Credit-based shaper-oriented parameter dynamic reconfiguration method and related equipment

Abstract

The embodiment of the invention provides a credit-based shaper-oriented parameter dynamic reconfiguration method and related equipment, and belongs to the technical field of communication networks. According to the method, the problem of service interruption or unsafe transient behavior during online adjustment of the CBS bandwidth parameters is solved. The core of the method is that idle slope parameters and internal credit states of the CBS queues are updated atomically and cooperatively at the global updating moment. Two updating strategies of forced zero setting and scaling are provided, and the two updating strategies can achieve remarkable smooth transition according to the principle that zero returning time is not changed. The influence is quantized by introducing a time penalty term, and a time-varying server model is constructed, so that a provable transient worst-case time delay upper bound is derived, and the security of a reconfiguration process is ensured. The control plane is based on a centralized network controller (CNC) and a NETCONF / YANG protocol, and realizes cross-device synchronous transaction updating. According to the method and the device, the security bandwidth dynamic adjustment under the condition that the service is not interrupted is realized, and the flexibility and the certainty of the TSN network are improved.

Description

Technical Field

[0001] This application relates to the field of communication network technology, and in particular to a method and related equipment for dynamic parameter reconfiguration of a credit-based shaper. Background Technology

[0002] Time-Sensitive Networking (TSN) is a series of standards developed by the IEEE 802.1 working group. It aims to provide Ethernet with deterministic low-latency, low-jitter, and high-reliability data transmission capabilities, and is widely used in industrial automation, automotive electronics, avionics, and other fields. The Credit-Based Shaper (CBS) is a key shaping mechanism in TSN used to manage Audio / Video Bridging (AVB) traffic. It controls the sending behavior of queues through a credit variable, ensuring that high-priority traffic receives bounded latency and jitter.

[0003] CBS behavior is controlled by two key parameters: idle slope and send slope. The idle slope determines the rate at which credits accumulate in the queue while it is in a waiting state, and is directly related to the bandwidth guaranteed for that queue. In traditional TSN deployments, CBS parameters (especially the idle slope) are typically statically configured when a flow is established and remain unchanged throughout the flow's lifecycle.

[0004] However, as network applications become increasingly dynamic (such as adaptive bitrate video streaming, dynamically load-adjusted industrial control, and 5G+TSN converged systems), it is necessary to dynamically adjust the Quality of Service (QoS) guarantees during continuous streaming, for example, by increasing or decreasing the reserved bandwidth. Existing technologies typically use a "delete the old stream first, then establish the new stream" approach to achieve bandwidth adjustment, but this leads to interruptions in service communication and cannot meet the needs of scenarios with stringent continuity requirements.

[0005] Attempting to directly update the idleSlope parameter of the CBS during streaming without considering the synchronization and coordination of its internal credit status will introduce uncertainties in transient behavior. For example, reducing bandwidth when the credit is negative (payment period) may prolong the quiet period and violate the upper limit of latency; changing parameters when the credit is positive (sending period) may introduce unplanned traffic bursts, interfering with other traffic. Currently, there is a lack of a formalized, provably secure method to achieve dynamic reconfiguration of CBS parameters during operation.

[0006] Therefore, how to safely and predictably adjust the bandwidth guarantee parameters of CBS online while ensuring uninterrupted service has become a key challenge in TSN dynamic management. Summary of the Invention

[0007] The main objective of this application is to propose a method, electronic device, storage medium, and program product for dynamic parameter reconfiguration of a credit-based shaper, in order to solve the technical problem of safely adjusting its bandwidth guarantee parameters without interrupting services during continuous streaming and ensuring transient latency determinism.

