CN115794373A - Calculation force resource hierarchical scheduling method, system, electronic equipment and storage medium - Google Patents
Calculation force resource hierarchical scheduling method, system, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN115794373A CN115794373A CN202211287320.0A CN202211287320A CN115794373A CN 115794373 A CN115794373 A CN 115794373A CN 202211287320 A CN202211287320 A CN 202211287320A CN 115794373 A CN115794373 A CN 115794373A
- Authority
- CN
- China
- Prior art keywords
- scheduling
- node
- current
- computing power
- computing
- 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
- 238000004364 calculation method Methods 0.000 title claims abstract description 97
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000004590 computer program Methods 0.000 claims description 14
- 238000010586 diagram Methods 0.000 description 15
- 238000012544 monitoring process Methods 0.000 description 10
- 238000004458 analytical method Methods 0.000 description 9
- 230000008447 perception Effects 0.000 description 8
- 238000004891 communication Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 230000036541 health Effects 0.000 description 6
- 238000001514 detection method Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000012216 screening Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005538 encapsulation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to the field of computers, and provides a calculation power resource hierarchical scheduling method, a calculation power resource hierarchical scheduling system, electronic equipment and a storage medium, wherein the method comprises the following steps: determining that the current computing power resource of the current scheduling node does not meet the current computing power requirement according to the local computing power information and the computing power requirement information of the current scheduling node, and determining an interconnection scheduling node, a regional scheduling node, a cross-domain cooperative scheduling node or a global scheduling network of the current scheduling node; and performing hierarchical scheduling of computing resources based on the interconnection scheduling node, the regional scheduling node, the cross-domain cooperative scheduling node or the universe scheduling network and the current computing demand. According to the hierarchical scheduling method for the computing power resources, when the current computing power resources do not meet the current computing power requirements, the computing power requirements are transmitted to the upper layer direction along topological connection through the interconnection scheduling nodes, the regional scheduling nodes, the cross-domain cooperative scheduling nodes or the universe scheduling network in sequence to find the computing power nodes meeting the requirements, and therefore the cross-region and cross-layer scheduling of the computing power resources is achieved.
Description
Technical Field
The present invention relates to the field of computers, and in particular, to the field of computing power networks, and in particular, to a computing power resource hierarchical scheduling method, system, electronic device, and storage medium.
Background
The traditional computational resource scheduling method is mainly a centralized scheduling method, the centralized scheduling method generally uniformly collects resources to be scheduled to a central control node by adopting a timing or real-time synchronization mechanism for uniform storage and management, and after a scheduling requirement is generated, the central control node performs comprehensive analysis to select the resources meeting the condition for scheduling and opening. The computational power network faces the trans-regional and trans-domain ubiquitous computational power resources, including general computational power, intelligent computational power, super computational power and the like, and has the characteristics of wide regional distribution range, heterogeneous types, large scale quantity and the like, and the traditional computational power resource scheduling method cannot perform trans-regional and trans-hierarchical scheduling of computational power resources.
Disclosure of Invention
The invention provides a method, a system, electronic equipment and a storage medium for hierarchical scheduling of computing resources, and aims to realize cross-region and cross-layer scheduling of computing resources.
In a first aspect, the present invention provides a method for hierarchical scheduling of computing resources, including:
determining whether the current computing power resource of the current scheduling node meets the current computing power requirement or not according to the local computing power information and the computing power requirement information of the current scheduling node;
if the current computing power resource does not meet the current computing power requirement, determining an interconnection scheduling node, a regional scheduling node, a cross-domain cooperative scheduling node or a global scheduling network of the current scheduling node;
and performing calculation resource hierarchical scheduling based on the interconnection scheduling node, the regional scheduling node, the cross-domain cooperative scheduling node or the universe scheduling network and the current calculation requirement.
In one embodiment, performing hierarchical scheduling of computing resources based on the interconnected scheduling nodes and the current computing demand comprises:
determining distributable computing power resources of the interconnected scheduling nodes and determining whether the distributable computing power resources meet the current computing power requirement;
and if the distributable computing power resource meets the current computing power requirement, determining a first target computing power node in the interconnected scheduling nodes, and calling the computing power resource of the first target computing power node.
Performing hierarchical scheduling of computing power resources based on the regional scheduling nodes and the current computing power demand, comprising:
if the distributable computing power resource does not meet the current computing power requirement, determining whether computing power resource scheduling is allowed to go out of the local area;
if the local resource scheduling is allowed, determining each local node to be scheduled which belongs to the same scheduling region as the current scheduling node based on the region scheduling node;
and performing calculation power resource hierarchical scheduling based on each local node to be scheduled and the current calculation power requirement.
The step of performing calculation power resource hierarchical scheduling based on each local node to be scheduled and the current calculation power demand comprises the following steps:
determining whether a local target scheduling node exists in each local node to be scheduled, wherein the local computing power resource of the local target scheduling node meets the current computing power requirement;
and if the local target scheduling node exists, determining a second target computational power node in the local target scheduling node, and calling computational power resources of the second target computational power node.
Performing hierarchical scheduling of computing power resources based on the cross-domain cooperative scheduling node and the current computing power demand, including:
if the local target scheduling node does not exist, determining each associated region to-be-scheduled node associated with the region scheduling node based on the cross-domain cooperative scheduling node;
determining whether each associated region target scheduling node exists in the nodes to be scheduled in the associated region, wherein the region computing power resource of the associated region target scheduling node meets the current computing power requirement;
and if the associated region target scheduling node exists, determining a third target computing power node in the associated region target scheduling node, and calling computing power resources of the third target computing power node.
Performing hierarchical scheduling of computing resources based on the global scheduling network and the current computing demand, comprising:
if the associated regional target scheduling node does not exist, searching a global node to be scheduled through interconnected cross-domain cooperative scheduling nodes in a global range based on the global scheduling network;
determining whether a global target scheduling node exists in the global node to be scheduled, wherein the computing power resource of the global target scheduling node meets the current computing power requirement;
and if the global target scheduling node exists, determining a fourth target computing power node in the global target scheduling node, and calling computing power resources of the fourth target computing power node.
