing power resource demand types and computing power key indexes;
the matching unit is used for matching a task receiving unit which meets the calculation power resource demand type and calculation power key indexes from the calculation power network system based on calculation power load and network load information of calculation power nodes at all levels in the calculation power network system, and the task receiving unit comprises multiple levels of calculation power nodes with different calculation power demand types;
and the arrangement unit is used for arranging the multi-level computing power nodes with different computing power demand types in the task receiving unit to the deployment position of the service.
In one possible implementation manner, the obtaining unit is configured to obtain scene information of the user equipment accessing the computing power network, and includes:
the acquisition unit is used for calculating the times of cell switching of the user equipment in unit time;
and determining the scene information corresponding to the times based on the corresponding relation between the times of cell switching and the scene information.
In one possible implementation, the computing power resource demand type includes at least one of a type of computing power node and a device type; the computing power key index comprises at least one of average time delay from the computing power node to the organization arrangement module, computing power accounting, storage accounting, energy consumption and algorithm type supported by the computing power node.
In one possible implementation, the arrangement device further includes a clustering module, a computing module, and a connecting module;
the system comprises an acquisition unit, a calculation power network system and a calculation power management unit, wherein the acquisition unit is also used for acquiring characteristic data and calculation power types of all calculation power nodes in the calculation power network system, the characteristic data comprises calculation power key indexes, and the calculation power types are used for matching calculation power resource demand types;
the clustering module is used for clustering the characteristic data of each power computing node to obtain characteristic data sets of the power computing nodes of a plurality of clustered clusters;
the computing module is used for computing the mean value and standard deviation of the characteristic data of the computing force nodes in each cluster;
the calculation module is also used for calculating a target value based on the mean value and the standard deviation;
the connection module is used for carrying out calculation force node connection according to the calculation force type of the calculation force nodes in each cluster and the target value of the calculation force nodes to obtain a plurality of task receiving units, wherein the task receiving units comprise calculation force nodes of which the levels are divided according to the calculation force type, and the calculation force nodes of different levels are connected through IP routing.
In a third aspect, an embodiment of the present application provides an electronic device, including: a processor and a memory storing computer program instructions;
the processor, when executing the computer program instructions, implements the first aspect and any possible network computing power orchestration method of the first aspect.
In a fourth aspect, embodiments of the present application provide a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the network computing power orchestration method according to the first aspect, and any one of the possible implementations of the first aspect.
The network computing power arranging method, the device, the equipment and the computer storage medium of the embodiment of the application acquire scene information of the user equipment accessing to the computing power network; acquiring computing power resource information required to be called by the service in progress of the user equipment according to the scene information, wherein the computing power resource information comprises computing power resource demand types and computing power key indexes; based on the calculation force load and network load information of calculation force nodes at each level in the calculation force network system, a task receiving unit which meets the calculation force resource demand type and calculation force key index is matched in the calculation force network system, the task receiving unit comprises multi-level calculation force nodes with different calculation force demand types, the multi-level calculation force nodes with different calculation force demand types are arranged to the deployment position of the service, the arranged calculation force nodes can meet the service demand type and calculation force key index, and the performance characteristic demand and the utilization rate of calculation force resources are improved.
Drawings
In order to more clearly illustrate the technical solution of the embodiments of the present application, the drawings that are needed to be used in the embodiments of the present application will be briefly described, and it is possible for a person skilled in the art to obtain other drawings according to these drawings without inventive effort.
FIG. 1 is a schematic diagram of a network computing orchestration system according to one embodiment of the present application;
FIG. 2 is a flow chart of a method for orchestrating network computing according to one embodiment of the present application;
FIG. 3 is a schematic diagram of a task receiving unit after network computing power arrangement according to an embodiment of the present application;
FIG. 4 is a schematic diagram of a task receiving unit according to an embodiment of the present application;
FIG. 5 is a schematic illustration of an apparatus according to one embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
Features and exemplary embodiments of various aspects of the present application will be described in detail below, and in order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described in further detail below with reference to the accompanying drawings and the detailed embodiments. It should be understood that the particular embodiments described herein are meant to be illustrative of the application only and not limiting. It will be apparent to one skilled in the art that the present application may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the application by showing examples of the application.