[0008] To achieve the above objectives, one aspect of this application proposes a parameter dynamic reconfiguration method for credit-based shapers, applied to switches in Time-Sensitive Networks (TSNs). The method includes: At the global update time for adjusting the bandwidth guarantee parameters of the credit-based shaper CBS queue, an atomic state-parameter reconstruction operation is performed. This atomic state-parameter reconstruction operation is configured to take effect synchronously within the same scheduling cycle, including: Update the idle slope parameter of the CBS queue from the first value to the second value; Based on the credit status value before the update, the first value, and the second value, the credit status value of the CBS queue is synchronously updated to a new credit status value that matches the second value, so as to prevent the CBS queue from being in an intermediate state where the second value and the credit status value before the update coexist.

[0009] In some embodiments, the step of synchronously updating the credit status value of the CBS queue to a new credit status value matching the second value based on the previous credit status value, the first value, and the second value includes: The previous credit status value is reset to zero, which is then used as the new credit status value.

[0010] In some embodiments, the step of synchronously updating the credit status value of the CBS queue to a new credit status value matching the second value based on the previous credit status value, the first value, and the second value includes: Based on the stage of credit change in the credit status value before the update, the credit status value before the update is scaled proportionally to obtain a scaled credit value. Wherein, if the credit status value before the update is negative, the scaled credit value is equal to the credit status value before the update multiplied by the ratio of the second value to the first value; If the credit status value before the update is non-negative, the scaled credit value is equal to the credit status value before the update multiplied by the ratio of the first difference to the second difference, where the first difference is the difference between the port physical rate and the second value, and the second difference is the difference between the port physical rate and the first value. Compare the scaled credit value with the upper and lower bounds of the credit value corresponding to the second value. If the value exceeds the range, the boundary value is taken; otherwise, the original value is kept, and the new credit status value is obtained.

[0011] In some embodiments, the method further includes: Based on the credit status value before the update, the first value, and the second value, determine the time penalty introduced due to performing the atomic state-parameter reconstruction operation; Based on the time penalty and the steady-state service curve determined based on the second value, the transient minimum service curve of the CBS queue after the global update time is determined.

[0012] In some embodiments, determining the time penalty introduced due to performing the atomic state-parameter reconstruction operation based on the credit status value before the update, the first value, and the second value includes: When a strategy of resetting the credit status value to zero is adopted, the time penalty is equal to the positive part of the credit status value before the update divided by the difference between the port physical rate and the first value. When a scaling strategy is used, the time penalty is equal to the portion of the scaling credit value that exceeds the credit upper bound corresponding to the second value divided by the difference between the port physical rate and the second value. If the scaling credit value does not exceed the credit upper bound, the time penalty is zero.

[0013] In some embodiments, the method further includes: Based on the time-varying server model spanning the global update time, the arrival curve constraint corresponding to the CBS queue, and the transient minimum service curve, the worst-case upper bound of the delay of the CBS queue during the parameter reconfiguration transient process is determined.

[0014] In some embodiments, determining the worst-case upper bound of the delay of the CBS queue during the parameter reconfiguration transient process includes: If the second value is greater than the first value and a preset condition is met, the worst-case latency upper bound is determined by the queue emptying rate calculated based on the second value. The preset condition includes that the traffic burst that causes the maximum latency arrives mainly after the global update time.

[0015] In some embodiments, the method further includes: Assess the additional burst impact of the atomic state-parameter reconstruction operation on other priority traffic on the port to which the CBS queue belongs; When the credit status value before the update is negative, if a strategy of resetting the credit status value to zero is adopted, the upper bound of the additional burst impact is the ratio of the port physical rate multiplied by the absolute value of the credit status value before the update to the first value.

[0016] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0017] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described above.

[0018] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer program product, including a computer program that, when executed by a processor, implements the method described above.

[0019] Compared with the prior art, this application has the following beneficial effects: 1) Business continuity assurance: The bandwidth guarantee of the stream can be adjusted online without interruption or reconstruction of the stream, which meets the stringent requirements for communication continuity in scenarios such as automotive and industrial industries.

[0020] 2) Transient security is provable: Through formalized time penalty terms and time-varying server models, a computable upper bound for transient delay is provided for the dynamic reconfiguration process, eliminating the uncertainty of "black box" operations and enabling the system to still have verifiable determinism during reconfiguration.