After determining whether a global target scheduling node exists in the global node to be scheduled, the method further includes:
and if the global target scheduling node does not exist, determining that the computing resource scheduling of the current computing requirement fails.
In a second aspect, the present invention provides a hierarchical scheduling system for computing resources, comprising:
the decision module is used for determining whether the current computing power resource of the current scheduling node meets the current computing power requirement or not according to the local computing power information and the computing power requirement information of the current scheduling node;
a determining module, configured to determine an interconnection scheduling node, a local scheduling node, a cross-domain cooperative scheduling node, or a global scheduling network of the current scheduling node if the current computational resource does not meet the current computational demand;
and the hierarchical scheduling module is used for performing hierarchical scheduling on the computing resources based on the interconnection scheduling node, the regional scheduling node, the cross-domain cooperative scheduling node or the universe scheduling network and the current computing demand.
In a third aspect, the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the method for hierarchical scheduling of computational resources according to the first aspect is implemented.
In a fourth aspect, the present invention also provides a non-transitory computer-readable storage medium comprising a computer program which, when executed by the processor, implements the hierarchical scheduling method of computational resources of the first aspect.
In a fifth aspect, the present invention further provides a computer program product comprising a computer program, which when executed by the processor implements the hierarchical scheduling method for computational resources of the first aspect.
The invention provides a computing power resource hierarchical scheduling method, a computing power resource hierarchical scheduling system, electronic equipment and a storage medium, wherein whether the current computing power resource of the current scheduling node meets the current computing power requirement is determined according to the local computing power information and the computing power requirement information of the current scheduling node; if the current computing power resource does not meet the current computing power requirement, determining an interconnection scheduling node, a regional scheduling node, a cross-domain cooperative scheduling node or a global scheduling network of the current scheduling node; and performing hierarchical scheduling of the computing resources based on the interconnection scheduling node, the regional scheduling node, the cross-domain cooperative scheduling node or the universe scheduling network and the current computing demand. In the process of hierarchical scheduling of the computational power resources, when the current computational power resources of the current scheduling nodes do not meet the current computational power requirements, the cross-region and cross-level scheduling computational power resources are sequentially carried out through the interconnected scheduling nodes, the regional scheduling nodes, the cross-region collaborative scheduling nodes or the universe scheduling network, the computational power requirements are transmitted to the upper layer direction along the topological connection to find the computational power nodes meeting the requirements, and the cross-region and cross-level scheduling computational power resources are realized.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed for the description of the embodiment or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a schematic flow chart of a hierarchical scheduling method for computing resources according to the present invention;
FIG. 2 is a functional diagram of a scheduling node provided by the present invention;
FIG. 3 is a schematic diagram of a hierarchical scheduling node relationship provided by the present invention;
FIG. 4 is a schematic diagram of a local node interconnection relationship provided by the present invention;
FIG. 5 is a schematic illustration of local scheduling provided by the present invention;
FIG. 6 is a schematic diagram of scheduling node interconnection provided by the present invention;
FIG. 7 is a schematic diagram of cross-domain scheduling provided by the present invention;
FIG. 8 is a schematic structural diagram of a hierarchical resource scheduling system provided by the present invention;
fig. 9 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
While a logical order is shown in the flow chart, in some cases, the steps shown or described may be performed in a different order than that shown or described.
Referring to fig. 1, fig. 1 is a schematic flow chart of a computational resource hierarchical scheduling method provided by the present invention. The method for hierarchical scheduling of computing resources provided by the embodiment of the invention comprises the following steps:
It should be noted that, for convenience of description, the computationally resource hierarchical scheduling system of the embodiment of the present invention can be described as a resource scheduling system.
Referring to fig. 2, fig. 2 is a functional schematic diagram of a scheduling node provided by the present invention, and a resource scheduling system includes a control plane and a service plane, where the control plane includes a sensing module, a decision module, a connection module, and a monitoring module, and the service plane includes a requirement analysis module, a data configuration module, and a service provisioning module. Further, referring to fig. 3, fig. 3 is a schematic diagram of a hierarchical scheduling node relationship provided by the present invention, therefore, a hierarchical scheduling architecture may be constructed by deploying a plurality of scheduling nodes according to an actual scene, scheduling nodes of each level in the hierarchical scheduling architecture may be specifically divided into local scheduling nodes, cross-domain cooperative scheduling nodes, and a global scheduling network, and cooperative scheduling of different levels is implemented by the hierarchical scheduling architecture.
Furthermore, the perception module mainly collects the information of computing power, network, environment and the like under the current scheduling node, the collected data comprise resources, capacity, performance, faults, events, states and the like, the collected data are analyzed and standardized to obtain standardized perception data, the standardized perception data mainly comprise computing power types, computing power sizes, residual capacities, operating states, health degrees and the like, the standardized perception data can be issued to the scheduling network through the connection module, and the superior scheduling node or the interconnected scheduling node can store and analyze the received perception data to provide data for computing power decision and selection.
Further, data collection is mainly performed through a butt joint open interface to periodically collect information such as computing power, networks and environments in the current node, the collection period is default to 5 minutes, and custom setting is supported. Data with high real-time requirements, such as capacity, faults, events and states, can be acquired in real time by adopting a message queue mode.
Furthermore, the perception analysis of the perception module mainly collects and correlates the collected data, and comprehensively analyzes the operation health degree, capacity, fault root cause and the like through a relevant AI model and algorithm, so as to provide reference data for the calculation power decision.
Further, the standardization processing of the perception module mainly processes the computing power and related performance indexes of the network, including computing power type, computing power size, residual capacity, exit network bandwidth, network average delay, network jitter and the like. The residual capacity and the health degree of each type of computational power, the residual capacity and the health degree of each type of storage, the bandwidth of an egress network, the average time delay, the health degree and other operation environment data, such as a container, a DevOps, a TensorFlow, a CAFFE and the like, are obtained by calculating the related indexes, namely the local computational power information of the current scheduling node is obtained.
