Message processing method and device based on SD-WAN controller, medium and product
By dividing related services into fragments in the SD-WAN controller and combining them with a three-dimensional hierarchical circuit breaker mechanism, the problem of overload in the SD-WAN controller was solved, and efficient and orderly recovery of core services and resource optimization were achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING QINGWANG TECH CORP
- Filing Date
- 2026-03-16
- Publication Date
- 2026-05-01
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing technologies have failed to effectively address the overload issues caused by differences in business relevance, business priority, and message type in the message processing of SD-WAN controllers, resulting in low fault recovery efficiency, core business delays, and logical errors.
By dividing related business segments based on business association identifiers and business types, and combining two-dimensional linkage indicators and three-dimensional hierarchical circuit breaker mechanisms, precise expansion and circuit breaker operations are carried out to ensure the priority and sequential processing of core businesses.
It enables efficient and orderly recovery of core business operations, reduces fault recovery time, improves resource utilization, and meets the high availability requirements of SD-WAN scenarios.
Smart Images
Figure CN121967230A_ABST
Abstract
Description
A message processing method, device, medium, and product based on an SD-WAN controller Technical Field
[0001] This application relates to the field of communications, and in particular to a message processing method, device, medium, and product based on an SD-WAN controller. Background Technology
[0002] The SD-WAN (Software-Defined Wide Area Network) controller is the core hub of software-defined wide area network services. Its connection forwarding nodes need to support massive client connections and high-concurrency message processing. Messages in this scenario have three unique characteristics: ① Strong business correlation (network configuration and device status messages within the same business link must be executed sequentially, otherwise business logic will be disrupted); ② High priority for core businesses (the loss from interrupting core network services far exceeds that of ordinary services); ③ Significant differences in message types (network configuration cannot be interrupted, but log uploads can be temporarily rate-limited).
[0003] In practical applications, connection forwarding nodes may experience sudden load spikes due to hardware failures, resource exhaustion, or other anomalies, triggering a chain reaction of problems: message backlog, core business delays, disordered recovery, and logical errors. Existing technologies address these issues using scaling techniques. One general elastic scaling technology monitors basic resource metrics such as server CPU utilization, memory usage, and network bandwidth, setting fixed thresholds (e.g., CPU > 80%). When these metrics exceed the thresholds, it automatically adds connection forwarding nodes or expands the container size. After scaling, the load is distributed using common algorithms such as round-robin or hashing. This technology only focuses on matching resource supply with basic load, without considering business relevance or priority. Specifically, the following drawbacks exist: (1) No business priority awareness: After expansion, the load is evenly distributed, and core businesses cannot obtain priority resource guarantees. They may still face message processing delays and cannot meet the high SLA requirements of core businesses in the SD-WAN scenario; (2) Ignoring message correlation: The message of the same business link is not considered to be executed sequentially. Load distribution may cause related messages to be scattered to multiple connection forwarding nodes, causing network configuration logic errors and abnormal device status synchronization; (3) Low expansion accuracy: It only relies on resource indicators and does not combine message fragment backlog status and business weight, resulting in the resource mismatch problem of "core fragment backlog and ordinary fragment resources idle after expansion"; (4) No linkage mechanism: It only solves the resource expansion problem and does not match the flow limiting and pressure reduction, message recovery and other mechanisms, and cannot systematically solve the chain failure caused by overload.
[0004] However, existing expansion technologies are designed for general scenarios and are not adapted to the associated message processing characteristics of SD-WAN controllers. They can only solve the problem of single resource supply and cannot form full-link management and control, resulting in low fault recovery efficiency, poor core business experience, and high business logic risk. Summary of the Invention
[0005] The purpose of this application is to provide a message processing method, device, medium, and product based on an SD-WAN controller, which can form end-to-end control and improve fault recovery efficiency and communication efficiency.
