Resource allocation method, device, equipment, medium and product

Abstract

The invention provides a resource allocation method and device, equipment, a medium and a product, and relates to the field of network resource management. Comprising the following steps: acquiring measurement data of all network nodes in a to-be-distributed network system; splitting the measurement data according to the network slice dimension of the to-be-distributed network system to obtain a data queue corresponding to each network slice; according to a preset monitoring period, performing aggregation processing on the slice measurement data to obtain slice aggregation data of the network slice in the preset monitoring period; according to the slice aggregation data, calculating a resource violation amount of the to-be-allocated network system; and if the resource violation amount is greater than 0, triggering resource allocation of the to-be-allocated network system so as to generate a resource scheduling strategy for the network nodes in the to-be-allocated network system through the data queue. The technical problem of low resource allocation efficiency caused by low network resource utilization rate due to dependence on a static or semi-static strategy and poor flexibility in the prior art is solved.

Description

Technical Field

[0001] This application relates to the field of network resource management, and in particular to a resource allocation method, apparatus, equipment, medium and product. Background Technology

[0002] With the continuous evolution of mobile communication technology, software-defined networking, and network function virtualization, network slicing has become a core means to meet the needs of different business scenarios (such as enhanced mobile broadband, ultra-reliable low-latency communication, and massive machine-type communication) through logical isolation and multi-tenant management of network resources.

[0003] In existing network slicing resource allocation methods, static or semi-static strategies are mainly relied upon: traditional slice configuration allocates resources through predefined bandwidth and computing resources, and its slice scheduling strategy adjusts the processing speed of services based on priority through basic traffic awareness technology.

[0004] However, existing technologies rely on static or semi-static strategies, which are inflexible and result in low network resource utilization, leading to low resource allocation efficiency. Summary of the Invention

[0005] This application provides a resource allocation method, apparatus, device, medium, and product to solve the problem that existing technologies rely on static or semi-static strategies, which have poor flexibility, resulting in low network resource utilization and thus low resource allocation efficiency.

[0006] Firstly, this application provides a resource allocation method, including:

[0007] Obtain the metric data of all network nodes in the network system to be assigned;

[0008] The metric data is split according to the network slice dimension of the network system to be assigned, so as to obtain the data queue corresponding to each network slice; wherein, each data queue includes multiple slice metric data corresponding to the network slice.

[0009] According to the preset monitoring period, the slice measurement data is aggregated to obtain the slice aggregate data of the network slice within the preset monitoring period.

[0010] Calculate the number of resource violations in the network system to be allocated based on the sliced ​​aggregated data;

[0011] If the number of resource violations is greater than 0, resource allocation for the network system to be allocated is triggered, so as to generate a resource scheduling policy for the network nodes in the network system to be allocated through the data queue.

[0012] In one possible design, the metric data includes quintuple data, latency data, and packet loss counts.

[0013] In one possible design, the metric data is split according to the network slice dimension of the network system to be assigned, to obtain the data queue corresponding to each network slice, including:

[0014] Preprocessing is performed on the measurement data to obtain preprocessed measurement data; wherein, the preprocessing includes at least one of message unpacking, deduplication, and tagging.

[0015] Based on the network slice dimension of the network system to be assigned, identify the network slice corresponding to the preprocessed metric data.

[0016] Based on the network slices corresponding to the preprocessed metric data, the preprocessed metric data is split to obtain multiple slice metric data corresponding to each network slice.

[0017] Based on the multiple slice metric data corresponding to the network slice, determine the data queue corresponding to the network slice, and write the data queue to the time series database or message queue.

[0018] In one possible design, the slice metric data is aggregated according to a preset monitoring period to obtain aggregated slice data of the network slice within the preset monitoring period, including:

[0019] Based on the preset monitoring cycle, determine the sliding aggregation window and / or the fixed aggregation window;

[0020] Based on the sliding aggregation window and / or fixed aggregation window, the slice metric data is aggregated to obtain the original data of the slice set of network slices within the preset detection period.

[0021] The original slice aggregation data is subjected to performance quantization processing to obtain slice aggregation data; the slice aggregation data includes arrival rate, average queue length and average latency.

[0022] In one possible design, the number of resource violations in the network system to be allocated is calculated based on the sliced ​​aggregated data, including:

[0023] Based on the preset latency upper limit threshold and the corresponding average latency of the network slice, determine the number of resource violations of the network slice in the network system to be allocated.

[0024] In one possible design, network nodes include edge-side network nodes and core-side network nodes;

[0025] Accordingly, the resource allocation of the network system to be allocated is triggered, including:

[0026] Obtain the first slice metric data of the network slice at the edge-side network node and the second slice metric data of the network slice at the core-side network node;

[0027] Based on the first slice metric data, determine the edge node inflow rate, edge latency, and edge queue length of the network slice;

[0028] Based on the preset first objective function based on dual decomposition and Lagrange multipliers, the inflow rate of edge nodes, the edge delay, and the edge queue length, the edge Lagrange multipliers of the network slice are determined;

[0029] Obtain the local Lagrange multipliers of network slices at the edge-side network nodes;

[0030] Based on the second slice metric data, the total available resource information and core-side Lagrange multipliers of the core-side network nodes are calculated.

[0031] Based on the total available resource information, the core-side Lagrange multiplier, the edge Lagrange multiplier, and the local Lagrange multiplier, the resource information allocated to the network slice by the core-side network node is calculated.

[0032] Based on the resource information allocated to network slices by core-side network nodes, a resource allocation strategy is generated.

[0033] In one possible design, after generating the resource allocation strategy based on the resource information allocated to network slices by the core-side network nodes, the following is also included:

[0034] Resource allocation policies are issued via flow tables through the southbound application programming interface of the software-defined network controller to adjust traffic forwarding paths or bandwidth limits for network slices; and / or,

[0035] Resource allocation policies are issued through the elastic scaling interface of the Network Functions Virtualization Orchestrator to trigger the expansion / shrinkage of the number of virtual network function instances or to adjust the resource quota of network function virtualization.