Furthermore, the decision module mainly determines whether the current computational power resource of the current scheduling node meets the current computational power requirement according to the computational power requirement information analyzed by the requirement analysis module and the local computational power information of the current scheduling node. If the calculation can be satisfied, selecting proper target calculation power to be delivered to service opening for instantiation; if the current scheduling node cannot meet the requirement but an interconnection scheduling node meeting the requirement exists, sending a request to a specified interconnection node; otherwise, the computing power request is forwarded to the connection module according to the hierarchical identification to request the superior scheduling node to make a decision, if the hierarchical identification does not allow forwarding, the scheduling fails, and failure information and suggestions are provided to the superior application.
Further, the decision method of the decision module is mainly expressed as: firstly, screening out target calculation force nodes meeting the current calculation force demand according to related indexes, wherein the target calculation force nodes comprise conditions such as calculation force type, calculation force size, network bandwidth, storage type, storage size and environment type; then, calculating scores and sequencing by adopting an index weighted average mode, wherein the calculation rule of the weighted average can be configured through a service interface, such as residual capacity, utilization rate, time delay, average fault time, health degree and the like; and finally, selecting five nodes with the highest scores, screening again according to the performance requirement and the using mode of the scheduling request, performing weighted calculation again on indexes such as time delay, jitter and fault time if high stability is required, increasing space position parameters like a machine room, the same frame and the like if a data source position is provided and the data source is required to be close to, obtaining a final calculation force node, and scheduling the calculation force resource of the final calculation force node.
Further, the connection module mainly has functions of topology connection, message broadcasting, data channel and heartbeat detection, and each function is specifically analyzed as follows: the topological connection is mainly used for interconnection among scheduling nodes to construct a hierarchical scheduling topology. And according to the rule of hierarchical scheduling, the method is used for connecting the upper and lower scheduling nodes and interconnecting the leaf scheduling nodes in the local scheduling node. In order to prevent broadcast storm, it is suggested that only leaf scheduling nodes and cross-domain collaborative scheduling nodes may be connected in a pairwise interconnection manner, other scheduling nodes adopt an up-down connection relationship, and use of left-right connection is prohibited, referring to fig. 4, fig. 4 is a schematic diagram of a local node interconnection relationship provided by the present invention, and a data structure of a topological connection is: the system comprises an upper connection node, a lower connection node and an interconnection node, wherein the upper connection node is only one at most, and the lower connection node and the interconnection node can be zero or more.
The message broadcasting mainly broadcasts the computing power state data after the current scheduling node is standardized to other scheduling nodes connected with the current scheduling node according to the topological connection relation, and receives the computing power state data of other connected scheduling nodes and caches the computing power state data into the storage of the sensing module. In the context of the previous and subsequent relationships, status message data is passed up, and the subsequent node does not receive the computing power message of the previous node. In a peer-to-peer interconnection relationship, the interconnection nodes may send and receive messages of each other.
The data channel is mainly used for establishing the data channel of the access scheduling node and the target scheduling node when the scheduling node is crossed so as to directly transmit data. The data channel adopts CS (Client-Server) architecture design, when being scheduled, a target scheduling node starts monitoring service, a scheduling node is accessed to start a Client, the target scheduling node and the Client establish network link in a Socket mode, communication is carried out based on a TCP (Transmission Control Protocol) Protocol at present, and SRv6 related tunnel technology can be further considered.
The heartbeat detection mainly maintains the accessibility of the connection of the scheduling nodes, the connection is maintained by periodically sending heartbeats to the connected scheduling nodes according to the topology connection data, and the monitoring module is informed to carry out further processing under the condition of connection interruption or high time delay. The heartbeat detection firstly broadcasts heartbeat information to the outside at the default frequency of every 2 seconds, and after heartbeat detection modules of other nodes receive the heartbeat information, the active state and the latest heartbeat time of related nodes are updated. If a heartbeat is not received for 10 consecutive times (20 seconds), the associated node is marked as lost, then the monitoring module is notified and an attempt reconnect request is initiated. And if the heartbeat information is not received within 2 minutes, marking the offline node as an offline state, informing the monitoring module, and removing the offline node from the topological connection.
Furthermore, the monitoring module mainly tracks and monitors the operation of the scheduling node from end to end, so that the scheduling process and the scheduling result can be conveniently subjected to statistical analysis. The monitoring module performs standardized encapsulation on the monitoring data and provides an open interface, except for internal use, the monitoring module can be opened to systems such as Prometheus + Grafana and ELK to perform monitoring of the whole scheduling network, and the metrics interface is realized to provide running state data for Prometheus.
Further, the data configuration module is mainly used for transmitting, receiving and configuring data after the data channel crossing the nodes is established.
Furthermore, the demand analysis module analyzes the calculation force demand information, splits the calculation force demand information into atomic demands according to analysis rules, filters noise data and performs standardized packaging of the requests. The standardized field includes: regional limitations (local, regional, cross-domain, global), security limitations (encryption, authentication, etc.), computing power types (CPU, GPU, DPU, TPU, FPGA, ASIC, etc.), computing power size, network bandwidth, storage type, storage size, environment types (cloud host, bare metal, container, tensrflow, CAFFE, etc.), mirror addresses, etc., to obtain the current computing power demand of the computing power demand information.
Furthermore, the service provision module is mainly used for arranging the calculation power according to the target calculation power node selected by the decision module and the current calculation power requirement output by the analysis module, and applying and providing the calculation power, deploying the application environment, configuring the service and the like. The method comprises the steps that in a resource scheduling system, the scheduling of service fulfillment can be carried out by calling the capacity of a scheduling center, the service fulfillment sends target calculation force nodes and current calculation force requirements to the scheduling center, the scheduling center carries out combination and flow creation, the service fulfillment carries out scheduling execution according to scheduling results, and finally the service fulfillment is completed.
Specifically, a decision module in the resource scheduling system determines whether the current computing power resource of the current scheduling node meets the current computing power requirement or not according to the current computing power information and the computing power requirement information of the current scheduling node. Referring to fig. 5, fig. 5 is a schematic diagram of local scheduling provided by the present invention, if it is determined that the current computational power resource of the current scheduling node meets the current computational power demand according to the current computational power information and the computational power demand information of the current scheduling node, the decision module selects a target computational power node through local scheduling, and then completes service provisioning through the target computational power node and the current computational power demand.