[0021] 3) Performance optimization: The preferred "proportional scaling" strategy can significantly reduce service loss and additional traffic bursts caused by parameter updates. Compared with a simple reset strategy, it can provide tighter latency guarantees and has less impact on other priority flows on the same port.

[0022] 4) Controllable impact on other flows: The additional burst upper bound that the reconfiguration operation may cause to low-priority flows such as Best-Effort is clearly analyzed and quantified, which facilitates system-level interference assessment and security design.

[0023] 5) Standard compatibility and ease of implementation: The control plane solution is based entirely on the CNC architecture defined by the TSN standard and the NETCONF / YANG management protocol of the IETF standard. Utilizing its native transaction and synchronization mechanisms, it is easy to integrate into the existing TSN network management system, and the engineering implementation path is clear. Attached Figure Description

[0024] Figure 1This is a flowchart illustrating the steps of a dynamic parameter reconfiguration method for a credit-based shaper in an embodiment of this application.

[0025] Figure 2 This is a schematic diagram comparing the trajectory of credit value changes over time under two credit status update strategies in the embodiments of this application.

[0026] Figure 3 This is a schematic diagram of the equivalent transient service curves of the CBS queue before and after parameter reconfiguration in an embodiment of this application.

[0027] Figure 4 This is a control plane system architecture diagram based on CNC and NETCONF / YANG in the embodiments of this application. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0030] (1) Time-Sensitive Networks and Credit-Based Shapers Time-Sensitive Networking (TSN) is a next-generation deterministic Ethernet technology designed for scenarios such as automotive electronics, the Industrial Internet of Things (IIoT), and Industry 4.0. Typical applications include: a) Deterministic communication between domain controllers within the vehicle; b) Plug-and-play access to modular equipment within the factory production line; c) Mixed handling of periodic and sudden business in industrial control systems.

[0031] In the TSN standard system, the Credit-Based Shaper (CBS) defined in IEEE 802.1Qav is an important mechanism for providing bounded latency and low jitter for AVB / SR class traffic. The CBS maintains a credit variable that evolves over time. The sending behavior of the queue is controlled by two core parameters: a) Free slope When the queue is in a waiting state The cumulative rate; b) Sending slope When the queue sends frames The consumption rate generally satisfies:

[0032] in This refers to the physical speed of the port.

[0033] In existing engineering practices, Typically, configuration is statically based on end-to-end bandwidth reservation results, for example:

[0034] in The percentage of bandwidth reserved for this stream (or queue).

[0035] (2) Limitations of existing dynamic reconfiguration Existing research on TSN "dynamic reconstruction" mainly focuses on the following types of issues: 1) Dynamic flow management: including online access control, dynamic path selection, and the addition and deletion of flows; 2) TAS (IEEE 802.1Qbv) Gated List (GCL) Update: Update the gating schedule of the time-aware shaper according to the global clock synchronization; 3) Fault tolerance and rerouting: Incremental rerouting and scheduling adjustments are performed when links or nodes fail.

[0036] Most of these studies treat shapers such as CBS / TAS as "black boxes," focusing on how to "calculate parameters" and "issue parameters when creating / deleting a stream." The implicit assumption is that "modification" of a stream can be accomplished through a destructive process of "deleting the old stream and creating a new stream."

[0037] However, for applications that require continuous service but also need to adjust the quality of service (QoS) online (such as adaptive bitrate video streaming), this "tear-down and rebuild" approach can lead to service interruption (glitch) and fail to meet the system's requirements for continuity and security.

[0038] (3) Transient problems in online modification of CBS parameters When an existing flow needs to adjust its bandwidth guarantees, the ideal network behavior is: a) Update it without interrupting the flow. ; b) The promised upper bound of worst-case delay (WCD) is maintained throughout the entire transient process.