[0036] In one possible design, both the elastic scaling interface of the network function virtualization orchestrator and the southbound application programming interface of the software-defined network controller use encrypted channels to implement policy distribution, and both the elastic scaling interface of the network function virtualization orchestrator and the southbound application programming interface of the software-defined network controller are configured with automatic certificate rotation cycles.

[0037] In one possible design, metric data for all network nodes in the network system to be assigned is obtained, including:

[0038] Obtain a list of all network nodes in the network system to be assigned;

[0039] Based on the node list, determine the location to deploy the traffic probes;

[0040] Deploy the traffic probe to the location where the traffic probe will be deployed;

[0041] A traffic probe is used to collect metric data from all network nodes in the network system to be assigned.

[0042] Secondly, this application provides a resource allocation device, comprising:

[0043] The acquisition module is used to acquire metric data for all network nodes in the network system to be assigned.

[0044] The splitting processing module is used to split the metric data according to the network slice dimension of the network system to be assigned, so as to obtain the data queue corresponding to each network slice; wherein, each data queue includes multiple slice metric data corresponding to the network slice;

[0045] The aggregation processing module is used to aggregate the slice measurement data according to the preset monitoring period to obtain the slice aggregate data of the network slice within the preset monitoring period.

[0046] The calculation module is used to calculate the number of resource violations in the network system to be allocated based on the sliced ​​aggregated data;

[0047] The allocation module is used to trigger resource allocation for the network system to be allocated if the number of resource violations is greater than 0, so as to generate a resource scheduling policy for the network nodes in the network system to be allocated through the data queue.

[0048] Thirdly, this application provides a resource allocation device, including: a memory and a processor;

[0049] The memory stores the instructions that the computer executes;

[0050] The processor executes computer execution instructions stored in memory, causing the processor to perform the resource allocation method as described in the first aspect of the invention.

[0051] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the resource allocation method as described in the first aspect of the invention.

[0052] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the resource allocation method described in the first aspect of the invention.

[0053] This application provides a resource allocation method, apparatus, device, medium, and product, comprising: acquiring measurement data of all network nodes in a network system to be allocated; splitting the measurement data according to the network slice dimension of the network system to be allocated to obtain a data queue corresponding to each network slice; aggregating the slice measurement data according to a preset monitoring period to obtain slice aggregated data of the network slice within the preset monitoring period; calculating the resource violation amount of the network system to be allocated based on the slice aggregated data; if the resource violation amount is greater than 0, triggering resource allocation of the network system to be allocated, thereby generating a resource scheduling strategy for network nodes in the network system to be allocated through the data queue. Compared with existing technologies that rely on static or semi-static strategies, which have poor flexibility and result in low network resource utilization, thus leading to low resource allocation efficiency, this application achieves flexible network resource scheduling through traffic awareness, slice quantification, violation detection, and resource scheduling, thereby improving network resource utilization and improving resource allocation efficiency. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 A schematic diagram of a system architecture for a resource allocation method provided in an embodiment of this application;

[0056] Figure 2 A schematic flowchart of a resource allocation method provided in this application embodiment. Figure 1 ;

[0057] Figure 3 A schematic flowchart of a resource allocation method provided in this application embodiment. Figure 2 ;

[0058] Figure 4 A schematic flowchart of a resource allocation method provided in this application embodiment. Figure 3 ;

[0059] Figure 5 This is a schematic diagram of the structure of the resource allocation device provided in the embodiments of this application;

[0060] Figure 6 This is a schematic diagram of the structure of a resource allocation device provided in an embodiment of this application. Detailed Implementation

[0061] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0062] In the embodiments of this application, the terms "first" and "second" are used to distinguish identical or similar items with substantially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, nor do they necessarily imply difference. It should be noted that in the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner. In the embodiments of this application, "at least one" refers to one or more, and "more than one" refers to two or more.

[0063] It should be noted that the phrase "at...time" in the embodiments of this application can refer to the instant at which a certain situation occurs, or to a period of time after the occurrence of a certain situation; the embodiments of this application do not specifically limit this. Furthermore, the resource allocation method provided in the embodiments of this application is merely an example, and resource allocation methods may include more or fewer elements.

[0064] With the continuous evolution of mobile communication technology, software-defined networking, and network function virtualization, network slicing has become an important means to meet diverse business needs and stringent quality assurance. In 5G and its evolved versions, network slicing, through logical isolation of network resources and multi-tenant management, enables the parallel deployment of different business scenarios such as enhanced mobile broadband, ultra-reliable low-latency communication, and massive machine-type communication on the same physical infrastructure.

[0065] Meanwhile, traffic awareness technology has also made significant progress in the field of network monitoring and management. Traffic identification methods based on deep packet inspection, time-series traffic analysis, and machine learning provide a theoretical foundation and practical path for dynamically understanding business characteristics and predicting network load trends. In terms of resource isolation and lifecycle management of network slicing, as well as traffic scheduling based on software-defined networking / network function virtualization, the industry has proposed various static or semi-static resource allocation strategies, and has completed initial deployment and functional verification in experimental networks and some commercial networks.

[0066] Existing network slicing resource allocation methods have the following main shortcomings:

[0067] On the one hand, traditional slicing configurations often rely on predefined bandwidth and computing resources, lacking real-time perception and closed-loop feedback of time-varying traffic, making it difficult to cope with sudden traffic fluctuations or sudden increases in service quality demands from the business side.

[0068] On the one hand, the resource utilization rate among multiple slices is low, and the allocation of downlink / uplink resources is prone to the "island effect", resulting in some slice resources being idle while other slices are saturated, thereby causing loss of business service quality requirements or waste of resources.