It should be noted that local scheduling may be used in a scenario where the user data compliance requirement is limited to local, nearby service, or performance priority. The local scheduling node may be designed according to an actual scene, and if there are a plurality of separated computation nodes or data centers for management, the local scheduling node may be expanded into a plurality of sub-scheduling nodes and interconnected, referring to fig. 4.
Further, if the current computing power resource does not meet the current computing power requirement, the decision module determines an interconnection scheduling node, a local scheduling node, a cross-domain cooperative scheduling node or a global scheduling network of the current scheduling node, and performs computing power resource hierarchical scheduling on the current computing power requirement sequentially through the interconnection scheduling node, the local scheduling node, the cross-domain cooperative scheduling node or the global scheduling network.
The invention provides a hierarchical scheduling method of computing power resources, which determines whether the current computing power resources of the current scheduling nodes meet the current computing power requirements or not according to the local computing power information and the computing power requirement information of the current scheduling nodes; if the current computing power resource does not meet the current computing power requirement, sequentially determining an interconnection scheduling node, a regional scheduling node, a cross-domain cooperative scheduling node or a global scheduling network of the current scheduling node; and performing hierarchical scheduling of the computing resources based on the interconnection scheduling node, the regional scheduling node, the cross-domain cooperative scheduling node or the universe scheduling network and the current computing demand.
In the process of hierarchical scheduling of the computing power resources, when the current computing power resources of the current scheduling nodes do not meet the current computing power requirements, the computing power resources are sequentially scheduled across regions and levels through the interconnection scheduling nodes, the regional scheduling nodes, the cross-domain cooperative scheduling nodes or the universe scheduling network, the computing power requirements are transmitted to the upper layer direction along the topological connection to search the computing power nodes meeting the requirements, and the cross-region and cross-level scheduling of the computing power resources is achieved.
Further, the hierarchical scheduling of computing power resources based on the interconnection scheduling node and the current computing power demand, which is described in step 103, includes:
determining distributable computing resources of the interconnected scheduling nodes, and determining whether the distributable computing resources meet the current computing requirement;
and if the distributable computing power resource meets the current computing power requirement, determining a first target computing power node in the interconnected scheduling nodes, and calling the computing power resource of the first target computing power node.
Specifically, if the current computing power resource does not meet the current computing power requirement, the decision module determines whether a local interconnected scheduling node interconnected with the current scheduling node exists, and if the local interconnected scheduling node interconnected with the current scheduling node does not exist and the computing power resource is not allowed to be dispatched, the decision module determines that the computing power resource scheduling of the current computing power requirement fails. And if the interconnected scheduling node interconnected with the current scheduling node does not exist, but the computational resource scheduling is allowed to be local, the decision module carries out computational resource scheduling on the current computational demand through the regional scheduling node.
Further, if an interconnection scheduling node interconnected with the current scheduling node exists, the decision module firstly carries out calculation resource scheduling on the current calculation requirement through the interconnection scheduling node, and under the condition that the calculation resource of the interconnection scheduling node does not meet the current calculation requirement, the calculation resource is allowed to go out of the local area, and then the regional scheduling node carries out calculation resource scheduling on the current calculation requirement.
Further, the decision module determines the allocable computational power resources of the interconnected scheduling nodes through the cache in the sensing module, and determines whether the allocable computational power resources meet the current computational power requirement.
Referring to fig. 6, fig. 6 is a schematic diagram of scheduling node interconnection scheduling provided by the present invention, and if it is determined that the allocable computing power resource meets the current computing power demand, the decision module sends the current computing power demand to the interconnection scheduling node. And a decision module of the interconnected dispatching nodes determines a first target computing power node in the interconnected dispatching nodes according to the current computing power demand, and performs computing power resource dispatching on the current computing power demand through the first target computing power node. That is, it can be understood that the decision module of the current scheduling node schedules the computing resources of the first target computing resource node in the interconnected scheduling nodes by hierarchy.
The embodiment of the invention carries out calculation resource scheduling on the current calculation demand through the interconnected scheduling nodes interconnected with the current scheduling node, thereby realizing cross-region cross-level scheduling of calculation resources. Meanwhile, by adopting a hierarchical scheduling mode, data is stored in a hierarchical and dispersed mode, and scheduling nodes are used for performing cooperative scheduling, so that the management problem, the safety problem and the compliance problem caused by data centralized storage adopted by an original scheduling mode are effectively solved.
Further, the hierarchical scheduling of computing power resources based on the regional scheduling nodes and the current computing power demand, which is described in step 103, includes:
if the distributable computing power resource does not meet the current computing power requirement, determining whether computing power resource scheduling is allowed to go out of the local area;
if the local resource scheduling is allowed, determining each local node to be scheduled which belongs to the same scheduling region as the current scheduling node based on the region scheduling node;
and performing calculation power resource hierarchical scheduling based on each local node to be scheduled and the current calculation power requirement.
It should be noted that, the regional scheduling may perform coordinated management on a plurality of local scheduling nodes, so that the computational power demand may be more evenly scheduled and distributed, and computational power resources in a wider range may be fully utilized, for example, the computational power demand of large-scale computation may be scheduled to a superior intelligent computation center. After the local scheduling node is accessed to analyze the computational power requirement, when the higher-level scheduling node is found to be required to perform scheduling according to the computational power requirement upgrading strategy and the user requirement analysis, the scheduling request can be sent to the higher-level regional scheduling node for processing.
After receiving the request, the superior regional scheduling node searches for computational power resources in the regional range, selects the subordinate scheduling node meeting the requirement, sends the request to the designated subordinate scheduling node, and carries out further processing by the subordinate scheduling node. The subordinate scheduling node and the access scheduling node directly establish a logic channel through the access module to carry out message interaction and data transmission.