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 … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.
The development of the computing power network improves the efficiency and benefit of the social application of computing power and promotes the fusion innovation of information technology. The method is not only beneficial to implementing the deployment requirement of the national 'new capital construction', promotes the accurate configuration and the on-demand acquisition of the computational resources, but also promotes the extension and expansion of the eastern digital economic industry chain to the west, effectively reduces the computational energy consumption, and achieves the coordinated development of the assistance area and the national carbon reaching peak and carbon neutralization targets. The power computing network can realize the organic integration and the integrated symbiosis of the network and the power computing, and becomes the main form of the future information infrastructure. The innovation and development of the computing network quickens the intelligent transformation of the numbers in various fields of the economic society and the civil life, and injects new kinetic energy into the digital economic development.
The continuous expansion of the scale of the computing power network and the continuous increase of the computing power requirement of new application lead to the rapid increase of the equipment scale of the computing power network requiring the nano-tubes and the number of the users to be supported, the centralized management can not meet the requirement, and the organization arrangement mode of the computing power network needs to be adjusted. Because the computing power has no standard and unified layout mode, but is distributed at each position node such as the end, the side, the cloud and the like of the network, the existing computing power arrangement method for dividing computing power task receiving units according to the physical address of the computing power facility distributes computing power resources in the task receiving units, network resources and other resources, and the computing power resources are seriously uneven due to the fact that the computing power resources at each physical position have larger difference, and the task receiving units formed after arrangement; and because the computing power resources in the same physical location generally have the same computing performance characteristics, the single computing performance characteristics cannot cope with the increasingly complex computing performance characteristics requirements of the current and future applications, so that the resource utilization efficiency is lower.
Aiming at the problems of uneven computing power resources and low resource utilization of the existing computing power task receiving unit based on physical address allocation, the embodiment of the application divides a computing power network into a multi-level structure, the computing power shows a trend of increasing along with the deepening of the layers, and the computing power resources can be belonged to the cloud edge end in general. Because the computing power nodes at the cloud edge end have the characteristics, the computing power nodes at the cloud edge end are far away from the computing power nodes at the cloud edge end but have strong computing power, and the computing power nodes at the edge end are close to the computing power nodes at the edge end but have weak computing power, the computing power nodes at the cloud edge end are used for multiprocessing global tasks, and the computing power nodes at the edge end are used for multiprocessing local real-time tasks. Aiming at a multi-layer structure of the computational power network, a hierarchical clustering algorithm (Hierarchical Clustering) is utilized to identify the computational power network of the multi-layer structure based on the position, time delay and computational power of each computational power resource in the layer structure, so that the organization arrangement mode of the computational power network is adjusted.
Dividing the power computing network into a plurality of layers of task receiving units with different power computing types based on characteristic data and power computing types of each node in the power computing network, and then accessing scene information of the power computing network through the acquired user equipment; acquiring computing power resource information required to be called by the service in progress of the user equipment according to the scene information, wherein the computing power resource information comprises computing power resource demand types and computing power key indexes; based on the calculation force load and network load information of calculation force nodes at all levels in the calculation force network system, matching a task receiving unit meeting calculation force resource demand types and calculation force key indexes from the calculation force network system, wherein the task receiving unit comprises multi-level calculation force nodes with different calculation force demand types; the multi-level computing nodes with different computing power demand types in the task receiving unit are arranged to the deployment position of the service, so that the arranged computing power nodes can meet the service demand types and computing power key indexes, and the performance characteristic demands and the utilization rate of computing power resources are improved.
The following describes a network computing power arranging method, device, equipment and computer storage medium provided by the embodiment of the application with reference to the accompanying drawings.
Fig. 1 is a schematic diagram of a network computing power orchestration system according to an embodiment of the present application. As shown in fig. 1, the orchestration system may include a user interfacing module, an application interfacing module, an organization orchestration module, and a network scheduling module.
The user interface module is used for providing a network interface of a user and judging scene information of the user.
The application docking module is used for acquiring the computational power resources required to be invoked by the service in progress of the user equipment and providing the deployment positions of the computational power nodes corresponding to the computational power resources.