[0069] On the other hand, the lack of refined consideration of traffic attributes (such as quintuples, application categories, latency sensitivity, etc.) in the slice scheduling strategy makes it difficult to achieve dynamic acceleration of high-priority services and reliable deceleration of low-priority services, thus often facing trade-off problems when meeting SLA (Service Level Agreement).

[0070] To address the aforementioned problems, the inventors, during their research on the low efficiency of resource allocation, discovered that existing technologies rely on static or semi-static strategies, resulting in poor flexibility and low network resource utilization. Therefore, the inventors considered using traffic awareness, slice quantization, violation detection, and resource scheduling to achieve flexible network resource scheduling and improve network resource utilization. Based on this, embodiments of this application provide a resource allocation method, apparatus, device, medium, and product, applicable to the field of network resource management, aiming to solve the problem of low resource allocation efficiency in existing technologies.

[0071] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0072] Figure 1 This is a schematic diagram of a system architecture for a resource allocation method provided in an embodiment of this application. The resource allocation system is a computer device. Figure 1 In the above architecture, at least one of data acquisition device 101, processing device 102 and display device 103 is included.

[0073] It is understood that the structures illustrated in the embodiments of this application do not constitute a specific limitation on the processing system architecture of the resource allocation method. In other feasible embodiments of this application, the above architecture may include more or fewer components than illustrated, or combine some components, or divide some components, or arrange different components, which can be determined according to the actual application scenario and is not limited here. Figure 1The components shown can be implemented in hardware, software, or a combination of both.

[0074] In the specific implementation process, the data acquisition device 101 may include an input / output interface or a communication interface. The data acquisition device 101 can be connected to the processing device through the input / output interface or the communication interface to obtain the measurement data of all network nodes in the network system to be allocated.

[0075] The processing device 102 can generate a resource scheduling strategy for the network nodes in the network system to be allocated based on the metric data of all network nodes in the network system to be allocated.

[0076] The display device 103 can also be a touch screen or the screen of a terminal device, used to receive user commands while displaying the above-mentioned content, so as to realize interaction with the user.

[0077] It should be understood that the aforementioned processing device can be implemented by a processor reading instructions from memory and executing those instructions, or it can be implemented by a chip circuit.

[0078] Furthermore, the network architecture and business scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0079] The technical solution of this application will be described in detail below with reference to specific embodiments:

[0080] Figure 2 A schematic flowchart of a resource allocation method provided in this application embodiment. Figure 1 ,like Figure 2 As shown, the method includes:

[0081] S201. Obtain the measurement data of all network nodes in the network system to be assigned.

[0082] The metrics include quintuple data, latency data, and packet loss counts.

[0083] It should be noted that the following steps are included after step S201:

[0084] Specifically, P4 probes or eBPF probes are deployed on each switching node or virtual network element, according to a fixed period. (e.g., 100ms) or trigger threshold over-limit mode, the captured quintuple information, queue depth, packet loss count and other raw metrics are pushed to the telemetry aggregation layer through the remote procedure call framework or distributed stream processing platform.

[0085] S202. Based on the network slice dimension of the network system to be assigned, the metric data is split to obtain the data queue corresponding to each network slice.

[0086] Each data queue contains multiple slice metric data corresponding to a network slice.

[0087] Specifically, the measurement data is preprocessed to obtain preprocessed measurement data.

[0088] Preprocessing includes at least one of message unpacking, deduplication, and tagging.

[0089] Specifically, based on the network slice dimension of the network system to be assigned, the network slice corresponding to the preprocessed metric data is identified.

[0090] Specifically, based on the network slices corresponding to the preprocessed metric data, the preprocessed metric data is split to obtain multiple slice metric data corresponding to each network slice.

[0091] Specifically, based on the multiple slice metric data corresponding to the network slice, the data queue corresponding to the network slice is determined, and the data queue is written to the time series database or message queue.

[0092] In this embodiment, specifically, the telemetry aggregation layer performs message unpacking, deduplication, and tagging (slice identifier, virtual terminal network instance identifier) ​​on the reported data, and stores it in a time-series database, message queue, or memory queue.

[0093] In addition, preprocessing includes traffic feature preprocessing, which classifies the original packet traffic according to the slice dimension: all packets belonging to the same network slice s and within the time window The data packets arriving within the region are counted as a set. .

[0094] It should be noted that by preprocessing the metric data based on the network slice dimension (such as unpacking, deduplication, and labeling) and accurately splitting it into each slice data queue and writing it into a time-series database or message queue, standardized storage, efficient querying, and real-time analysis capabilities of network slice metrics are achieved, thereby supporting slice-level performance monitoring, resource optimization, and rapid location of fault root causes.

[0095] S203. According to the preset monitoring period, the slice measurement data is aggregated to obtain the slice aggregate data of the network slice within the preset monitoring period.

[0096] Specifically, based on the preset monitoring cycle, a sliding aggregation window and / or a fixed aggregation window are determined.

[0097] Specifically, the slice metric data is aggregated based on a sliding aggregation window and / or a fixed aggregation window to obtain the original slice set data of the network slice within a preset detection period.

[0098] Specifically, the raw data of the slice aggregation is subjected to performance quantization processing to obtain slice aggregated data; the slice aggregated data includes arrival rate, average queue length and average latency.

[0099] In this embodiment, a sliding window or fixed window method is used for aggregation. After each window ends, the cache is cleared to prepare for the next period of statistics.

[0100] Among them, real-time performance quantization calculates the arrival rate. The formula for dividing the number of slices *s* within a window period by the window length is:

[0101]

[0102] Among them, the average queue length is calculated. : Queue depth reported by all probe nodes m The formula for calculating the arithmetic mean is:

[0103]

[0104] Among them, the average delay is calculated. The formula for averaging the inbound and outbound timestamp differences of all data packets within the same window is as follows:

[0105]

[0106] In this embodiment, by using a sliding / fixed aggregation window based on a preset monitoring period to aggregate and quantify slice metric data (such as calculating arrival rate, average queue length, and average latency), the key performance indicators of network slices within the period are accurately quantified and comprehensively evaluated, thereby supporting the refinement of slice-level performance monitoring, the scientific nature of resource optimization, and the speed of fault location.