Specifically, if it is determined that the allocable computing resource does not meet the current computing requirement, the decision module of the current scheduling node determines whether the computing resource scheduling is allowed to go out of the local area. And if the local computer determines that the computational resource scheduling is not allowed, the decision module of the current scheduling node determines that the computational resource scheduling of the current computational demand fails. And if the local calculation resource scheduling permission is determined, the decision module of the current scheduling node sends the current calculation power requirement to the regional scheduling node of the current scheduling node. And the regional scheduling node determines each local node to be scheduled which belongs to the same scheduling region as the current scheduling node. Furthermore, the regional scheduling nodes perform calculation power resource hierarchical scheduling according to each local node to be scheduled and the current calculation power requirement.
The embodiment of the invention carries out calculation resource scheduling on the current calculation requirement through the regional scheduling node of the current scheduling node, thereby realizing cross-region cross-hierarchy scheduling of calculation resources. Meanwhile, by adopting a hierarchical scheduling mode, data is stored in a hierarchical and dispersed mode, and scheduling nodes are used for performing cooperative scheduling, so that the management problem, the safety problem and the compliance problem caused by data centralized storage adopted by an original scheduling mode are effectively solved.
Further, performing computation power resource hierarchical scheduling based on each local node to be scheduled and the current computation power demand, including:
determining whether a local target scheduling node exists in each local node to be scheduled, wherein the local computing power resource of the local target scheduling node meets the current computing power requirement;
and if the local target scheduling node exists, determining a second target computing power node in the local target scheduling node, and calling computing power resources of the second target computing power node.
Specifically, the regional scheduling node determines whether a local target scheduling node exists in each local node to be scheduled, wherein the local computing resource meets the current computing power requirement.
And if the local target scheduling node exists in each local node to be scheduled, the regional scheduling node sends the current computational power requirement to the local target scheduling node.
Further, a decision module in the local target scheduling node determines a second target calculation force node in the local target scheduling node according to the current calculation force demand, and performs calculation force resource scheduling on the current calculation force demand through the second target calculation force node. That is, it can be understood that the decision module of the current scheduling node schedules the computing resources of the second target computing resource node in the local target scheduling node, which belong to the same scheduling region as the current scheduling node, through hierarchical scheduling.
The embodiment of the invention carries out calculation resource scheduling on the current calculation demand through the local target scheduling node which belongs to the same scheduling region as the current scheduling node, thereby realizing cross-region and cross-level scheduling of calculation resources. Meanwhile, by adopting a hierarchical scheduling mode, data is stored in a hierarchical and dispersed mode, and scheduling nodes are used for performing cooperative scheduling, so that the management problem, the safety problem and the compliance problem caused by data centralized storage adopted by an original scheduling mode are effectively solved.
Further, the step 103 of performing hierarchical scheduling of computing resources based on the cross-domain cooperative scheduling node and the current computing demand includes:
if the local target scheduling node does not exist, determining nodes to be scheduled in each association area associated with the area scheduling node based on the cross-domain cooperative scheduling node;
determining whether a related area target scheduling node exists in the nodes to be scheduled in the related area, wherein the area computing power resource of the related area target scheduling node meets the current computing power requirement;
and if the associated region target scheduling node exists, determining a third target computational power node in the associated region target scheduling node, and calling computational power resources of the third target computational power node.
It should be noted that cross-domain scheduling generally refers to a scenario in which several areas perform joint coordination, such as long triangle, jingjin Ji, and the like. The scheduling process is basically the same as the regional scheduling, a higher-level cross-domain collaborative scheduling node is responsible for forwarding the current computing power requirement in the associated region, and the cross-domain scheduling generally performs comprehensive analysis according to the aspects of price, resources, performance and the like to select the optimal computing power distribution mode for scheduling. If the region has no over calculation power, high-performance calculation needs to be carried out across the regions; or the user mainly saves expenditure, the calculation force request can be cooperatively scheduled by crossing domains, and the scheme with the most cost performance is selected.
After receiving the current computing power demand, the cross-domain cooperative scheduling node selects a target scheduling node meeting the demand according to the computing power state data of each node cached by the sensing module, forwards the current computing power demand to the target scheduling node, and further processes the current computing power demand by the target scheduling node. Meanwhile, in cross-domain scheduling, the current computing power requirement can be split according to the computing power resource matching condition, the split computing power requirement is divided into a plurality of subtasks to be issued to a plurality of target scheduling nodes for cooperative execution, the split result is notified to the access scheduling node, and a logic channel is directly established by the target scheduling node and the access scheduling node to perform message interaction and data transmission.
Specifically, if it is determined that no local target scheduling node exists in each local node to be scheduled, the regional scheduling node sends the current computational power requirement to the cross-domain cooperative scheduling node.
Further, the cross-domain collaborative scheduling node determines each associated region to-be-scheduled node associated with the region scheduling node, and determines that there is an associated region target scheduling node with region computing resources meeting the current computing power requirement in each associated region to-be-scheduled node.
Further, if it is determined that the associated region target scheduling node exists in the nodes to be scheduled in each associated region, the cross-domain cooperative scheduling node sends the current calculation requirement to the associated region target scheduling node.
And a decision module in the target scheduling node of the associated region determines a third target calculation force node in the target scheduling node of the associated region according to the current calculation force demand, and performs calculation force resource scheduling on the current calculation force demand through the third target calculation force node. That is, it can be understood that the decision module of the current scheduling node associates the computational resource of the third target computational resource node in the target scheduling nodes in the region by hierarchically scheduling the computational resources which belong to the same collaborative region as the current scheduling node.
In an embodiment, referring to fig. 7, fig. 7 is a schematic cross-domain scheduling diagram provided by the present invention, where a cross-domain cooperative scheduling node 1 includes a region scheduling node 1 and a region scheduling node 2, the region scheduling node 1 includes a local scheduling node 1 and a local scheduling node 2, and the region scheduling node 2 includes a local scheduling node 3 and a local scheduling node 4. And if the current computing power resource of the local scheduling node 1 does not meet the current computing power requirement, the current computing power requirement is sent to the regional scheduling node 1, and if the regional scheduling node 1 determines that the computing power resources of all the local scheduling nodes managed by the regional scheduling node do not meet the current computing power requirement, the current computing power requirement is sent to the superior cross-domain collaborative scheduling node 1.