The organization arrangement module is used for managing the computing power network and the computing power resources, and comprises sensing and measuring of the computing power resources, dividing of the computing power resources and computing power nodes in the computing power network, and arrangement of the computing power nodes and the computing power resources in the computing power network.
Because the node computing forces of different computing force types in the computing force network have differences, such as the computing force provided by an access node (such as CPE, WIFI and other equipment) is the edge computing force, and the computing force provided by a core cloud node is the computing force of a cloud computing center. There is therefore a need to perceive and measure the computational power resources in a computational power network. The sensing and measurement of the computing power resources are realized by uniformly computing the power quantity of hardware devices (such as Field Programmable Gate Arrays (FPGA), NPU, GPU, CPU and the like) in the computing power network through the resource characteristic data of the computing power nodes, so that the uniform resource description of the physical resources of the hardware devices can be effectively provided.
The resource characteristic data of the computing power comprise node type, average response time delay, equipment type, computing power accounting, storage accounting, energy consumption, AI algorithm supporting type and the like.
Node types such as access node, edge node (e.g., MEC), edge cloud node, core cloud node, top level node (i.e., supercomputer), etc. computing node types. The average response time delay refers to the average time delay from the computing force node to the organization module. The device type includes statistics including, but not limited to, the type and number of hardware resources such as FPGA, NPU, GPU, CPU. The calculation force accounting comprises a unified calculation force network calculation force accounting mode, and single-precision calculation force (32 bits, FP 32) is adopted to measure the calculation force of calculation force nodes. The storage accounting includes the memory resource case of the system computation force node. The energy consumption comprises the daily average energy consumption condition of the statistical power calculation nodes. The AI algorithm support types include AI algorithm types supported by the statistical power node.
The network scheduling module is used for scheduling the computing power resources in the computing power network.
Fig. 2 is a flow chart of a network computing power arranging method according to an embodiment of the present application, as shown in fig. 2, the network computing power arranging method includes the following steps:
s110, acquiring scene information of the user equipment accessing the power computing network.
The user device accesses the computing network through a network interface provided by the network computing orchestration system shown in fig. 1. And under the condition that the user equipment is accessed to the power computing network, acquiring scene information of the user equipment accessing to the power computing network.
The motion condition of the user equipment can influence the distribution of the computational power resources, the higher the motion state of the user equipment is, the later the computational power resource distribution is, and the less sensitive the computational power nodes of the higher level are to the movement change of the user equipment under the condition of the same service requirement, so that the scenes can be divided into static scenes, low-speed motion scenes and high-speed motion scenes in the computational power network. The scene information is information corresponding to a scene.
Specifically, the higher the motion state of the user equipment, the faster the wireless network interface connected with the user equipment changes, for example, the user equipment is currently in the cell a, the computing power is provided by using the edge computing node of the cell a, in a very short time, the user equipment moves from the cell a to the cell B, when no direct route exists between the computing power edge nodes of the cell a to the cell B, the user equipment is switched from the computing power provided by using the edge computing node of the cell a to the computing power provided by using the edge computing node of the cell B, and then the computing power needs to be transmitted from the edge computing node of the cell a to the edge computing power node of the cell B through the higher computing power node; the user equipment moves to other cells when the edge computing node of the cell A sends to the edge computing node of the cell B through the higher computing node, so that the nodes providing computing power for the user equipment are difficult to match, and meanwhile, the computing power information is forwarded to the user equipment through the higher computing node frequently and forwarded to the switching cell, so that computing power resource waste is caused. Thus, the computing power resources are allocated based on the movement situation of the user equipment.
In some embodiments, obtaining scene information of a user device accessing a computing power network includes:
calculating the times of cell switching of the user equipment in unit time;
and determining the scene information corresponding to the times based on the corresponding relation between the times of cell switching and the scene information.
In one example, the correspondence between the number of cell handovers and the scenario information is:
static scene: the switching times of the user equipment are less than or equal to a/min;
low-speed motion scene: a/min < the number of user switching times is less than or equal to b/min;
high-speed motion scene: b/min < the number of user switching times is less than or equal to c/min;
wherein a, b and c are positive numbers.
S120, acquiring computing power resource information required to be called by the service in progress of the user equipment according to the scene information, wherein the computing power resource information comprises computing power resource demand types and computing power key indexes.