[0039] However, in only updating Without processing When in a state, the following transient ambiguity problem may occur: 1) If at the update time , (The queue is in the "repayment" phase) and new parameters (If bandwidth is reduced), the queue will be restored to its original state. The increased time required may lead to transient WCD violations; 2) If (The queue is in the "continuous sending" phase), modify Will change at the same time This causes unpredictable jumps in the credit trajectory corresponding to the currently being sent data packets, thereby introducing instantaneous bursts that exceed the original network calculation and analysis. 3) Even Simple update While locally safe, the requirement to update only when the queue is idle significantly reduces the responsiveness of dynamic refactoring.

[0040] The existing publicly available literature does not provide information on "modifying running CBS parameters online (especially...)". Formal analysis of the transient behavior at time 0 ... a) Static or quasi-static Calculation and configuration; b) Or adjust the bandwidth indirectly by creating / deleting streams.

[0041] Therefore, how can we securely modify the switch online without interrupting the streaming service? and its related This status represents a gap in the current TSN reconstruction field.

[0042] To address the aforementioned issues, this embodiment provides a transient security dynamic reconfiguration scheme for TSN CBS, which includes the following key points: 1) Introduce a consistent state-parameter reconfiguration mechanism: during global update... ,right and The state is updated atomically. 2) Two specific collaborative update strategies are proposed: 2.1) Baseline strategy: Forced zeroing. Set it directly to 0; 2.2) Optimal Strategy: Scaling, following the "Time-to-Zero Invariant" principle. Perform scaling and match with the new parameters; 3) Introduce a time penalty. The concept of ) equates the service loss during the reconfiguration period to a time shift of the entire service curve; 4) Construct a Time-Varying Server model that spans update points. The lower bound of the transient service curve is derived, and the upper bound of the worst-case transient latency and the additional burst upper bound caused to other service flows are further derived. 5) At the control plane level, based on the TSN CNC and NETCONF / YANG protocols, design an implementation process that supports atomic and transactional parameter updates to ensure that the above reconfiguration operations can be triggered by precise timestamps in the actual network.

[0043] This embodiment provides a parameter dynamic reconfiguration method for credit-based shapers, applied to TSN switches. The core of the method is to perform an atomic collaborative update operation on the idle slope parameter and credit status value of the CBS queue at a preset global update time.

[0044] like Figure 1 As shown, this atomic operation consists of two indivisible steps: S1: Update the idle slope parameter of the CBS queue from the current value (first value) to the target value (second value).

[0045] S2: Based on the credit status value before the update, the first value, and the second value, calculate a new credit status value that matches the new idle slope parameter, and update the credit status register synchronously.

[0046] In some embodiments, the credit status update strategy includes at least two types: a) Forced Zeroing Strategy: Directly resets the credit status value to zero. This strategy is simple to implement, but may introduce large transient disturbances.

[0047] b) Scaling Strategy: Based on the principle that "the time required for credit to reach zero remains unchanged," the credit status value before the update is scaled proportionally. Specifically, if the credit before the update is negative (in the credit recovery phase), it is multiplied by (second value / first value); if the credit before the update is non-negative (in the credit consumption phase), it is multiplied by ((C - second value) / (C - first value)), where C is the port physical rate. The scaled value must be limited to the upper and lower bounds of the credit corresponding to the new parameters. This strategy can maximize the smooth transition and reduce service loss and additional bursts.

[0048] In some embodiments, to quantify the impact of reconfiguration operations, this embodiment introduces the concept of a time penalty. The time penalty represents the equivalent overall service curve delay caused by the mismatch between credit status and parameters. Based on the time penalty and the new idle slope parameter, the transient minimum service curve of the CBS queue after reconfiguration can be derived.

[0049] Furthermore, this embodiment constructs a time-varying server model spanning update moments, unifying the service capabilities before and after reconfiguration within a single analytical framework. By combining flow arrival curve constraints (such as the token bucket model), the upper bound of the worst-case delay (WCD) for the entire transient process can be calculated, thus proving the security of the reconfiguration process. For scenarios with increased bandwidth, a more compact method for calculating the upper bound of delay is also provided.