The cross-domain cooperative scheduling node 1 determines the calculation resources of the local scheduling node 3 and the local scheduling node 4 in the regional scheduling node 2, determines that the calculation resources of the local scheduling node 4 meet the current calculation requirement, and performs calculation resource scheduling on the current calculation requirement through the local scheduling node 4. It can also be understood that the local scheduling node 1 schedules the computational demands to the computational resources of the cross-domain local scheduling node 4 by hierarchical scheduling.
The embodiment of the invention carries out calculation resource scheduling on the current calculation demand through the associated regional target scheduling node which belongs to the same collaborative region with the current scheduling node, thereby realizing cross-region and cross-level scheduling of calculation resources. Meanwhile, by adopting a hierarchical scheduling mode, data is stored in a hierarchical and dispersed mode, and scheduling nodes are used for performing cooperative scheduling, so that the management problem, the safety problem and the compliance problem caused by data centralized storage adopted by an original scheduling mode are effectively solved.
Further, the hierarchical scheduling of computing resources based on the global scheduling network and the current computing demand in step 103 includes:
if the associated regional target scheduling node does not exist, searching a global node to be scheduled through interconnected cross-domain cooperative scheduling nodes in a global range based on the global scheduling network;
determining whether a global target scheduling node exists in the global node to be scheduled, wherein the computing power resource of the global target scheduling node meets the current computing power requirement;
and if the global target scheduling node exists, determining a fourth target computing power node in the global target scheduling node, and calling computing power resources of the fourth target computing power node.
The global scheduling refers to scheduling and distributing current computing power requirements in a global wide computing power network, and makes full use of the advantages of the computing power network. The global scheduling network is interconnected through cross-domain collaborative scheduling nodes, after receiving the current calculation power requirement, the access scheduling node issues a request to the scheduling network, and after the request reaches the cross-domain collaborative scheduling node connected with the access scheduling node, the access cross-domain collaborative scheduling node broadcasts a message to other interconnected cross-domain collaborative scheduling nodes, and the current calculation power requirement is spread in the whole calculation power network. After receiving the current computing power demand, other cross-domain cooperative scheduling nodes analyze and make decisions according to the computing power data in the range, select computing power information meeting part or all of the current computing power demand and uniformly send the information to the access cross-domain cooperative scheduling nodes for screening.
Furthermore, after the decision of the access cross-domain cooperative scheduling node is completed, the decision results are respectively notified to the corresponding cross-domain cooperative scheduling nodes to forward in the range of the cross-domain cooperative scheduling nodes, and the whole scheduling execution process is monitored.
Specifically, if it is determined that a local target scheduling node does not exist in each local node to be scheduled, the cross-domain cooperative scheduling node sends the current calculation power requirement to the global scheduling network.
The universe scheduling network searches all universe nodes to be scheduled in the universe range through the interconnected cross-domain cooperative scheduling nodes, and determines whether universe nodes to be scheduled have universe target scheduling nodes with calculation resources meeting the current calculation requirements.
Further, if it is determined that a global target scheduling node with computational power resources meeting the current computational power requirement exists in the global node to be scheduled, the global scheduling network sends the current computational power requirement to the global target scheduling node through a cross-domain cooperative scheduling node.
And a decision module in the global target scheduling node determines a fourth target calculation power node in the global target scheduling node according to the current calculation power requirement, and performs calculation power resource scheduling on the current calculation power requirement through the fourth target calculation power node. That is, it can be understood that the decision module of the current scheduling node schedules the computing resources of the fourth target computing resource node in the global target scheduling nodes of the global scheduling network in a hierarchical manner.
If it is determined that there is no global target scheduling node whose computational resources meet the current computational demand in the global node to be scheduled, that is, the computational resources of all nodes in the global scheduling network do not meet the current computational demand, the resource scheduling system determines that the computational resource scheduling of the current computational demand fails.
The embodiment of the invention carries out calculation power resource scheduling on the current calculation power demand through the global target scheduling node of the global scheduling network, thereby realizing cross-region and cross-hierarchy scheduling of calculation power resources. Meanwhile, by adopting a hierarchical scheduling mode, data is stored in a hierarchical and dispersed mode, and scheduling nodes are used for performing cooperative scheduling, so that the management problem, the safety problem and the compliance problem caused by data centralized storage adopted by an original scheduling mode are effectively solved.
Furthermore, the computational resource hierarchical scheduling system provided by the invention and the computational resource hierarchical scheduling method provided by the invention are in mutual corresponding reference.
Fig. 8 is a schematic structural diagram of a computational resource hierarchical scheduling system provided in the present invention, where the computational resource hierarchical scheduling system includes:
a decision module 801, configured to determine whether a current computational resource of a current scheduling node meets a current computational demand according to local computational information and computational demand information of the current scheduling node;
a determining module 802, configured to determine an interconnection scheduling node, a local scheduling node, a cross-domain cooperative scheduling node, or a global scheduling network of the current scheduling node if the current computational resource does not meet the current computational demand;
a hierarchical scheduling module 803, configured to perform hierarchical scheduling of computational resources based on the interconnection scheduling node, the local scheduling node, the cross-domain cooperative scheduling node, or the global scheduling network, and the current computational demand.
Further, the hierarchical scheduling module 803 is further configured to:
determining distributable computing power resources of the interconnected scheduling nodes and determining whether the distributable computing power resources meet the current computing power requirement;
and if the distributable computing power resource meets the current computing power requirement, determining a first target computing power node in the interconnected scheduling nodes, and calling the computing power resource of the first target computing power node.
Further, the hierarchical scheduling module 803 is further configured to:
if the distributable computing power resource does not meet the current computing power requirement, determining whether computing power resource scheduling is allowed to go out of the local area;
if the local resource scheduling is allowed, determining each local node to be scheduled which belongs to the same scheduling region as the current scheduling node based on the region scheduling node;
and performing calculation power resource hierarchical scheduling based on each local node to be scheduled and the current calculation power requirement.
Further, the hierarchical scheduling module 803 is further configured to:
determining whether a local target scheduling node exists in each local node to be scheduled, wherein the local computing power resource of the local target scheduling node meets the current computing power requirement;
and if the local target scheduling node exists, determining a second target computing power node in the local target scheduling node, and calling computing power resources of the second target computing power node.