Based on the current scene information of the user equipment, acquiring information of computational power resources required to be invoked by the current ongoing service of the user equipment, wherein the computational power resource information comprises computational power resource requirement types and computational power key indexes required by the ongoing service.
The computing power resource demand type, i.e., the computing power resource type, may include at least one of a node type and a device type.
The power key indicators may include at least one of average time delay of the power node to the organization arrangement module, power accounting, storage accounting, energy consumption, algorithm type supported by the power node.
In some embodiments, the computing power resource information may further include a location of a computing power node corresponding to the computing power resource demand type, so as to facilitate subsequent orchestration of the computing power node based on the location of the computing power node.
S130, based on the calculation force load and network load information of calculation force nodes at all levels in the calculation force network system, matching task receiving units meeting calculation force resource demand types and calculation force key indexes from the calculation force network system, wherein the task receiving units comprise multi-level calculation force nodes with different calculation force demand types.
And acquiring the computational power load and network load information of all levels of computational power nodes in the computational power network, and matching task receiving units matched with the computational power resource demand type and the computational power key index from a plurality of task receiving units in the computational power network.
The computing network includes a plurality of task receiving units forming a tree schema (dendrogram) as shown in fig. 3. The task receiving unit is a unit based on characteristic data of each computing node in the computing network and computing type clustering division. Each task receiving unit includes a plurality of hierarchical computing nodes of different computing types. In a tree graph, the different types of computing nodes act as leaf nodes (i.e., the lowest level of computing nodes) in the tree, with the top level of the tree being the other different types of computing nodes with higher computing power.
S140, arranging the multi-level computing power nodes with different computing power demand types in the task receiving unit to the deployment position of the service.
And arranging the computing power nodes included in the task receiving units to the deployment positions of the services under the condition that the task receiving units corresponding to the computing power resources required by the services in progress of the user equipment are determined, so as to provide computing power for the services in progress of the user equipment.
Acquiring scene information of user equipment accessing to an algorithm network; acquiring computing power resource information required to be called by the service in progress of the user equipment according to the scene information, wherein the computing power resource information comprises computing power resource demand types and computing power key indexes; based on the calculation force load and network load information of calculation force nodes at all levels in the calculation force network system, matching a task receiving unit meeting calculation force resource demand types and calculation force key indexes from the calculation force network system, wherein the task receiving unit comprises multi-level calculation force nodes with different calculation force demand types; the multi-level computing nodes with different computing power demand types in the task receiving unit are arranged to the deployment position of the service, so that the arranged computing power nodes can meet the service demand types and computing power key indexes, and the performance characteristic demands and the utilization rate of computing power resources are improved.
In some embodiments, before matching task receiving units that satisfy the computing power resource demand type and the computing power key index from the computing power network system based on the computing power load and network load information of each level of computing power nodes in the computing power network system, the method further comprises:
acquiring characteristic data and calculation force types of all calculation force nodes in a calculation force network system, wherein the characteristic data comprises calculation force key indexes, and the calculation force types are used for matching calculation force resource demand types; clustering the characteristic data of each computing node to obtain a characteristic data set of computing nodes of a plurality of clustered clusters; calculating the mean value and standard deviation of the characteristic data of the computing force nodes in each cluster; calculating a target value based on the mean and standard deviation; and performing calculation force node connection according to the calculation force type of the calculation force nodes in each cluster and the target value of the calculation force nodes to obtain a plurality of task receiving units, wherein the task receiving units comprise calculation force nodes of which the layers are divided according to the calculation force type, and the calculation force nodes of different layers are connected through IP routing.