[0050] In the control plane, this embodiment proposes an implementation scheme based on a centralized network controller (CNC) and the NETCONF / YANG protocol stack, using the TSN standard architecture. The CNC is responsible for calculating new parameters and coordinating updates across multiple devices. Through NETCONF's transaction and locking mechanisms, the atomicity of parameter and credit status updates within the switch is ensured. Combined with the IEEE 802.1AS Precision Time Protocol (gPTP), a synchronized global update time can be specified for multiple switches, enabling path-level coordinated reconfiguration.

[0051] Below, in conjunction with Figure 2 , Figure 3 , Figure 4 The present application provides a detailed description and explanation of the solutions in its embodiments, along with specific application examples.

[0052] (I) System Model and Symbol Conventions 1.1) Flow Model and Arrival Curve Consider using the output port of a certain TSN switch A single CBS traffic class (e.g., AVB Class A). Input traffic satisfies token bucket arrival curve constraints:

[0053] in: Average arrival rate; For sudden size.

[0054] 1.2) Physical Properties and Interference Model Assume the port physical rate is The interference of high-priority TT flow and low-priority BE flow on this flow is equivalent to an inherent physical delay term. :

[0055] in , These are the maximum frame lengths for the TT stream and the BE stream, respectively.

[0056] 1.3) CBS Parameters and Steady-State Service Curve CBS behavior is determined by the idle slope (recorded as) ) and sending slope Decision. Assuming global synchronization time... At this point, the system switches the parameters from configuration A to configuration B: a) Status A (Old configuration, ): The corresponding steady-state minimum service curve ; b) Status B (New Configuration) ): The corresponding steady-state minimum service curve .

[0057] In fixed configuration Under these conditions, CBS's steady-state service curve adopts the standard rate-latency form:

[0058] in

[0059] (ii) Consistent State-Parameter Reconfiguration Mechanism 2.1) Atomic Reconstruction Operation This embodiment is in At time t, the parameter update is modeled as an atomic state transition, which includes at least: 1) CBS's from Updated to (and corresponding) ); 2) Synchronize and update credit status This ensures consistency with the new parameter configuration and avoids mismatches between historical states and new parameters.

[0060] The credit status before the update was:

[0061] The updated credit status is as follows:

[0062] 2.2) Strategy I: Forced Zeroing As a baseline implementation, this embodiment defines the Zeroing strategy as follows: 1) Update rules:

[0063] This strategy will place CBS in Resetting the time interval to the state equivalent to "the queue has just been served" has the advantages of being simple to implement and intuitive for formal analysis.

[0064] 2) The physical meaning includes: a) When At that time, negative credit is cleared, the queue is granted early sending qualification, which may cause additional bursts to downstream nodes; b) When When positive credits are cleared, it is equivalent to erasing the time window that should have been able to send continuously, resulting in a loss of service for this stream in the transient state.

[0065] 2.3) Strategy II: Scaling To reduce service loss and additional disruptions, this embodiment proposes a preferred scaling strategy, the core principle of which is: Time-to-Zero Invariant: the physical time required for credit to return to zero or recover to 0 remains unchanged before and after parameter reconfiguration.

[0066] Under this principle, the ideal scaling credit value is defined. for:

[0067] To satisfy the upper and lower bound constraints of credit in the hardware implementation The final update rules are as follows:

[0068] in:

[0069] The above design achieves the following: a) For The recovery period, ensuring recovery time

[0070] The process remains unchanged before and after the reassignment, thus smoothly continuing the original "waiting period"; b) For The exhaustion period ensures continuous transmission duration.

[0071] The data remains unchanged before and after reconfiguration, thus smoothly continuing the original "continuous transmission period".

[0072] The scaling strategy can significantly reduce transient disturbances caused by parameter mismatch without triggering upper and lower bound truncation.