Further, the hierarchical scheduling module 803 is further configured to:
if the local target scheduling node does not exist, determining each associated region to-be-scheduled node associated with the region scheduling node based on the cross-domain cooperative scheduling node;
determining whether a related area target scheduling node exists in the nodes to be scheduled in the related area, wherein the area computing power resource of the related area target scheduling node meets the current computing power requirement;
and if the associated region target scheduling node exists, determining a third target computing power node in the associated region target scheduling node, and calling computing power resources of the third target computing power node.
Further, the hierarchical scheduling module 803 is further configured to:
if the associated regional target scheduling node does not exist, searching a global node to be scheduled through interconnected cross-domain cooperative scheduling nodes in a global range based on the global scheduling network;
determining whether a global target scheduling node exists in the global node to be scheduled, wherein the computing power resource of the global target scheduling node meets the current computing power requirement;
and if the global target scheduling node exists, determining a fourth target computational power node in the global target scheduling node, and calling computational power resources of the fourth target computational power node.
Further, the determining module 802 is further configured to:
and if the global target scheduling node does not exist, determining that the computing power resource scheduling of the current computing power requirement fails.
The specific embodiment of the hierarchical scheduling system for computing resources provided by the present invention is basically the same as the embodiments of the hierarchical scheduling method for computing resources described above, and is not described herein again.
Fig. 9 illustrates a physical structure diagram of an electronic device, and as shown in fig. 9, the electronic device may include: a processor (processor) 910, a communication Interface (Communications Interface) 920, a memory (memory) 930, and a communication bus 940, wherein the processor 910, the communication Interface 920, and the memory 930 are coupled for communication via the communication bus 940. Processor 910 may invoke logic instructions in memory 930 to perform a computational resource hierarchical scheduling method comprising:
determining whether the current computing power resource of the current scheduling node meets the current computing power requirement or not according to the local computing power information and the computing power requirement information of the current scheduling node;
if the current computing power resource does not meet the current computing power requirement, determining an interconnection scheduling node, a regional scheduling node, a cross-domain cooperative scheduling node or a global scheduling network of the current scheduling node;
and performing calculation resource hierarchical scheduling based on the interconnection scheduling node, the regional scheduling node, the cross-domain cooperative scheduling node or the universe scheduling network and the current calculation requirement.
Furthermore, the logic instructions in the memory 930 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention or a part thereof which substantially contributes to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk, and various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions, which when executed by a computer, enable the computer to perform the hierarchical scheduling method for computational resources provided by the above methods, the method comprising:
determining whether the current computing power resource of the current scheduling node meets the current computing power requirement or not according to the local computing power information and the computing power requirement information of the current scheduling node;
if the current computing power resource does not meet the current computing power requirement, determining an interconnection scheduling node, a regional scheduling node, a cross-domain cooperative scheduling node or a global scheduling network of the current scheduling node;
and performing calculation resource hierarchical scheduling based on the interconnection scheduling node, the regional scheduling node, the cross-domain cooperative scheduling node or the universe scheduling network and the current calculation requirement.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium, on which a computer program is stored, the computer program being implemented by a processor to execute the hierarchical scheduling method for computing resources provided by the above methods, the method including:
determining whether the current computing power resource of the current scheduling node meets the current computing power requirement or not according to the local computing power information and the computing power requirement information of the current scheduling node;
if the current computing power resource does not meet the current computing power requirement, determining an interconnection scheduling node, a regional scheduling node, a cross-domain cooperative scheduling node or a global scheduling network of the current scheduling node;
and performing hierarchical scheduling of computing power resources based on the interconnection scheduling node, the regional scheduling node, the cross-domain cooperative scheduling node or the global scheduling network and the current computing power requirement.
The above-described system embodiments are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding, the above technical solutions may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (personal computer, server, network device, etc.) to execute the methods according to the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, and not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for hierarchical scheduling of computing resources is characterized by comprising the following steps:
determining whether the current computing power resource of the current scheduling node meets the current computing power requirement or not according to the local computing power information and the computing power requirement information of the current scheduling node;
if the current computing power resource does not meet the current computing power requirement, determining an interconnection scheduling node, a regional scheduling node, a cross-domain cooperative scheduling node or a global scheduling network of the current scheduling node;
and performing hierarchical scheduling of computing power resources based on the interconnection scheduling node, the regional scheduling node, the cross-domain cooperative scheduling node or the global scheduling network and the current computing power requirement.
2. The hierarchical computing resource scheduling method according to claim 1, wherein performing hierarchical computing resource scheduling based on the interconnected scheduling nodes and the current computing demand comprises:
determining distributable computing resources of the interconnected scheduling nodes, and determining whether the distributable computing resources meet the current computing requirement;
and if the distributable computing power resource meets the current computing power requirement, determining a first target computing power node in the interconnected scheduling nodes, and calling the computing power resource of the first target computing power node.
3. The hierarchical power resource scheduling method according to claim 2, wherein performing hierarchical power resource scheduling based on the regional scheduling nodes and the current power demand comprises:
if the distributable computing power resource does not meet the current computing power requirement, determining whether computing power resource scheduling is allowed to go out of the local area;
if the local resource scheduling is allowed, determining each local node to be scheduled which belongs to the same scheduling region as the current scheduling node based on the region scheduling node;
and performing calculation power resource hierarchical scheduling based on each local node to be scheduled and the current calculation power requirement.
4. The hierarchical power resource scheduling method according to claim 3, wherein the performing power resource scheduling based on each local node to be scheduled and the current power demand comprises:
determining whether a local target scheduling node exists in each local node to be scheduled, wherein the local computing power resource of the local target scheduling node meets the current computing power requirement;
and if the local target scheduling node exists, determining a second target computing power node in the local target scheduling node, and calling computing power resources of the second target computing power node.