Specifically, in one example, each of the computing power network systems is acquiredCharacteristic data of the force calculation node, wherein the characteristic data set is assumed to be X= { X 1 ,...,x l Selecting cluster number K, initializingCharacteristic data of clustered computing force nodes are obtained to obtain cluster S i ＝{x i { i=1,. }, l }; then initialize +.>Obtaining S j = { j }, { j = 1,..; find the nearest two clusters S i And S is equal to j Wherein the nearest two clusters S i And S is equal to j The following conditions are satisfied:
obtaining two similar clusters S i And S is equal to j After that, fuse S i And S is equal to j The method comprises the steps of carrying out a first treatment on the surface of the Fusion S i And S is equal to j Namely, calculate S i And S is equal to j Delete duplicated S after union of (C) j S, i.e i ＝S j ∪S i and delete S j 。
If it isInitializing +.>Clustering is then performed again, and then the two nearest clusters are calculated until +.>At->Outputting S corresponding to K i 。
At the acquisition cluster S i After that, S i The characteristic data of the power calculation node in (a) is mapped to a standard normal distribution, then the characteristic data label value (which can also be the mean value of the characteristic data) mu is calculated, and the characteristic data label standard deviation (which can also be the characteristic dataStandard deviation) δ; then calculate normalized value X new ＝(X-μ)/δ。
Then, the method is implemented in R language, the parameter mode (method) is set as average (average), the computing nodes are gradually connected by using the method of mean clustering connection, a tree diagram (dendrogram) is obtained as shown in FIG. 4, FIG. 4 is a characteristic connection diagram taking a pattern example as 72 computing nodes, and the connection diagram contains the average value of the parameter mode.
Then dividing the tree diagram into a plurality of task connection units, wherein the task receiving units after division are shown in fig. 3, and meanwhile, setting layering parameters of the tree diagram (dendrogram) as a specific total layer number (layer) of the computing network. The computing power network is divided into computing power network task receiving units with mutually exclusive layers.
And then the divided tree diagram is subjected to self-configuration of the network IP routing connection of each computing node resource by the driving network scheduling module through the organization arrangement module of FIG. 1, so that a task receiving unit is actually formed, and the task receiving unit is managed by the management unit of the organization arrangement module, so that the balance of computing network load is realized.
Based on the same inventive concept, the embodiment of the present application further provides a network computing power arranging device 200, as shown in fig. 5, which includes an acquisition unit 210, a matching unit 220, and an arranging unit 330.
An obtaining unit 210, configured to obtain scene information of the user equipment accessing the power computing network;
the obtaining unit 210 is further configured to obtain, according to the scenario information, computing power resource information that needs to be invoked by the service that is being performed by the user equipment, where the computing power resource information includes a computing power resource requirement type and a computing power key indicator;
the matching unit 220 is configured to match, from the computing power network system, a task receiving unit that meets a computing power resource demand type and a computing power key index, where the task receiving unit includes multiple levels of computing power nodes with different computing power demand types, based on computing power load and network load information of computing power nodes at each level in the computing power network system;
an orchestration unit 230 for orchestrating the multi-level computing nodes of different computing power demand types in the task receiving unit to the deployment location of the service.
In one embodiment, the obtaining unit is configured to obtain scene information of the user equipment accessing the computing power network, and includes:
the acquisition unit is used for calculating the times of cell switching of the user equipment in unit time;
and determining the scene information corresponding to the times based on the corresponding relation between the times of cell switching and the scene information.
In one embodiment, the computing power resource demand type includes at least one of a type of computing power node and a device type; the computing power key index comprises at least one of average time delay from the computing power node to the organization arrangement module, computing power accounting, storage accounting, energy consumption and algorithm type supported by the computing power node.
In one embodiment, the orchestration device further comprises a clustering module, a computing module, and a connecting module;
the system comprises an acquisition unit, a calculation power network system and a calculation power management unit, wherein the acquisition unit is also used for acquiring characteristic data and calculation power types of all calculation power nodes in the calculation power network system, the characteristic data comprises calculation power key indexes, and the calculation power types are used for matching calculation power resource demand types;
the clustering module is used for clustering the characteristic data of each power computing node to obtain characteristic data sets of the power computing nodes of a plurality of clustered clusters;
the computing module is used for computing the mean value and standard deviation of the characteristic data of the computing force nodes in each cluster;
the calculation module is also used for calculating a target value based on the mean value and the standard deviation;
the connection module is used for carrying out calculation force node connection according to the calculation force type of the calculation force nodes in each cluster and the target value of the calculation force nodes to obtain a plurality of task receiving units, wherein the task receiving units comprise calculation force nodes of which the levels are divided according to the calculation force type, and the calculation force nodes of different levels are connected through IP routing.