[0006] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a message processing method based on an SD-WAN controller, applied to a connection forwarding node cluster of an SD-WAN controller. The message processing method based on the SD-WAN controller includes: dividing the message data stream sent by the client into different associated service fragments based on service association identifiers and service types; sending the fragmented message data streams to the connection forwarding node cluster, while synchronizing the basic information of each associated service fragment; sorting messages within the same associated service fragment by timestamp; the basic information includes: fragment ID, service level, and message quantity; constructing a two-dimensional linkage indicator by combining the acquired message layer indicators and connection forwarding node layer indicators, and monitoring the backlog status of associated service fragments and the load status of the connection forwarding node cluster in real time; the two-dimensional linkage indicator includes: fragment backlog amount and arbitrary fragment backlog growth rate. The system considers the weighted load of connection forwarding nodes and thread busy rate. If the two-dimensional linkage indicators meet the preset circuit breaker triggering conditions, a three-dimensional tiered circuit breaker operation is performed. The three-dimensional tiered circuit breaker operation determines the circuit breaker object based on business level, business type, and fragment priority, and only blocks the queuing of messages from low-priority fragments of ordinary business, and temporarily stores the corresponding circuit breaker messages in a cold queue according to the fragment ID. If the fragment backlog of the core business fragment does not decrease within a preset time after the circuit breaker, a weighted linkage expansion operation is performed. The weighted linkage expansion operation determines the number of connection forwarding nodes to be expanded according to the fragment backlog and business priority weight, and prioritizes the allocation of the newly added connection forwarding nodes to the core business fragment. If the weighted load of the connection forwarding nodes recovers to a preset normal state and continues for a preset duration, a fragment order self-healing operation is performed. The fragment order self-healing operation restores the temporarily stored circuit breaker messages from the cold queue according to priority order and timestamp order.
[0007] Secondly, this application provides a message processing device based on an SD-WAN controller, applied to a connection forwarding node cluster of the SD-WAN controller. The message processing device includes: an associated service sharding module, used to divide the message data stream sent by the client into different associated service shards based on service association identifiers and service types; and to send the sharded message data streams to the connection forwarding node cluster, while synchronizing the basic information of each associated service shard; messages within the same associated service shard are sorted by timestamp; the basic information includes: shard ID, service level, and message quantity; and a two-dimensional linkage monitoring module, used to construct two-dimensional linkage indicators from the acquired message layer indicators and connection forwarding node layer indicators, and to monitor the backlog status of associated service shards and the load status of the connection forwarding node cluster in real time; the two-dimensional linkage indicators include: shard backlog amount, arbitrary shard backlog growth rate, weighted load of connection forwarding nodes, and thread count. Busyness rate; a three-dimensional hierarchical circuit breaker module, used to perform a three-dimensional hierarchical circuit breaker operation if the two-dimensional linkage indicator meets the preset circuit breaker triggering conditions; the three-dimensional hierarchical circuit breaker operation determines the circuit breaker object based on service level, service type and fragment priority and only blocks the queuing of messages from low-priority fragments of ordinary services, and temporarily stores the corresponding circuit breaker messages in a cold queue according to fragment ID; a weighted linkage expansion module, used to perform a weighted linkage expansion operation if the fragment backlog of core service fragments does not decrease within a preset time after the circuit breaker; the weighted linkage expansion operation determines the number of expansion connection forwarding nodes according to the fragment backlog and service priority weight, and prioritizes the allocation of the newly added connection forwarding nodes to core service fragments; a fragment order self-healing module, used to perform a fragment order self-healing operation if the weighted load of the connection forwarding nodes recovers to a preset normal state and continues for a preset duration; the fragment order self-healing operation restores the temporarily stored circuit breaker messages from the cold queue according to priority order and timestamp order.
[0008] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the message processing method based on the SD-WAN controller described in the claims.
[0009] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned message processing method based on an SD-WAN controller.
[0010] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the aforementioned message processing method based on an SD-WAN controller.
[0011] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a message processing method, device, medium, and product based on an SD-WAN controller. Through business association awareness and priority-weighted precise expansion, it ensures that core businesses obtain computing resources first, and that the order of associated messages is not disordered. Through a three-dimensional hierarchical circuit breaker mechanism based on service level, service type, and fragmentation, it achieves zero circuit breaker for core business fragmentation and precise circuit breaker for low-priority fragmentation of ordinary services, avoiding false circuit breakers and over-protection. By establishing a closed-loop linkage of fragmentation circuit breaker-weighted expansion-sequential self-healing, it systematically solves the chain problems of message backlog, disordered recovery, core business delays, and logical errors. Attached Figure Description
[0012] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0013] Figure 1 is a schematic flowchart of a message processing method based on an SD-WAN controller in one embodiment of this application; Figure 2 is an overall schematic flowchart of a message processing method based on an SD-WAN controller in one embodiment of this application (the nodes in Figure 2 are connection forwarding nodes); Figure 3 is a schematic flowchart of a message processing device based on an SD-WAN controller in one embodiment of this application. Detailed Implementation
[0014] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0015] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0016] In an exemplary embodiment, as shown in Figures 1 and 2, a message processing method based on an SD-WAN controller is provided, applied to a connection forwarding node cluster of an SD-WAN controller. The method includes the following steps S101 to S105: S101: Based on the service association identifier and service type, the message data stream sent by the client is divided into different associated service fragments; the fragmented message data stream is then sent to the connection forwarding node cluster, while synchronizing the basic information of each associated service fragment; messages within the same associated service fragment are sorted by timestamp; the basic information includes: fragment ID, service level, and message quantity; S101 specifically includes: S11: Obtaining the service association identifier and service type of each message in the message data stream; Fragment ID = hash(service association identifier) mod total number of fragments + service type code; where hash is a hash function that converts the service association identifier into a fixed value, and mod is a modulo operation to ensure the result falls within a reasonable range.