[0073] (III) Analysis of Time Penalty and Transient Service Curve To formally characterize the service loss caused by parameter updates, this embodiment introduces the concept of a time penalty.

[0074] 3.1) Definition of Time Penalty Time penalty This indicates the amount of time it takes for the equivalent service curve to shift to the right due to the reconfiguration operation.

[0075] remember:

[0076] Then, for different strategies, the time penalty can be defined as: a) Zeroing strategy:

[0077] b) Scaling strategy:

[0078] That is, a penalty is only incurred if the scaled credit exceeds the upper bound and is truncated; otherwise, no penalty is imposed. .

[0079] 3.2) Equivalent Transient Service Curve After introducing the time penalty, the parameters are updated ( The equivalent minimum service curve of () can be expressed as:

[0080] This means that the impact of reconfiguration is reflected in the service curve as an additional latency term. This allows for unified analysis within the standard network computation framework.

[0081] 3.3) Lower Bound of Time-Varying Servers To analyze the overall service capability across update times, this invention constructs a time-varying server. , indicating in the interval The minimum service volume provided. Based on the concatenation property of service curves, we have:

[0082] in This is a minimum convolution operation. Further, a lower bound of the Rate-Latency form can be obtained:

[0083] It can be seen that the equivalent service across reconfiguration points can still be provided by a rate of Delay is The lower bound of the server.

[0084] (iv) Transient delay and additional burst limits 4.1) Transient worst-case delay Based on the above time-varying server lower bound and arrival curve:

[0085] The maximum transient queuing delay can be obtained. Upper bound:

[0086] This formula shows that transient delay consists of the following three parts: 1) Inherent physical delay ; 2) Time penalty caused by parameter reconfiguration ; 3) Sudden burst from the token bucket Difference with available service speed The time required to empty the queue is determined.

[0087] Under the Zeroing strategy, because Follow The instantaneous delay is significantly increased with increasing speed, and its transient delay limit is relatively conservative; under the scaling strategy, if truncation is not triggered, it can achieve... This allows for a tighter transient delay limit without sacrificing security.

[0088] 4.2) Compact Boundaries in Bandwidth Expansion Scenarios In bandwidth expansion scenarios ( If still using As for service rate, the queue emptying acceleration effect brought about by capacity expansion will be ignored, resulting in an overly conservative limit. Based on the assumption in actual engineering that "bursts are mainly served after reconfiguration", this embodiment further derives a tight upper bound for capacity expansion scenarios.

[0089] Under the following conditions: 1) Sudden events causing maximum latency Mainly in Then arrive and be served; 2) Transient "service pauses" caused by parameter reconfiguration have been penalized with time penalties. Complete absorption; The maximum transient delay can be further tightened to:

[0090] This limit more accurately reflects the increase in queue emptying rate after bandwidth expansion, which is conducive to fully releasing the performance benefits brought by expansion while ensuring security.

[0091] 4.3) Additional burst effects on other flows When in During the recovery period, parameter updates may be performed, and reconfiguration could shorten the quiescent period of the CBS queue, thus causing additional bursts of traffic to other flows such as Best-Effort. Therefore, this embodiment provides a solution for dealing with these additional bursts. Upper bound analysis.

[0092] For the Zeroing strategy, when When the queue resumes transmission ahead of schedule, its physical rate The additional burst upper bound that can be generated is:

[0093] For the scaling strategy, without triggering the credit upper bound truncation (i.e. Under these conditions, the original silence time can be kept unchanged, thus theoretically achieving:

[0094] Therefore, under the same conditions, the Scaling strategy proposed in this embodiment is not only more beneficial to the transient delay of the target flow, but also causes less additional burst interference to other flows, making it more suitable as the preferred engineering solution.

[0095] (v) Control plane implementation scheme This embodiment not only provides a transient safe reconfiguration mechanism at the data plane level, but also provides a practical implementation method at the control plane level.