The modules shown in fig. 5 have the arrangement method for realizing the network computing power shown in fig. 1 to 4, and the technical effects achieved are the same, so that the description is omitted herein for brevity.
Fig. 6 shows a schematic hardware structure of an electronic device according to an embodiment of the present application.
The electronic device may comprise a processor 301 and a memory 302 storing computer program instructions.
In particular, the processor 301 may include a Central Processing Unit (CPU), or an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), or may be configured as one or more integrated circuits that implement embodiments of the present application.
Memory 302 may include mass storage for data or instructions. By way of example, and not limitation, memory 302 may comprise a Hard Disk Drive (HDD), floppy Disk Drive, flash memory, optical Disk, magneto-optical Disk, magnetic tape, or universal serial bus (Universal Serial Bus, USB) Drive, or a combination of two or more of the foregoing. Memory 302 may include removable or non-removable (or fixed) media, where appropriate. Memory 302 may be internal or external to the integrated gateway disaster recovery device, where appropriate. In a particular embodiment, the memory 302 is a non-volatile solid-state memory.
The memory may include Read Only Memory (ROM), random Access Memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical, or other physical/tangible memory storage devices. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software comprising computer-executable instructions and when the software is executed (e.g., by one or more processors) it is operable to perform the operations described with reference to a method in accordance with an aspect of the application.
The processor 301 implements the network computing arrangement of any of the above embodiments by reading and executing computer program instructions stored in the memory 302.
As one example, the electronic device may also include a communication interface 303 and a bus 310. As shown in fig. 3, the processor 301, the memory 302, and the communication interface 303 are connected to each other through a bus 3410 and perform communication with each other.
The communication interface 303 is mainly used to implement communication between each module, device, unit and/or apparatus in the embodiment of the present application.
Bus 310 includes hardware, software, or both that couple the components of the online data flow billing device to each other. By way of example, and not limitation, the buses may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an infiniband interconnect, a Low Pin Count (LPC) bus, a memory bus, a micro channel architecture (MCa) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a video electronics standards association local (VLB) bus, or other suitable bus, or a combination of two or more of the above. Bus 310 may include one or more buses, where appropriate. Although embodiments of the application have been described and illustrated with respect to a particular bus, the application contemplates any suitable bus or interconnect.
The electronic device may execute the information acquisition method in the embodiment of the present application, thereby implementing the arrangement method of the information network computing power described in connection with fig. 1 to 4.
In addition, in combination with the information acquisition method in the above embodiment, the embodiment of the present application may be implemented by providing a computer storage medium. The computer storage medium has stored thereon computer program instructions; the computer program instructions, when executed by a processor, implement a network computing arrangement method of any of the above embodiments.
It should be understood that the application is not limited to the particular arrangements and instrumentality described above and shown in the drawings. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the present application are not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between steps, after appreciating the spirit of the present application.
The functional blocks shown in the above block diagrams may be implemented in hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, a plug-in, a function card, or the like. When implemented in software, the elements of the application are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine readable medium or transmitted over transmission media or communication links by a data signal carried in a carrier wave. A "machine-readable medium" may include any medium that can store or transfer information. Examples of machine-readable media include electronic circuitry, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio Frequency (RF) links, and the like. The code segments may be downloaded via computer networks such as the internet, intranets, etc.
It should also be noted that the exemplary embodiments mentioned in this disclosure describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, or may be performed in a different order from the order in the embodiments, or several steps may be performed simultaneously.
Aspects of the present disclosure are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions/acts specified in the flowchart and/or block diagram block or blocks. Such a processor may be, but is not limited to being, a general purpose processor, a special purpose processor, an application specific processor, or a field programmable logic circuit. It will also be understood that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware which performs the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In the foregoing, only the specific embodiments of the present application are described, and it will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the systems, modules and units described above may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein. It should be understood that the scope of the present application is not limited thereto, and any equivalent modifications or substitutions can be easily made by those skilled in the art within the technical scope of the present application, and they should be included in the scope of the present application.