[0017] The service association identifier is a unique combination of characteristic fields associated with a certain type / line of SD-WAN services. It must meet the requirements of uniqueness, stability, and hashability. Common combinations include: tenant ID + service line ID, link ID + device ID, and application type ID + session ID. The voice service association identifier for enterprise SD-WAN can be defined as Tenant ID: 1001 + Application Type ID: VoIP + Link ID: LINK-002; S12, generate a fragment ID based on the service association identifier, total number of fragments, and service type code for each message; and group messages with the same service association identifier and the same service type into the same fragment; fragment priority (from high to low): Core service P0 (core network configuration) fragment > Core service P1 (device status synchronization) fragment > Ordinary service P0 (core network configuration) fragment > Ordinary service P1 (device status synchronization) fragment > Ordinary service P2 (non-core services such as log upload, i.e., low priority ordinary services) fragment; among them, core services: affect the basic connectivity of SD-WAN and critical user services, have high SLA requirements, and interruptions will cause system or core service paralysis (such as link status synchronization, critical traffic scheduling); ordinary services: auxiliary operation and maintenance / management functions, have low SLA requirements, and interruptions only affect local efficiency (such as non-critical log collection, configuration backup notification).
[0018] S13: Sort messages within the same related business segment by timestamp.
[0019] Among them, the service types include the following types: (1) Link status type: carrying data such as the connectivity status, bandwidth utilization, latency / packet loss rate, etc. of the links between SD-WAN devices; such as link online / offline notifications and link quality detection reports.
[0020] (2) Traffic scheduling: used to issue traffic policies, path optimization instructions, and traffic splitting / aggregation rules; such as application-based routing policy messages and link load balancing adjustment instructions.
[0021] (3) Device configuration: includes device configuration distribution, configuration synchronization, and configuration rollback requests; such as tunnel parameter configuration, security policy configuration, and device firmware upgrade instructions.
[0022] (4) Security Policy: Carries security-related messages such as firewall rules, encrypted tunnel key updates, DDoS protection alarms, and access control list (ACL) synchronization.
[0023] (5) Operation and Maintenance Monitoring: Covers operation and maintenance data S102 such as equipment operating status, resource utilization (CPU / memory), log collection, and fault alarms (such as equipment downtime, tunnel interruption). The acquired message layer indicators and connection forwarding node layer indicators are combined to form a two-dimensional linkage indicator, and the backlog status of related business shards and the load status of the connection forwarding node cluster are monitored in real time. The two-dimensional linkage indicator includes: shard backlog amount, arbitrary shard backlog growth rate, connection forwarding node weighted load, and thread busy rate. S102 collects message layer indicators and node layer indicators, determines whether the circuit breaker threshold or expansion threshold is triggered, and the monitored data is shown in Table 1. Table 1
[0024] S103, if the two-dimensional linkage indicator meets the preset circuit breaker triggering conditions, a three-dimensional hierarchical circuit breaker operation is performed, as shown in Table 2. The three-dimensional hierarchical circuit breaker operation determines the circuit breaker object based on the service level, service type, and fragment priority, and only blocks the queuing of messages from low-priority fragments of ordinary services. The corresponding circuit breaker messages are temporarily stored in a cold queue according to the fragment ID, while retaining the timestamp. The circuit breaker triggering conditions are that any message layer indicator exceeds the limit and any connection forwarding node layer indicator exceeds the limit, and the duration reaches 30 seconds. The cold queue is a low-priority, persistent queue for temporarily storing messages after the circuit breaker is triggered. It has low resource consumption, stores messages on disk rather than in memory, does not consume them in real time, and can revert messages after the circuit breaker is restored. Messages that have expired can be cleaned up / archived.