[0107] S204. Calculate the number of resource violations in the network system to be allocated based on the slice aggregation data.

[0108] Specifically, the number of resource violations of the network slice in the network system to be allocated is determined based on the preset latency upper limit threshold and the corresponding average latency of the network slice.

[0109] For example, performing Service Level Agreement Violation Detection (SLA Violation Detection): setting a latency cap for slice s. Calculate the number of violations The formula is expressed as:

[0110]

[0111] Among them, if If this occurs, it will trigger the upper-level control strategy or alarm mechanism.

[0112] In one possible embodiment, the specific algorithm flow includes:

[0113] First, initialization: setting the monitoring cycle. Initialize the time-series database and message queue.

[0114] Secondly, each probe is activated and connected to the telemetry aggregation layer to begin real-time reporting of the quintuple and queue status.

[0115] Next, the aggregation layer groups the received metric data by slice identifier and, in each slice... Internal summary.

[0116] Next, calculate the value of each slice. , and .

[0117] Next, a violation judgment is made, and the violation is compared. and ,get .

[0118] Then, the above metrics are stored in the database and pushed to the strategy engine or large-screen monitoring platform for subsequent decision-making.

[0119] Finally, after the window closes, clear the temporary cache and proceed to the next cycle.

[0120] S205. If the number of resource violations is greater than 0, the resource allocation of the network system to be allocated is triggered, so as to generate a resource scheduling policy for the network nodes in the network system to be allocated through the data queue.

[0121] This embodiment provides a resource allocation method, comprising: acquiring metric data of all network nodes in a network system to be allocated; splitting the metric data according to the network slice dimension of the network system to be allocated to obtain a data queue corresponding to each network slice; aggregating the slice metric data according to a preset monitoring period to obtain slice aggregate data of the network slice within the preset monitoring period; calculating the resource violation count of the network system to be allocated based on the slice aggregate data; if the resource violation count is greater than 0, triggering resource allocation for the network system to be allocated, thereby generating a resource scheduling strategy for network nodes in the network system to be allocated through the data queue. Compared with existing technologies that rely on static or semi-static strategies, which have poor flexibility and result in low network resource utilization, leading to low resource allocation efficiency, this application achieves flexible network resource scheduling through traffic awareness, slice quantification, violation detection, and resource scheduling, thereby improving network resource utilization and thus enhancing resource allocation efficiency.

[0122] Figure 3 A schematic flowchart of a resource allocation method provided in this application embodiment. Figure 2 ,like Figure 3 As shown, network nodes include edge-side network nodes and core-side network nodes; therefore, step S205 specifically includes:

[0123] It should be noted that step S205 is performed on the edge computing node side, and its cycle is... .

[0124] Among them, the edge side refers to the edge computing node side.

[0125] S301. Obtain the first slice metric data of the network slice on the edge-side network node and the second slice metric data of the network slice on the core-side network node.

[0126] S302. Based on the first slice metric data, determine the edge node inflow rate, edge latency, and edge queue length of the network slice.

[0127] For example, based on the first slice metric data, the inflow rate of edge nodes in slice s is obtained from a local programmable probe. Edge latency and edge queue length .

[0128] S303. Based on the preset first objective function based on dual decomposition and Lagrange multipliers, the inflow rate of edge nodes, the edge delay, and the edge queue length, determine the edge Lagrange multipliers of the network slice.

[0129] Specifically, local capacity budgeting reserves total available resources on edge computing nodes. Based on the dual decomposition concept, local processing only considers the "edge capacity". Optimize the allocation to satisfy the formula:

[0130]

[0131] Furthermore, local iterative optimization utilizes locally summarized data and the marginal Lagrange multipliers obtained from the previous iteration. Quick solution:

[0132]

[0133] S304. Obtain the local Lagrange multipliers of the network slice at the edge-side network nodes.

[0134] In this embodiment, the marginal Lagrange multipliers are based on the above-solved method. Update local Lagrange multipliers Obtain the local Lagrange multipliers of the network slices at the edge-side network nodes.

[0135] S305. Based on the second slice metric data, calculate the total available resource information and core-side Lagrange multipliers of the core-side network nodes.

[0136] It should be noted that the core side refers to the data center side.

[0137] S306. Based on the total available resource information, the core-side Lagrange multiplier, the edge Lagrange multiplier, and the local Lagrange multiplier, calculate the resource information allocated to the network slice by the core-side network node.

[0138] Specifically, the data collection core control center periodically (e.g., every 5 minutes) collects data reported by all edge computing nodes. , and the local Lagrange multipliers of each edge computing node .

[0139] Furthermore, the global dual iteration will slice the overall requirements at the core node: Lagrange multipliers on the core side of the previous moment Integrate both into the dual framework and update the core-side network node allocation. With Grandes multipliers The calculation formula is:

[0140]

[0141] S307. Generate a resource allocation strategy based on the resource information allocated to network slices by the core-side network nodes.

[0142] In one possible embodiment, the algorithm description process of step S205 is as follows:

[0143] Optionally, a dual decomposition is performed, and the original optimization objective formula is expressed as:

[0144]

[0145] The constraints are as follows:

[0146]

[0147]

[0148] Optional, Lagrange function:

[0149]

[0150] in, .

[0151] Optionally, during initialization, Take the smallest positive number and let each It can be evenly distributed.

[0152] Optionally, the local Lagrange multiplier problem is solved at the edge computing nodes for each slice. :

[0153]

[0154] Optional, update edge Lagrange multipliers:

[0155]

[0156] Optional, data collection on the data center side for solving the core Lagrange multiplier problem. Then, calculate the total population. The formula is expressed as:

[0157]

[0158] Optional, update the core Lagrange multipliers:

[0159]

[0160] Optional, convergence determination, if and All below the threshold Stop iteration; otherwise, return to resolve the edge computing node pair for each slice s based on the local Lagrange multiplier problem.