5. The hierarchical computing resource scheduling method according to claim 4, wherein performing hierarchical computing resource scheduling based on the cross-domain collaborative scheduling node and the current computing demand comprises:
if the local target scheduling node does not exist, determining each associated region to-be-scheduled node associated with the region scheduling node based on the cross-domain cooperative scheduling node;
determining whether each associated region target scheduling node exists in the nodes to be scheduled in the associated region, wherein the region computing power resource of the associated region target scheduling node meets the current computing power requirement;
and if the associated region target scheduling node exists, determining a third target computing power node in the associated region target scheduling node, and calling computing power resources of the third target computing power node.
6. The hierarchical computing resource scheduling method according to claim 5, wherein performing hierarchical computing resource scheduling based on the global scheduling network and the current computing demand comprises:
if the associated regional target scheduling node does not exist, searching a global node to be scheduled through interconnected cross-domain cooperative scheduling nodes in a global range based on the global scheduling network;
determining whether a global target scheduling node exists in the global node to be scheduled, wherein the computing power resource of the global target scheduling node meets the current computing power requirement;
and if the global target scheduling node exists, determining a fourth target computing power node in the global target scheduling node, and calling computing power resources of the fourth target computing power node.
7. The computational resource hierarchical scheduling method according to any one of claims 1 to 6, wherein after determining whether there is a global target scheduling node in the global nodes to be scheduled, further comprising:
and if the global target scheduling node does not exist, determining that the computing resource scheduling of the current computing requirement fails.
8. A hierarchical scheduling system for computing resources, comprising:
the decision module is used for determining whether the current computing power resource of the current scheduling node meets the current computing power requirement or not according to the local computing power information and the computing power requirement information of the current scheduling node;
a determining module, configured to determine an interconnection scheduling node, a local scheduling node, a cross-domain cooperative scheduling node, or a global scheduling network of the current scheduling node if the current computational resource does not meet the current computational demand;
and the hierarchical scheduling module is used for performing hierarchical scheduling on the computing resources based on the interconnection scheduling node, the regional scheduling node, the cross-domain cooperative scheduling node or the universe scheduling network and the current computing demand.
9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the computational resource hierarchical scheduling method of any one of claims 1 to 7 when executing the computer program.
10. A non-transitory computer readable storage medium comprising a computer program, wherein the computer program when executed by a processor implements the hierarchical scheduling method of computational resources of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211287320.0A CN115794373A (en) | 2022-10-20 | 2022-10-20 | Calculation force resource hierarchical scheduling method, system, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211287320.0A CN115794373A (en) | 2022-10-20 | 2022-10-20 | Calculation force resource hierarchical scheduling method, system, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115794373A true CN115794373A (en) | 2023-03-14 |
Family
ID=85433310
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211287320.0A Pending CN115794373A (en) | 2022-10-20 | 2022-10-20 | Calculation force resource hierarchical scheduling method, system, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115794373A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116302580A (en) * | 2023-05-25 | 2023-06-23 | 南方电网数字电网研究院有限公司 | Method and device for scheduling calculation force resources of nano relay |
| CN116436800A (en) * | 2023-06-13 | 2023-07-14 | 新华三技术有限公司 | Resource scheduling method and device |
| CN117749500A (en) * | 2023-12-25 | 2024-03-22 | 深业智慧科技(深圳)有限公司 | Intelligent park operation fusion communication method and system |
-
2022
- 2022-10-20 CN CN202211287320.0A patent/CN115794373A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116302580A (en) * | 2023-05-25 | 2023-06-23 | 南方电网数字电网研究院有限公司 | Method and device for scheduling calculation force resources of nano relay |
| CN116302580B (en) * | 2023-05-25 | 2023-09-15 | 南方电网数字电网研究院有限公司 | Method and device for scheduling calculation force resources of nano relay |
| CN116436800A (en) * | 2023-06-13 | 2023-07-14 | 新华三技术有限公司 | Resource scheduling method and device |
| CN116436800B (en) * | 2023-06-13 | 2023-09-19 | 新华三技术有限公司 | Resource scheduling method and device |
| CN117749500A (en) * | 2023-12-25 | 2024-03-22 | 深业智慧科技(深圳)有限公司 | Intelligent park operation fusion communication method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Aral et al. | A decentralized replica placement algorithm for edge computing | |
| CN115794373A (en) | Calculation force resource hierarchical scheduling method, system, electronic equipment and storage medium | |
| CN113448721A (en) | Network system for computing power processing and computing power processing method | |
| US8140666B2 (en) | Method and apparatus for network distribution and provisioning of applications across multiple domains | |
| US6665262B1 (en) | Distributed fault management architecture | |
| Al-Khafajiy et al. | Fog computing framework for internet of things applications | |
| CN105683918B (en) | Centralized networking configuration in distributed systems | |
| WO2023284830A1 (en) | Management and scheduling method and apparatus, node, and storage medium | |
| CN102045197B (en) | Alarm data synchronization method and network management system | |
| WO2019206100A1 (en) | Feature engineering programming method and apparatus | |
| Jazayeri et al. | A latency-aware and energy-efficient computation offloading in mobile fog computing: a hidden Markov model-based approach | |
| CN110602180B (en) | Big data user behavior analysis method based on edge calculation and electronic equipment | |
| CN115002862A (en) | Network system for computing power processing, service processing method and computing power network element node | |
| US11336536B1 (en) | Dynamic processing distribution for utility communication networks | |
| CN103795575A (en) | Multi-data-centre-oriented system monitoring method | |
| CN112994937A (en) | Deployment and migration system of virtual CDN in intelligent fusion identification network | |
| CN110855481B (en) | Data acquisition system and method | |
| Forestiero et al. | So-grid: A self-organizing grid featuring bio-inspired algorithms | |
| WO2020247713A1 (en) | Scalable hierarchical data automation in a network | |
| WO2022016969A1 (en) | Data processing method and device | |
| Gohar et al. | Minimizing the cost of 5G network slice broker | |
| Henna et al. | Distributed and collaborative high-speed inference deep learning for mobile edge with topological dependencies | |
| Ali et al. | Probabilistic normed load monitoring in large scale distributed systems using mobile agents | |
| Zeydan et al. | A multi-criteria decision making approach for scaling and placement of virtual network functions | |
| CN116684418A (en) | Calculation power arrangement scheduling method, calculation power network and device based on calculation power service gateway |
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 |