[0025] Table 2
[0026] S104, if the backlog of core service fragments does not decrease within a preset time (1 minute) after the circuit breaker is triggered, a weighted linkage expansion operation is performed. The weighted linkage expansion operation determines the number of expansion connection forwarding nodes based on the fragment backlog and service priority weight, and prioritizes the allocation of the newly added connection forwarding nodes to the core service fragments. The number of expansion connection forwarding nodes N is determined using the formula N=ceil(Σ(fragment backlog × service priority weight) × single connection forwarding node processing rate ÷ (bearing limit × 1.2)). Wherein, ceil is the floor function, and the result less than 1 is counted as 1 to ensure that the number of nodes is an integer; Σ is the summation of all message fragments.
[0027] Newly added connection forwarding nodes are given priority to carry core P0 / P1 shards, with the allocation ratio being the sum of core shard weights divided by the total weight.
[0028] As a specific implementation, the weighted linkage expansion operation includes: migrating the weighted load of the connection forwarding nodes corresponding to the core service shards within a set range (30%) to the newly added connection forwarding nodes; and synchronously updating the mapping relationship between the core service shards and the connection forwarding nodes.
[0029] S105, if the weighted load of the connection forwarding node recovers to the preset normal state and continues for a preset duration, then a fragmentation order self-healing operation is performed; the fragmentation order self-healing operation is to restore the temporarily stored circuit breaker messages from the cold queue in order of priority and timestamp.
[0030] S105 includes: S1, continuously monitoring the weighted load of the connection forwarding node; S2, if the weighted load of the connection forwarding node is ≤50% and continues for 10 minutes, release the resources of the redundant connection forwarding node, update the mapping table between the shard and the connection forwarding section, and return to normal monitoring status.
[0031] Based on the same inventive concept, this application also provides an SD-WAN controller-based message processing device for implementing the aforementioned message processing method based on an SD-WAN controller. The solution provided by this device is similar to the implementation described in the above method. Therefore, the specific limitations of one or more SD-WAN controller-based message processing device embodiments provided below can be found in the limitations of the SD-WAN controller-based message processing method described above, and will not be repeated here.
[0032] In an exemplary embodiment, a message processing device based on an SD-WAN controller is provided, applied to a connection forwarding node cluster of the SD-WAN controller. The message processing device includes: an associated service sharding module, used to divide the message data stream sent by the client into different associated service shards based on service association identifiers and service types; and to send the sharded message data streams to the connection forwarding node cluster, while synchronizing the basic information of each associated service shard; messages within the same associated service shard are sorted by timestamp; the basic information includes: shard ID, service level, and message quantity; and a two-dimensional linkage monitoring module, used to construct two-dimensional linkage indicators from the acquired message layer indicators and connection forwarding node layer indicators, and to monitor the backlog status of associated service shards and the load status of the connection forwarding node cluster in real time; the two-dimensional linkage indicators include: shard backlog amount, arbitrary shard backlog growth rate, and weighted load of connection forwarding nodes. Thread busy rate; a three-dimensional hierarchical circuit breaker module, used to perform a three-dimensional hierarchical circuit breaker operation if the two-dimensional linkage indicator meets the preset circuit breaker triggering conditions; the three-dimensional hierarchical circuit breaker operation determines the circuit breaker object based on business level, business type and shard priority and only blocks the queuing of messages from low-priority shards of ordinary business, and temporarily stores the corresponding circuit breaker messages in a cold queue according to the shard ID; a weighted linkage expansion module, used to perform a weighted linkage expansion operation if the shard backlog of the core business shard does not decrease within a preset time after the circuit breaker; the weighted linkage expansion operation determines the number of expansion connection forwarding nodes according to the shard backlog and business priority weight, and prioritizes the allocation of the newly added connection forwarding nodes to the core business shards; a shard order self-healing module, used to perform a shard order self-healing operation if the weighted load of the connection forwarding nodes recovers to a preset normal state and continues for a preset duration; the shard order self-healing operation restores the temporarily stored circuit breaker messages from the cold queue according to priority order and timestamp order.
[0033] The message flow data of the customers obtained in this application is divided by the associated business sharding module and then processed by the connection forwarding node cluster; the two-dimensional linkage monitoring module collects the sharding and connection forwarding node indicators and triggers the three-dimensional hierarchical circuit breaker module to execute precise circuit breaking; if the core business is still backlogged after the circuit breaker is triggered, the weighted linkage expansion is started; after the load is restored, the sharding sequence self-healing module restores the messages according to the rules; and it also includes a configuration management module; the configuration management module supports the initialization of parameters throughout the process.