[0161] It should be noted that after step S307, the following steps are also included:

[0162] Optionally, resource allocation policies can be issued through the southbound application programming interface of the software-defined network controller to adjust the traffic forwarding path or bandwidth limit of the network slice.

[0163] In this embodiment, the new , The application programming interface (API) is deployed to each edge computing node and the core network function virtualization platform to drive the elastic scaling of virtual network functions and the updating of software-defined network flow tables.

[0164] It should be noted that local execution: the edge computing node receives... Immediately afterwards, adjust the queue weights and bandwidth limits on the software-defined network / virtual switch.

[0165] It should be noted that global correction: core nodes receive local execution feedback (actual throughput and latency) for the next round of dual iteration, ensuring that the allocation of edge and core resources converges to the optimal balance state in the long term.

[0166] It should also be noted that high-availability network paths and accelerated redundant links are configured on each critical link. Multiple backup paths are pre-configured, and multi-protocol label switching traffic engineering is used to achieve rapid switching.

[0167] Specifically, the path switching algorithm includes: continuously monitoring the link packet loss rate. With delay .

[0168] Among them, if or Immediately reroute the traffic to the suboptimal path.

[0169] in, This represents the packet loss rate threshold. This indicates the time delay threshold.

[0170] Furthermore, and / or, resource allocation policies can be issued through the elastic scaling interface of the network function virtualization orchestrator to trigger the expansion / shrinkage of the number of virtual network function instances or adjust the resource quota of network function virtualization.

[0171] The network function virtualization orchestrator's elastic scaling interface and the software-defined network controller's southbound application programming interface both use encrypted channels to implement policy distribution, and the network function virtualization orchestrator's elastic scaling interface and the software-defined network controller's southbound application programming interface are configured with automatic certificate rotation cycles.

[0172] For example, to ensure that the overall system can still operate stably and reliably in the face of network jitter, node failure or security attacks, two aspects need to be addressed: first, component redundancy and failover; and second, communication encryption and certificate management.

[0173] Optional, for component redundancy and failover:

[0174] Specifically, a clustered deployment is carried out: a software-defined network controller and network function virtualization orchestrator cluster, with r instances (nodes) deployed on different physical / virtual machines to form a cluster.

[0175] Furthermore, nodes elect a leader through a consensus algorithm. The leader is responsible for receiving instructions from the upper layer and issuing flow tables or scaling commands. The remaining nodes act as followers and accept state replication from the leader.

[0176] Where p is the average annual availability of a single node (e.g., 0.99). r is the number of redundant replicas per node. A is the overall cluster availability, satisfying:

[0177]

[0178] Where the target availability rate is Then take

[0179]

[0180] Optional health check and failover algorithm:

[0181] Specifically, heart rate monitoring is performed every [time period]. (For example, 2 seconds) The Follower sends a heartbeat request to the Leader or the monitoring component sends a heartbeat request to all nodes; if a node does not respond for K consecutive times, it is determined that the node is offline.

[0182] in, is the heartbeat interval period. K is the threshold for consecutive lost heartbeats.

[0183] Furthermore, once a Leader anomaly is detected during failover (K ​​heartbeat timeouts), the remaining nodes execute the Leader election algorithm:

[0184] Specifically, each node waits for an election timeout period randomly. (For example, [150ms, 300ms]), the node that completes the wait first initiates a vote, and becomes the new leader after obtaining more than half of the votes.

[0185] in, Indicates the timeout period for the election.

[0186] Optional: Communication encryption and certificate management:

[0187] Specifically, a handshake process for establishing a session management interface is initiated through a two-way transport layer security protocol. The client initiates a client greeting, which includes a list of supported transport layer security protocol versions and encryption suites.

[0188] Furthermore, the server returns a server greeting and provides its own certificate (including the public key) and the selected encryption suite.

[0189] Furthermore, the two parties exchange messages regarding the key exchange, using the server's public key to encrypt the pre-master secret.

[0190] Furthermore, both parties derive their own master secrets and generate symmetric session keys.

[0191] Furthermore, the client and server mutually verify the end message, confirming that the handshake was successful.

[0192] Session reuse, for nodes with frequent communication, enables session recovery through session identifiers or session tickets, avoiding a full handshake on every occasion and reducing the average handshake latency. Reduced to:

[0193]

[0194] Optional certificate lifecycle management strategy:

[0195] Specifically, the automatic rotation algorithm is as follows: Periodic checks are performed, every [time period]... (e.g., 24 hours) Scan the remaining validity period of the local certificate. .

[0196] Among them, if (For example If the remaining period is less than 30%, the certificate renewal process will be triggered.

[0197] Specifically, domain / node authentication and new certificates are obtained through an automated certificate management environment protocol.

[0198] Specifically, a smooth replacement includes: loading the new certificate onto the transport layer security protocol port that is being listened to, and discarding the old certificate after a hot reload to ensure uninterrupted service.

[0199] in, The maximum validity period of the certificate (e.g., 90 days). This refers to the certificate inspection cycle. This is the certificate rotation trigger coefficient.

[0200] It should be noted that through the aforementioned redundant deployment, heartbeat fault detection, leader election, encrypted handshake and automatic certificate rotation, as well as rapid multi-path switching, comprehensive reliability and security guarantees are formed for the system control plane, orchestration plane, and data plane. All of the above parameters can be adjusted according to the actual business scale and service level agreement violation detection requirements to achieve the optimal balance between performance, cost, and availability.