[0096] 5.1) Control plane requirements To achieve the above atomic reassortment mechanism, the control plane needs to meet the following requirements: 1) Atomicity: Within a single switch, Updates and The update of a state must be an indivisible single operation, and intermediate states that are partially updated are not allowed. 2) Synchronization: When a flow needs to adjust its bandwidth simultaneously on multiple switches along the path, the synchronization of bandwidth on each switch... Alignment needs to be based on a global clock (such as IEEE 802.1AS); 3) Modeling Capability: The control plane protocol needs to be able to directly model and modify CBS queues. , , Parameters, not just abstract flow table entries.

[0097] 5.2) Implementation architecture based on CNC + NETCONF / YANG This embodiment preferably uses the centralized network controller (CNC) and NETCONF / YANG protocol stack defined in the TSN standardized architecture to implement the control plane part, mainly for the following reasons: a) NETCONF is a standard protocol for device configuration and status management, natively supporting transactions and locking mechanisms; b) The YANG model can precisely define CBS queue parameters (including...) (and credit upper and lower limits); c) Through <candidate> / <running>Data storage and <commit>The operation can achieve transaction semantics similar to ACID; It can be combined with the precise time synchronization provided by IEEE 802.1AS to achieve time-stamped synchronous submission.

[0098] 5.3) Control plane implementation sequence A single flow on several switches Taking capacity expansion as an example, one implementation sequence of this embodiment is as follows: 1) Preparation phase (CNC calculation): a) The CNC receives a bandwidth adjustment request from the application layer; b) The CNC calculates the new [data / mechanics] based on the overall resource status. ; c) The CNC uses the transient service curve and latency limits given in this embodiment to evaluate whether the WCD during the reconfiguration process is within an acceptable range.

[0099] 2) Locking Phase: a) CNC for the target switch (or set of switches) <candidate>Data storage initiates NETCONF <lock>ask; b) Ensure that there is no third-party configuration interference during the entire reconfiguration process.

[0100] 3) Reconfigure phase: a) CNC passing <edit-config>Target port / queue , , Wait for parameter updates to be written <candidate>; b) This configuration contains the necessary parameters for the data plane to execute the Zeroing or Scaling strategy.

[0101] 4) Commit Phase: a) CNC passing <commit>Trigger configuration from <candidate>Atomically applied to <running>; Optionally, CNC specifies an "execution timestamp" based on a global clock, enabling multiple switches to perform updates at the same absolute time; b) The internal implementation of the switch must ensure that execution... <commit>Simultaneously: b.1) Register updated to ; b.2) Update according to Zeroing or Scaling rules ; b.3) Ensure that the entire operation is completed within a single gated cycle or a single clock domain.

[0102] 5) Unlocking Phase: CNC passed <unlock>Release against <candidate>The device is locked and will then be allowed to accept other configuration requests.

[0103] Through the above process, this embodiment implements an "atomic parameter reconfiguration" mechanism at the control plane level that strictly corresponds to the transient model of the data plane.

[0104] (vi) Other embodiments and extensions 6.1) Multi-queue and multi-type stream expansion The mechanism in this embodiment is not only applicable to a single CBS queue, but can also apply the same reconfiguration rules individually or collaboratively to multiple CBS queues (e.g., Class A and Class B) on the same port. For multi-stream scenarios, new reconfiguration rules for each type of stream can be uniformly calculated at the CNC layer. And complete the coordinated reconfiguration of multiple queues through a single transactional update.

[0105] 6.2) Multi-hop paths and end-to-end analysis For end-to-end paths spanning multiple switches, this embodiment introduces a corresponding time penalty and transient service curve at each hop, and performs cascaded analysis on multiple nodes within a network calculus framework to obtain the upper bound of the end-to-end transient WCD. Correspondingly, CNC can coordinate multiple switches at the path level. This is to further reduce transient disturbances.

[0106] 6.3) Strategy Selection and Adaptive Control In certain scenarios, to further improve robustness, this embodiment can also be based on the current... , , It also allows for automatic selection of either Zeroing or Scaling strategies for transient WCD, or the setting of a conservative cutoff threshold for the Scaling strategy, thereby striking a balance between performance and security.