[0034] As shown in Figure 3, the processing flow of a message processing device based on an SD-WAN controller provided in this application is as follows: Step 1: System initialization configuration (execution subject: configuration management module) obtain the preset business priority weight configuration (core business weight 1.5, ordinary business weight 0.8), associated business sharding rules (sharding ID determination process), two-dimensional indicator thresholds (core P0 shard backlog ≤ 3000 messages, connection forwarding node weighted load ≤ bearing limit × 1.2, thread busy rate ≤ 85%), and single connection forwarding node basic parameters (bearing limit 4000 connections, message processing rate 10 messages / second).
[0035] The configuration management module loads the above parameters into local storage and sends an initialization completion signal to the associated business sharding module, two-dimensional linkage monitoring module, three-dimensional hierarchical circuit breaker module, weighted linkage expansion module, and sharding sequence self-healing module to synchronize the parameters required by each module.
[0036] Step 2: Related Business Sharding (Execution Entity: Related Business Sharding Module) Obtain message data streams sent by the client cluster (including business association identifier, business type, message content, and timestamp) and sharding rules issued by the configuration management module.
[0037] The associated business sharding module parses the business association identifier and business type of each message; generates a unique shard ID according to the shard ID calculation logic, and groups messages with the same business association identifier and the same business type into the same shard; sorts the messages in each shard according to the timestamp in a "first-in, first-out" manner; forwards the sharded message data stream to the connection forwarding node cluster, and simultaneously synchronizes the basic information of each shard (shard ID, business level, message quantity) to the two-dimensional linkage monitoring module to obtain the message data stream after sharding and the basic information table of each shard.
[0038] Step 3: Two-dimensional linkage indicator collection and monitoring (Execution subject: two-dimensional linkage monitoring module) to obtain real-time operating data of the connection forwarding node cluster (number of connection forwarding nodes, thread busy rate), sharding information table synchronized by the associated business sharding module, and indicator thresholds issued by the configuration management module.
[0039] Message layer metrics are collected once per second: real-time backlog of core P0 shards and backlog growth rate of all shards (calculated every 30 seconds); connection forwarding node layer metrics are collected once per second: weighted load of a single connection forwarding node (calculated as weighted load = Σ(number of shard connections × business priority weight)) and thread busy rate; the collected metric data are compared with preset thresholds in real time to obtain real-time monitoring results (meeting / exceeding standards).
[0040] Step 4: Judgment of dual indicators exceeding the standard (Execution subject: two-dimensional linkage monitoring module) Obtain the real-time monitoring results of the indicators.
[0041] Determine if the triggering conditions of "any message layer indicator exceeding the standard + any connection forwarding node layer indicator exceeding the standard" and "duration ≥ 30 seconds" are met: If not met, return to step 3 to continue data collection and monitoring; if met, send a circuit breaker trigger signal to the three-dimensional hierarchical circuit breaker module, along with details of the exceeding indicators (such as 3500 core P0 fragment backlogs and 5200 weighted loads on connection forwarding nodes), and obtain the circuit breaker trigger signal / continue monitoring instruction.
[0042] Step 5: P2 segmented precise circuit breaking for ordinary services (Execution subject: three-dimensional hierarchical circuit breaking module) Obtain the circuit breaking trigger signal sent by the two-dimensional linkage monitoring module and the hierarchical circuit breaking rules issued by the configuration management module.
[0043] Based on the hierarchical rules, only P2-level fragments of ordinary services are selected as circuit breaker targets, while all fragments of core services and P0 / P1 fragments of ordinary services continue to forward normally; message enqueueing of P2 fragments of ordinary services is blocked, such messages are temporarily stored in a cold queue according to fragment ID, and the fragment ID, timestamp, and service association identifier of each message are recorded; a circuit breaker execution completion signal is fed back to the two-dimensional linkage monitoring module, and the information of the circuit-breaker fragments is synchronized.
[0044] Output: Circuit breaker completion signal, and P2 fragmented messages of ordinary services temporarily stored in the cold queue.
[0045] Step 6: Determine the backlog status of the core P0 segment (Execution subject: two-dimensional linkage monitoring module) Obtain the fuse completion signal fed back by the three-dimensional hierarchical fuse module and the real-time backlog data of the core P0 segment.