Claims (10)
1. A method for orchestrating network computing power, comprising:
acquiring scene information of user equipment accessing to an algorithm network;
acquiring computing power resource information required to be called by the service in progress of the user equipment according to the scene information, wherein the computing power resource information comprises computing power resource demand types and computing power key indexes;
based on the calculation force load and network load information of calculation force nodes at all levels in a calculation force network system, matching a task receiving unit meeting the calculation force resource demand type and calculation force key indexes from the calculation force network system, wherein the task receiving unit comprises multiple levels of calculation force nodes with different calculation force demand types;
and arranging multi-level computing power nodes with different computing power demand types in the task receiving unit to a deployment position of the service.
2. The arrangement method according to claim 1, wherein the obtaining scene information of the user equipment accessing the computing power network comprises:
calculating the times of cell switching of the user equipment in unit time;
and determining the scene information corresponding to the times based on the corresponding relation between the times of cell switching and the scene information.
3. The method of claim 1 or 2, wherein the computing power resource demand type comprises at least one of a type of computing power node and a device type; the calculation power key index comprises at least one of average time delay from a calculation power node to an organization arrangement module, calculation power accounting, storage accounting, energy consumption and algorithm type supported by the calculation power node.
4. The orchestration method according to claim 1, wherein before matching from the computing network system task receiving units that meet the computing resource demand type and computing power key indicators based on computing power load and network load information of each level of computing power nodes in the computing network system, the method further comprises:
acquiring characteristic data and calculation force types of all calculation force nodes in the calculation force network system, wherein the characteristic data comprises calculation force key indexes, and the calculation force types are used for matching calculation force resource demand types;
clustering the characteristic data of each computing node to obtain a characteristic data set of computing nodes of a plurality of clustered clusters;
calculating the mean value and standard deviation of the characteristic data of the computing force nodes in each cluster;
calculating a target value based on the mean and standard deviation;
and performing calculation force node connection according to the calculation force type of the calculation force nodes in each cluster and the target value of the calculation force nodes to obtain a plurality of task receiving units, wherein the task receiving units comprise calculation force calculation nodes which are divided into layers according to the calculation force type, and the calculation force nodes of different layers are connected through IP routing.
5. A network computing power orchestration device, comprising:
the acquisition unit is used for acquiring scene information of the user equipment accessing the power calculation network;
the acquisition unit is further used for acquiring computing power resource information required to be invoked by the service in progress of the user equipment according to the scene information, wherein the computing power resource information comprises computing power resource demand types and computing power key indexes;
the matching unit is used for matching a task receiving unit which meets the computing power resource demand type and the computing power key index from the computing power network system based on the computing power load and network load information of computing power nodes at all levels in the computing power network system, and the task receiving unit comprises multiple levels of computing power nodes with different computing power demand types;
and the arrangement unit is used for arranging the multi-level computing force nodes with different computing force demand types in the task receiving unit to the deployment position of the service.
6. The arrangement device according to claim 5, wherein the obtaining unit is configured to obtain scene information of the user equipment accessing the computing power network, and the method comprises:
the acquisition unit is used for calculating the times of cell switching of the user equipment in unit time;
and determining the scene information corresponding to the times based on the corresponding relation between the times of cell switching and the scene information.
7. The orchestration apparatus according to claim 5 or 6, wherein the computing resource demand type comprises at least one of a type of computing node and a device type; the calculation power key index comprises at least one of average time delay from a calculation power node to an organization arrangement module, calculation power accounting, storage accounting, energy consumption and algorithm type supported by the calculation power node.
8. The orchestration device according to claim 6, further comprising a clustering module, a computing module, and a connection module;
the acquisition unit is further used for acquiring characteristic data and calculation force types of all calculation force nodes in the calculation force network system, wherein the characteristic data comprise calculation force key indexes, and the calculation force types are used for matching calculation force resource demand types;
the clustering module is used for clustering the characteristic data of each computing power node to obtain a characteristic data set of computing power nodes of a plurality of clustered clusters;
the computing module is used for computing the mean value and standard deviation of the characteristic data of the computing force nodes in each cluster;
the calculation module is further used for calculating a target value based on the mean value and the standard deviation;
the connection module is used for carrying out calculation force node connection according to the calculation force type of the calculation force nodes in each cluster and the target value of the calculation force nodes to obtain a plurality of task receiving units, wherein the task receiving units comprise calculation force nodes which are divided into layers according to the calculation force type, and the calculation force nodes of different layers are connected through IP routing.