[0201] In this embodiment, a resource allocation strategy is generated by combining dual decomposition and Lagrange multiplier optimization calculations based on edge and core-side dual-dimensional slice metric data. This strategy is then transmitted via an encrypted channel (with automatic certificate rotation) through the southbound application programming interface of the software-defined network controller and the elastic scaling interface of the network function virtualization orchestrator. This achieves precise dynamic allocation of network slice resources at the edge and core sides, intelligent optimization of traffic paths, and secure and reliable policy execution, thereby improving network resource utilization, service quality assurance capabilities, and the overall system's security resilience. This ultimately enhances resource allocation efficiency.

[0202] Figure 4 A schematic flowchart of a resource allocation method provided in this application embodiment. Figure 3 ,like Figure 4 As shown, the specific implementation steps of S201 above include:

[0203] S401. Obtain a list of all network nodes in the network system to be assigned.

[0204] In this embodiment, a list of all network nodes is obtained from the network management system or operation and maintenance database.

[0205] The list of all network nodes includes: all physical devices (switches, routers, servers, etc.) and a list of edge computing nodes (Multi-Access Edge Computing, MEC).

[0206] S402. Based on the node list, determine the location to deploy the traffic probes.

[0207] In this embodiment, the locations where traffic probes can be deployed are determined based on the node list (including programmable switching chips on physical switches, virtual switches, and host network cards running virtual terminal networks).

[0208] Furthermore, the above equipment list is mapped to a set of nodes V in a graph model, including edge nodes. Core nodes and measurement points .

[0209] The equipment list includes all network devices.

[0210] The network devices include at least one of switches, routers, servers, and edge computing nodes.

[0211] Each network device corresponds to one node—an edge node. Core nodes and measurement points .

[0212] Furthermore, by using the link discovery protocol or configuration file, the physical connections between each device and their bandwidth and latency information are obtained, and the edge set E in the graph model is constructed.

[0213] S403. Deploy the traffic probe to the location where the traffic probe is deployed.

[0214] The flow probe is a programmable probe.

[0215] The flow probes are P4 probes, eBPF probes, or eXDP probes.

[0216] Optionally, when the traffic probe is a P4 probe, on a switch that supports the P4 runtime control interface (P4Runtime), the probe program is compiled by the P4 compiler (P4Compiler) and sent to the switching chip. Then, the gRPC communication channel registry entry is used to match the five-tuple and count information such as queue length and packet loss count.

[0217] Optionally, when the traffic probe is an eBPF probe or an eXDP probe, a fast data path program is loaded in the Linux host network card driver layer, compiled using the underlying virtual machine toolchain, to capture network card ingress packets, extract metrics such as 5-tuples, queue latency, and packet loss count in real time, and pass them to the user-space daemon process through a performance buffer or a ring buffer.

[0218] It should be noted that in a network slicing environment, P4 probes, eBPF probes, or eXDP probes are integrated to simultaneously collect multi-dimensional, fine-grained traffic characteristics such as quintuples, queue length, and packet loss rate. This enables sub-millisecond-level traffic telemetry reporting, providing highly timely and accurate input for closed-loop scheduling.

[0219] S404. Use a traffic probe to collect measurement data of all network nodes in the network system to be assigned.

[0220] Specifically, a lightweight daemon (implemented in Go or C) runs at each measurement point, which is responsible for collecting measurement data from the probe program and sending the data to the central aggregation layer through a remote procedure call framework, a distributed stream processing platform, or a hypertext transfer protocol according to the configured reporting policy.

[0221] Furthermore, the central aggregation layer standardizes and tags the received multi-source time-series data (slice identifier, node identifier, timestamp), and writes it into a time-series database or message queue for subsequent real-time analysis.

[0222] Furthermore, export the device list from the network management system. and link list .

[0223] Specifically, for each device :

[0224] Optionally, if its role is an edge computing node, then add it as an edge node. .

[0225] Optionally, if it is a data center core switch / router, then add it to the core node. .

[0226] Optionally, if probes are configured, measurement points are added. .

[0227] Furthermore, for each link Read bandwidth attributes and latency attributes And add it to the edge set E.

[0228] Furthermore, deploy the probe mapping function, defining the mapping function as follows:

[0229]

[0230] in, This is the set of all data packets that have passed through node m.

[0231] Specifically, implement it in P4 or eBPF programs. When packet p arrives, it is simultaneously extracted and buffered. The quintuple, queue length Packet loss count And timestamp.

[0232] It should be noted that the metric reporting strategy is as follows:

[0233] Specifically, configure the periodic reporting interval. Every (For example, 100ms), the probe daemon will summarize the accumulated metrics into a vector according to the slice identifier. and call: It is then pushed into the telemetry aggregation layer.

[0234] Furthermore, configure threshold trigger sets. When a metric exceeds a corresponding threshold (such as queue length) or packet loss rate When this occurs, an immediate report is triggered.

[0235] In this embodiment, by systematically acquiring the list of network system nodes to be allocated and accurately determining the deployment location of traffic probes, full-coverage collection and real-time monitoring of network node metrics are achieved, thereby improving network performance optimization, rapid fault location, and security situation awareness capabilities. This ultimately enhances resource allocation efficiency.

[0236] Figure 5 This is a schematic diagram of the structure of the resource allocation device provided in the embodiments of this application, such as... Figure 5 As shown, the device includes: an acquisition module 51, a splitting processing module 52, an aggregation processing module 53, a calculation module 54, and an allocation module 55.

[0237] The acquisition module 51 is used to acquire the measurement data of all network nodes in the network system to be assigned;

[0238] The splitting processing module 52 is used to split the metric data according to the network slice dimension of the network system to be assigned, so as to obtain the data queue corresponding to each network slice; wherein, each data queue includes multiple slice metric data corresponding to the network slice;

[0239] The aggregation processing module 53 is used to aggregate the slice measurement data according to the preset monitoring period to obtain the slice aggregate data of the network slice within the preset monitoring period.