[0046] Continuously monitor the change in the backlog of core P0 shards within 1 minute: if the backlog drops to within the threshold (≤3000), send a waiting load recovery instruction to the sharding order self-healing module; if the backlog does not drop (still >3000), send an expansion trigger signal to the weighted linkage expansion module, along with parameters such as the core P0 shard backlog and shard weight, to obtain the expansion trigger signal / waiting load recovery instruction.
[0047] Step 7: Weighted linkage expansion execution (Execution subject: weighted linkage expansion module) Obtain the expansion trigger signal sent by the two-dimensional linkage monitoring module, the core P0 fragment backlog, and the expansion connection forwarding node number calculation model issued by the configuration management module.
[0048] The expansion number of connection forwarding nodes is calculated according to the model: Expansion number of connection forwarding nodes = ceil(Σ(fragment backlog × weight) × single connection forwarding node processing rate ÷ (bearing limit × 1.2)); A connection forwarding node creation request is sent to the cloud platform / connection forwarding node management system to generate a new connection forwarding node with the same configuration as the existing connection forwarding nodes; After the new connection forwarding node goes online, the program migrates 30% of the core P0 fragment load to the new connection forwarding node according to the core business priority allocation strategy, and updates the mapping relationship between fragments and connection forwarding nodes synchronously; The expansion completion signal is fed back to the two-dimensional linkage monitoring module to obtain the expansion completion signal and the updated fragment-connection forwarding node mapping table.
[0049] Step 8: Load recovery waiting (Execution subject: two-dimensional linkage monitoring module) Obtain the expansion completion signal, the waiting load recovery instruction issued in step 6, the real-time operating indicators of the connection forwarding node (weighted load, thread busy rate), and the core P0 fragment backlog.
[0050] Continuously monitor the load of connection forwarding nodes and the backlog of core P0 fragments until the basic conditions of "core P0 fragment backlog cleared + weighted load of connection forwarding nodes ≤ 60% + thread busy rate ≤ 70%" are met, and obtain real-time data on load recovery status.
[0051] Step 9: Determine the normal load status (Execution subject: Two-dimensional linkage monitoring module) Obtain load recovery status data.
[0052] Determine if the above basic conditions are met for 5 minutes: if not, return to step 8 to continue waiting for monitoring; if met, send a message recovery trigger signal to the sharding sequence self-healing module, along with the sharding information temporarily stored in the cold queue, and obtain the message recovery trigger signal / continue waiting instruction.
[0053] Step 10: Cold queue message order self-healing (execution subject: fragmented order self-healing module) obtains the message recovery trigger signal sent by the two-dimensional linkage monitoring module and the ordinary business P2 fragmented messages temporarily stored in the cold queue.
[0054] Message recovery is initiated according to the priority order of the fragments (core P1 → ordinary P0 → ordinary P1 → ordinary P2); messages within the same fragment are sorted by timestamp "first in first out" and resumed from the point of interruption based on message ID verification to avoid duplicate consumption; after recovery is completed, a message recovery completion signal is sent back to the two-dimensional linkage monitoring module, and a message recovery completion signal and a cold queue clearing instruction are received.
[0055] Step 11: On-demand scaling down and system regression (Execution subject: weighted linkage expansion module + two-dimensional linkage monitoring module) Obtain message recovery completion signal and connect to forwarding node real-time weighted load data.
[0056] The two-dimensional linkage monitoring module continuously monitors for 10 minutes. If the weighted load of the connection forwarding node is ≤50%, it sends a reduction trigger signal to the weighted linkage expansion module. The weighted linkage expansion module releases one redundant connection forwarding node resource and updates the shard-connection forwarding node mapping table. The two-dimensional linkage monitoring module switches to normal monitoring mode, and each module returns to the loop process of step 3 to obtain the system normal operation signal and return to the continuous monitoring state.
[0057] The following is an explanation through specific implementation examples: (1) Initial configuration of service weight: core service 1.5, ordinary service 0.8, which can be adjusted according to time needs, with core service being higher than ordinary service; sharding rules: total number 20, shard ID = hash(service association identifier) mod 20 + service code; threshold: core P0 shard backlog ≤ 3000, connection forwarding node weighted load ≤ 4800; basic parameters: single connection forwarding node carries 4000 connections, processing rate 10 messages / second.