9. An electronic device, the device comprising: a processor and a memory storing computer program instructions;
the processor, when executing the computer program instructions, implements the network computing power orchestration method according to any one of claims 1 to 4.
10. A computer readable storage medium, characterized in that it has stored thereon computer program instructions which, when executed by a processor, implement the network computing power orchestration method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211520184.5A CN116962528A (en) | 2022-11-30 | 2022-11-30 | Network computing power arranging method, device, equipment and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211520184.5A CN116962528A (en) | 2022-11-30 | 2022-11-30 | Network computing power arranging method, device, equipment and computer storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116962528A true CN116962528A (en) | 2023-10-27 |
Family
ID=88450043
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211520184.5A Pending CN116962528A (en) | 2022-11-30 | 2022-11-30 | Network computing power arranging method, device, equipment and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116962528A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117687798A (en) * | 2024-02-01 | 2024-03-12 | 浪潮通信信息系统有限公司 | Management and control method, system and storage medium for original application of computing power network |
| CN117994882A (en) * | 2024-04-02 | 2024-05-07 | 杭州海康威视数字技术股份有限公司 | Pass permission authentication equipment and method |
-
2022
- 2022-11-30 CN CN202211520184.5A patent/CN116962528A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117687798A (en) * | 2024-02-01 | 2024-03-12 | 浪潮通信信息系统有限公司 | Management and control method, system and storage medium for original application of computing power network |
| CN117687798B (en) * | 2024-02-01 | 2024-05-10 | 浪潮通信信息系统有限公司 | Management and control method, system and storage medium for original application of computing power network |
| CN117994882A (en) * | 2024-04-02 | 2024-05-07 | 杭州海康威视数字技术股份有限公司 | Pass permission authentication equipment and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116962528A (en) | Network computing power arranging method, device, equipment and computer storage medium | |
| Xia et al. | Federated-learning-based client scheduling for low-latency wireless communications | |
| CN111867049B (en) | Positioning method, positioning device and storage medium | |
| CN104995870A (en) | Multi-objective server placement determination | |
| CN108243245A (en) | The Radio Access Network and its resource allocation method calculated based on mixing fog | |
| CN108901075A (en) | A kind of resource allocation methods based on GS algorithm | |
| CN110061881B (en) | Energy consumption perception virtual network mapping algorithm based on Internet of things | |
| CN108764458B (en) | Method and system for reducing storage space consumption and calculation amount of mobile equipment | |
| Gao et al. | A credible and lightweight multidimensional trust evaluation mechanism for service-oriented IoT edge computing environment | |
| CN111404932A (en) | Method for accessing medical institution system to smart medical cloud service platform | |
| CN114118748B (en) | Service quality prediction method and device, electronic equipment and storage medium | |
| Abbasi et al. | An intelligent method for reducing the overhead of analysing big data flows in Openflow switch | |
| CN113438733A (en) | System and method for realizing indoor positioning based on 5G and WIFI and electronic equipment | |
| CN104331336B (en) | Be matched with the multilayer nest balancing method of loads of high-performance computer structure | |
| CN103631659B (en) | Schedule optimization method for communication energy consumption in on-chip network | |
| Sun | [Retracted] Certificateless Batch Authentication Scheme and Intrusion Detection Model Based on the Mobile Edge Computing Technology NDN‐IoT Environment | |
| CN109005060A (en) | A kind of deep learning optimizing application frame based on hierarchical high isomerism distributed system | |
| CN106550387B (en) | A kind of wireless sensor network routing layer QoS evaluating method | |
| CN113938992B (en) | Threshold determining method and device | |
| Li et al. | Advanced split BIRCH algorithm in reconfigurable network | |
| Jiang et al. | A Task Offloading Method with Edge for 5G‐Envisioned Cyber‐Physical‐Social Systems | |
| CN111278037B (en) | Cell value evaluation method and device | |
| Shirvani Moghaddam et al. | A threshold‐based sorting algorithm for dense wireless sensor systems and communication networks | |
| Chen et al. | A multi‐objective optimization model based on immune algorithm in wireless mesh networks | |
| WO2024067093A1 (en) | Communication load forecasting method and apparatus, device, and 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 |