[0240] Calculation module 54 is used to calculate the number of resource violations in the network system to be allocated based on the slice aggregation data;

[0241] The allocation module 55 is used to trigger resource allocation for the network system to be allocated if the number of resource violations is greater than 0, so as to generate a resource scheduling strategy for the network nodes in the network system to be allocated through the data queue.

[0242] In one possible design, the metric data includes quintuple data, latency data, and packet loss counts.

[0243] In one possible design, the metric data is split according to the network slice dimension of the network system to be assigned, to obtain the data queue corresponding to each network slice, including:

[0244] The splitting processing module 52 is also used to preprocess the measurement data to obtain preprocessed measurement data; wherein, the preprocessing includes at least one of message unpacking processing, deduplication processing and tagging processing;

[0245] Based on the network slice dimension of the network system to be assigned, identify the network slice corresponding to the preprocessed metric data.

[0246] Based on the network slices corresponding to the preprocessed metric data, the preprocessed metric data is split to obtain multiple slice metric data corresponding to each network slice.

[0247] Based on the multiple slice metric data corresponding to the network slice, determine the data queue corresponding to the network slice, and write the data queue to the time series database or message queue.

[0248] In one possible design, the slice metric data is aggregated according to a preset monitoring period to obtain aggregated slice data of the network slice within the preset monitoring period, including:

[0249] The aggregation processing module 53 is also used to determine the sliding aggregation window and / or the fixed aggregation window according to the preset monitoring cycle;

[0250] Based on the sliding aggregation window and / or fixed aggregation window, the slice metric data is aggregated to obtain the original data of the slice set of network slices within the preset detection period.

[0251] The original slice aggregation data is subjected to performance quantization processing to obtain slice aggregation data; the slice aggregation data includes arrival rate, average queue length and average latency.

[0252] In one possible design, the number of resource violations in the network system to be allocated is calculated based on the sliced ​​aggregated data, including:

[0253] The calculation module 54 is also used to determine the number of resource violations of the network slice in the network system to be allocated based on the preset latency upper limit threshold and the corresponding average latency of the network slice.

[0254] In one possible design, network nodes include edge-side network nodes and core-side network nodes;

[0255] Accordingly, the resource allocation of the network system to be allocated is triggered, including:

[0256] The allocation module 55 is also used to obtain the first slice metric data of the network slice at the edge-side network node and the second slice metric data at the core-side network node;

[0257] Based on the first slice metric data, determine the edge node inflow rate, edge latency, and edge queue length of the network slice;

[0258] Based on the preset first objective function based on dual decomposition and Lagrange multipliers, the inflow rate of edge nodes, the edge delay, and the edge queue length, the edge Lagrange multipliers of the network slice are determined;

[0259] Obtain the local Lagrange multipliers of network slices at the edge-side network nodes;

[0260] Based on the second slice metric data, the total available resource information and core-side Lagrange multipliers of the core-side network nodes are calculated.

[0261] Based on the total available resource information, the core-side Lagrange multiplier, the edge Lagrange multiplier, and the local Lagrange multiplier, the resource information allocated to the network slice by the core-side network node is calculated.

[0262] Based on the resource information allocated to network slices by core-side network nodes, a resource allocation strategy is generated.

[0263] In one possible design, after generating the resource allocation strategy based on the resource information allocated to network slices by the core-side network nodes, the following is also included:

[0264] Resource allocation policies are issued via flow tables through the southbound application programming interface of the software-defined network controller to adjust traffic forwarding paths or bandwidth limits for network slices; and / or,

[0265] Resource allocation policies are issued through the elastic scaling interface of the Network Functions Virtualization Orchestrator to trigger the expansion / shrinkage of the number of virtual network function instances or to adjust the resource quota of network function virtualization.

[0266] In one possible design, both the elastic scaling interface of the network function virtualization orchestrator and the southbound application programming interface of the software-defined network controller use encrypted channels to implement policy distribution, and both the elastic scaling interface of the network function virtualization orchestrator and the southbound application programming interface of the software-defined network controller are configured with automatic certificate rotation cycles.

[0267] In one possible design, metric data for all network nodes in the network system to be assigned is obtained, including:

[0268] The acquisition module 51 is also used to acquire a list of all network nodes in the network system to be assigned;

[0269] Based on the node list, determine the location to deploy the traffic probes;

[0270] Deploy the traffic probe to the location where the traffic probe will be deployed;

[0271] A traffic probe is used to collect metric data from all network nodes in the network system to be assigned.

[0272] The resource allocation device provided in this embodiment can execute a resource allocation method of the above embodiment. Its implementation principle and technical effect are similar, and will not be described again here.

[0273] In a specific implementation of the aforementioned resource allocation method, each module can be implemented as a processor. The processor can execute computer execution instructions stored in the memory, thereby enabling the processor to execute the aforementioned resource allocation method.

[0274] Figure 6 This is a schematic diagram of a resource allocation device provided in an embodiment of this application. Figure 6 As shown, the resource allocation device 60 includes at least one processor 61 and a memory 62. The resource allocation device 60 also includes a communication component 63. The processor 61, memory 62, and communication component 63 are connected via a bus 64.

[0275] In the specific implementation process, at least one processor 61 executes computer execution instructions stored in memory 62, causing at least one processor 61 to execute a method in the field of network resource management as executed by the resource allocation device as described above.

[0276] The specific implementation process of processor 61 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0277] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0278] The memory may include high-speed RAM, and may also include non-volatile storage (NVM), such as at least one disk storage.

[0279] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0280] The above description of the functions implemented by the resource allocation device and the main control device illustrates the solutions provided by the embodiments of the present invention. It is understood that, in order to achieve the above functions, the resource allocation device or the main control device includes hardware structures and / or software modules corresponding to the execution of each function. By combining the units and algorithm steps of the various examples described in the embodiments of the present invention, the embodiments of the present invention can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the technical solutions of the embodiments of the present invention.