[0058] (2) Abnormal Triggering and Handling Process Abnormal Monitoring: At time T0, the connection forwarding node fails and the load is transferred; at T0+1min, the core P0 fragment backlog is 3500 messages and the weighted load of the connection forwarding node is 5200, triggering a circuit breaker due to double overload; Tiered Circuit Breaker: at T0+1min30s, the P2 fragment of ordinary services is blocked, while the core fragment is processed normally; Weighted Expansion: at T0+2min30s, the core P0 fragment backlog has not decreased, and it is calculated that 2 expansions are needed; at T0+8min30s, the new connection forwarding node shares 30% of the core load, and the load of the connection forwarding node drops to 3640; Sequential Self-Healing: at T0+26min, the core P0 fragment backlog is cleared; at T0+27min, cold queue messages are restored according to priority; at T0+41min, all messages are restored; On-Demand Scaling Down: at T0+46min, the load remains at 50%, and 1 redundant connection forwarding node is released.
[0059] (3) Implementation effect: Business recovery time: shortened from several hours in the traditional solution to minutes, with a huge improvement in efficiency; Core business delay: low latency, zero interruption, and no order disorder; Message loss rate: 0%, lossless recovery of P2 fragmented messages for ordinary business; Resource utilization rate: expansion and contraction are more scientific and reasonable.
[0060] This application brings the following beneficial effects: (1) Core business protection capability is greatly improved: Associated sharding + weighted expansion realizes priority allocation of core resources, reduces processing latency, and meets the high SLA requirements of SD-WAN; (2) Fault recovery efficiency is significantly optimized: The full-link linkage closed loop shortens the recovery time and greatly reduces business losses; (3) Circuit breaking accuracy is greatly improved: The false circuit breaking rate of three-dimensional hierarchical circuit breaking is reduced to 0%, taking into account both system stability and business integrity; (4) Message recovery is orderly and controllable: The self-healing of sharding order completely solves the problem of logical disorder and improves the compliance of the execution order of associated messages; (5) Resource utilization is optimized: The weighted algorithm avoids resource mismatch and improves resource utilization.
[0061] As a specific implementation, sharding can also be achieved through business association identifiers and business links; the core association perception logic remains unchanged, adapting to complex business scenarios; as a specific implementation, business type weights can be superimposed (core P0 weight = 1.5 × 1.2 = 1.8), the expansion model remains unchanged, and the core priority is strengthened; as a specific implementation, expansion can also be achieved by replacing with K8s containerization, only adjusting the connection forwarding node creation interface, and the load distribution logic remains unchanged; as a specific implementation, sharding can be sorted according to the urgency of the business, and the timestamp execution mechanism within the shard remains unchanged, adapting to different business needs.
[0062] In an exemplary embodiment, a computer device is provided, which may be a server or a terminal. The computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I / O interfaces. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The I / O interfaces of the computer device are used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a message processing method based on an SD-WAN controller.
[0063] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.
[0064] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.
[0065] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.
[0066] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.
[0067] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).
[0068] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.
[0069] In this application, all actions to acquire signals, information, or data are carried out in compliance with the relevant data protection laws and policies of the country where the location is situated, and with the authorization granted by the owner of the relevant device.
[0070] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0071] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A message processing method based on an SD-WAN controller, characterized in that, An SD-WAN controller-based message processing method for connection forwarding node clusters includes: dividing client-sent message data streams into different associated service fragments based on service association identifiers and service types; sending the fragmented message data streams to the connection forwarding node cluster while synchronizing basic information for each associated service fragment; sorting messages within the same associated service fragment by timestamp; the basic information includes fragment ID, service level, and message quantity; constructing a two-dimensional linkage indicator by combining acquired message layer metrics and connection forwarding node layer metrics, and monitoring the backlog status of associated service fragments and the load status of the connection forwarding node cluster in real time; the two-dimensional linkage indicator includes fragment backlog amount, arbitrary fragment backlog growth rate, weighted load of connection forwarding nodes, and thread busy rate; if the two-dimensional linkage... If the indicators meet the preset circuit breaker trigger conditions, a three-dimensional tiered circuit breaker operation is performed. This operation determines the circuit breaker object based on service level, service type, and fragment priority, blocking only low-priority fragment messages from ordinary services from being queued, and temporarily storing the corresponding circuit breaker messages in a cold queue according to fragment ID. If the fragment backlog of the core service fragment does not decrease within a preset time after the circuit breaker, a weighted linkage expansion operation is performed. This operation determines the number of expansion connection forwarding nodes based on the fragment backlog and service priority weight, and prioritizes allocating the newly added connection forwarding nodes to the core service fragment. If the weighted load of the connection forwarding nodes recovers to a preset normal state for a preset duration, a fragment order self-healing operation is performed. This operation restores the temporarily stored circuit breaker messages from the cold queue in priority and timestamp order.
2. The message processing method based on an SD-WAN controller according to claim 1, characterized in that, The process of dividing the message data stream sent by the client into different associated business segments based on the business association identifier and business type includes: obtaining the business association identifier and business type of each message in the message data stream; generating a segment ID based on the business association identifier, the total number of segments, and the business type code of each message; grouping messages with the same business association identifier and the same business type into the same segment; and sorting messages within the same associated business segment by timestamp.
3. The message processing method based on an SD-WAN controller according to claim 1, characterized in that, The circuit breaker is triggered when any message layer indicator and any connection forwarding node layer indicator exceed the limit, and the duration reaches 30 seconds.
4. The message processing method based on an SD-WAN controller according to claim 1, characterized in that, The process of determining the number of forwarding nodes for expanded connections is as follows: The number of forwarding nodes N for expanded connections is determined using the formula N=ceil(Σ(fragment backlog × service priority weight) × single connection forwarding node processing rate ÷ (bearing limit × 1.2)); where ceil is the floor function, and the result is rounded up to 1 to ensure that the number of nodes is an integer; Σ is the sum of all message fragments.
5. The message processing method based on an SD-WAN controller according to claim 1, characterized in that, The weighted linkage expansion operation specifically includes: migrating the weighted load of the connection forwarding nodes corresponding to the core business shards within a defined range to the newly added connection forwarding nodes; and synchronously updating the mapping relationship between the core business shards and the connection forwarding nodes.
6. The message processing method based on an SD-WAN controller according to claim 1, characterized in that, If the weighted load of the connection forwarding node recovers to a preset normal state and remains so for a preset duration, a sharding sequence self-healing operation is performed. This operation then includes: continuously monitoring the weighted load of the connection forwarding node; if the weighted load of the connection forwarding node is ≤50% and remains so for 10 minutes, the resources of the redundant connection forwarding node are released, the mapping table between shards and connection forwarding nodes is updated, and the system returns to normal monitoring status.
7. A message processing device based on an SD-WAN controller, characterized in that, An SD-WAN controller-based message processing device for a connection forwarding node cluster includes: an associated service sharding module, used to divide the message data stream sent by the client into different associated service shards based on the service association identifier and service type; and to send the sharded message data stream to the connection forwarding node cluster, while synchronizing the basic information of each associated service shard; messages within the same associated service shard are sorted by timestamp; the basic information includes: shard ID, service level, and message quantity; a two-dimensional linkage monitoring module, used to construct two-dimensional linkage indicators from the acquired message layer indicators and connection forwarding node layer indicators, and to monitor the backlog status of associated service shards and the load status of the connection forwarding node cluster in real time; the two-dimensional linkage indicators include: shard backlog amount, arbitrary shard backlog growth rate, connection forwarding node weighted load, and thread busy rate; and a three-dimensional tiered circuit breaker module, used for... If the two-dimensional linkage indicator meets the preset circuit breaker triggering conditions, a three-dimensional hierarchical circuit breaker operation is performed. The three-dimensional hierarchical circuit breaker operation determines the circuit breaker object based on the service level, service type, and fragment priority, and only blocks the queuing of messages from low-priority fragments of ordinary services, and temporarily stores the corresponding circuit breaker messages in a cold queue according to the fragment ID. The weighted linkage expansion module is used to perform a weighted linkage expansion operation if the fragment backlog of the core service fragment does not decrease within a preset time after the circuit breaker. The weighted linkage expansion operation determines the number of expansion connection forwarding nodes based on the fragment backlog and service priority weight, and prioritizes the allocation of the newly added connection forwarding nodes to the core service fragment. The fragment order self-healing module is used to perform a fragment order self-healing operation if the weighted load of the connection forwarding nodes recovers to a preset normal state and remains so for a preset duration. The fragment order self-healing operation restores the temporarily stored circuit breaker messages from the cold queue in priority order and timestamp order.
8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the message processing method based on an SD-WAN controller as described in any one of claims 1-6.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the message processing method based on an SD-WAN controller as described in any one of claims 1-6.
10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the message processing method based on an SD-WAN controller as described in any one of claims 1-6.