[0281] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method in the field of network resource management as described above.

[0282] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0283] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in a resource allocation device or a master control device.

[0284] This application also provides a computer program product, comprising: a computer program stored in a readable storage medium, at least one processor of a resource allocation device being able to read the computer program from the readable storage medium, and the at least one processor executing the computer program causing the resource allocation device to perform the scheme provided in any of the above embodiments.

[0285] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disk, or optical disk.

[0286] The technical solutions of this application have been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of this application is obviously not limited to these specific embodiments. The above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A resource allocation method, characterized in that, include: Obtain the metric data of all network nodes in the network system to be assigned; The metric data is split according to the network slice dimension of the network system to be assigned, so as to obtain the data queue corresponding to each network slice; wherein, each data queue includes multiple slice metric data corresponding to the network slice; According to a preset monitoring period, the slice measurement data is aggregated to obtain the slice aggregate data of the network slice within the preset monitoring period; Based on the slice aggregation data, calculate the resource violation amount of the network system to be allocated; If the number of resource violations is greater than 0, the resource allocation of the network system to be allocated is triggered, so as to generate a resource scheduling strategy for the network nodes in the network system to be allocated through the data queue.

2. The method according to claim 1, characterized in that, The metrics include quintuple data, latency data, and packet loss counts.

3. The method according to claim 2, characterized in that, The step of splitting the metric data according to the network slice dimension of the network system to be assigned, to obtain the data queue corresponding to each network slice, includes: The measurement data is preprocessed to obtain preprocessed measurement data; wherein the preprocessing includes at least one of message unpacking, deduplication, and tagging. Based on the network slice dimension of the network system to be assigned, identify the network slice corresponding to the preprocessed metric data; Based on the network slices corresponding to the preprocessed metric data, the preprocessed metric data is split to obtain multiple slice metric data corresponding to each network slice. Based on the multiple slice metric data corresponding to the network slice, determine the data queue corresponding to the network slice, and write the data queue to a time-series database or a message queue.

4. The method according to claim 2, characterized in that, The step of aggregating the slice metric data according to a preset monitoring period to obtain the slice aggregated data of the network slice within the preset monitoring period includes: Based on the preset monitoring cycle, determine the sliding aggregation window and / or the fixed aggregation window; The slice metric data is aggregated according to the sliding aggregation window and / or fixed aggregation window to obtain the original data of the slice set of the network slice within the preset detection period; The original data of the slice aggregation is subjected to performance quantization processing to obtain slice aggregated data; wherein, the slice aggregated data includes arrival rate, average queue length and average latency.

5. The method according to claim 4, characterized in that, The step of calculating the resource violation amount of the network system to be allocated based on the slice aggregation data includes: Based on the preset latency upper limit threshold and the corresponding average latency of the network slice, the resource violation amount of the network slice in the network system to be allocated is determined.

6. The method according to any one of claims 2 to 5, characterized in that, The network nodes include edge-side network nodes and core-side network nodes; Accordingly, triggering the resource allocation of the network system to be allocated includes: Obtain the first slice metric data of the network slice at the edge-side network node and the second slice metric data at the core-side network node; Based on the first slice metric data, the edge node inflow rate, edge latency, and edge queue length of the network slice are determined; The edge Lagrange multipliers of the network slice are determined based on a preset first objective function based on dual decomposition and Lagrange multipliers, the inflow rate of the edge node, the edge delay, and the edge queue length. Obtain the local Lagrange multipliers of the network slice at the edge-side network nodes; Based on the second slice metric data, the total available resource information and core-side Lagrange multiplier of the core-side network node are calculated. Based on the total available resource information, the core-side Lagrange multiplier, the edge Lagrange multiplier, and the local Lagrange multiplier, the resource information allocated by the core-side network node to the network slice is calculated; A resource allocation strategy is generated based on the resource information allocated to the network slice by the core-side network node.

7. The method according to claim 6, characterized in that, After generating the resource allocation strategy based on the resource information allocated to the network slice by the core-side network node, the method further includes: The resource allocation policy is issued via the southbound application programming interface of the software-defined network controller, using flow tables, to adjust the traffic forwarding path or bandwidth limits of the network slice; and / or, The resource allocation strategy is issued through the elastic scaling interface of the network function virtualization orchestrator to trigger the expansion / shrinkage of the number of virtual network function instances or to adjust the resource quota of network function virtualization.

8. The method according to claim 7, characterized in that, Both the elastic scaling interface of the network function virtualization orchestrator and the southbound application programming interface of the software-defined network controller use encrypted channels to implement policy distribution, and both the elastic scaling interface of the network function virtualization orchestrator and the southbound application programming interface of the software-defined network controller are configured with automatic certificate rotation cycles.

9. The method according to any one of claims 1 to 5, characterized in that, The acquisition of metric data for all network nodes in the network system to be assigned includes: Obtain a list of all network nodes in the network system to be assigned; Based on the node list, determine the location to deploy the traffic probes; Deploy traffic probes to the locations where the traffic probes are deployed; The traffic probe is used to collect metric data from all network nodes in the network system to be assigned.

10. A resource allocation device, characterized in that, include: The acquisition module is used to acquire metric data for all network nodes in the network system to be assigned. The splitting processing module is used to split the metric data according to the network slice dimension of the network system to be allocated, so as to obtain the data queue corresponding to each network slice; wherein, each data queue includes multiple slice metric data corresponding to the network slice; An aggregation processing module is used to aggregate the slice measurement data according to a preset monitoring period to obtain the slice aggregate data of the network slice within the preset monitoring period. The calculation module is used to calculate the resource violation amount of the network system to be allocated based on the slice aggregated data; The allocation module is used to trigger resource allocation of the network system to be allocated if the number of resource violations is greater than 0, so as to generate a resource scheduling strategy for the network nodes in the network system to be allocated through the data queue.

11. A resource allocation device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-9.

13. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-9.

Images