[0107] In summary, this embodiment proposes a transient safety dynamic reconfiguration method for TSN CBS. From system modeling, atomic state-parameter reconstruction mechanism, time penalty term and time-varying server analysis, to the control plane implementation process based on CNC + NETCONF / YANG, a complete set of formally verifiable and engineering-applicable technical solutions is formed. Compared with existing technologies, the method of this embodiment has at least the following beneficial effects: 1) Transient security is provable: Through time penalty terms and time-varying server models, this embodiment provides calculable transient service curves and latency upper bounds for online adjustment of CBS parameters, avoiding the blind spot of "only looking at steady state" in analysis; 2) Uninterrupted service and guaranteed continuity: Bandwidth adjustment is not required through destructive methods such as "deleting old streams and creating new streams". Instead, parameter reconfiguration is completed without interrupting the transmission of service frames. 3) Tighter latency limits: The preferred scaling strategy can significantly reduce time penalties in most engineering scenarios. This yields a tighter transient WCD bound, which has a significant performance advantage compared to the simple Zeroing strategy; 4) Controllable impact on other services: Through additional burst upper bound analysis, the burst interference caused to other flows such as Best-Effort can be quantitatively constrained, which facilitates system-level security assessment; 5) Easy to implement in engineering: This embodiment selects the CNC + NETCONF / YANG solution that conforms to the TSN standard architecture at the control plane level, and uses NETCONF's transaction capabilities to achieve atomicity and synchronization of parameter updates, which is easy to integrate into the existing TSN management system.

[0108] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0109] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0110] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0111] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0112] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0113] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0114] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented in the embodiments of this program product are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments. The executable computer program code or "code" used to perform the various embodiments can be written in high-level programming languages ​​such as C, C++, Python, Smalltalk, Java, JavaScript, Visual Basic, Structured Query Language (e.g., Transact-SQL), Perl, or in various other programming languages.

[0115] The embodiments described in 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 by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0116] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0117] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0118] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0119] The terms "first," "second," "third," "fourth," etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0120] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0121] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0122] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0123] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0124] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0125] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.< / candidate> < / unlock> < / commit> < / running> < / candidate> < / commit> < / candidate> < / lock> < / candidate> < / commit> < / running> < / candidate>

Claims

1. A method for dynamic parameter reconfiguration for credit-based shapers, characterized in that, The method, applied to switches in a Time-Sensitive Network (TSN), includes: At the global update time for adjusting the bandwidth guarantee parameters of the credit-based shaper CBS queue, an atomic state-parameter reconstruction operation is performed; the atomic state-parameter reconstruction operation is configured to take effect synchronously within the same scheduling cycle, including: Update the idle slope parameter of the CBS queue from the first value to the second value; Based on the credit status value before the update, the first value, and the second value, the credit status value of the CBS queue is synchronously updated to a new credit status value that matches the second value, so as to prevent the CBS queue from being in an intermediate state where the second value and the credit status value before the update coexist.

2. The method according to claim 1, characterized in that, The step of synchronously updating the credit status value of the CBS queue to a new credit status value that matches the second value, based on the previous credit status value, the first value, and the second value, includes: The previous credit status value is reset to zero, which is then used as the new credit status value.

3. The method according to claim 1, characterized in that, The step of synchronously updating the credit status value of the CBS queue to a new credit status value that matches the second value, based on the previous credit status value, the first value, and the second value, includes: Based on the stage of credit change in the credit status value before the update, the credit status value before the update is scaled proportionally to obtain a scaled credit value. Wherein, if the credit status value before the update is negative, the scaled credit value is equal to the credit status value before the update multiplied by the ratio of the second value to the first value; If the credit status value before the update is non-negative, the scaled credit value is equal to the credit status value before the update multiplied by the ratio of the first difference to the second difference, where the first difference is the difference between the port physical rate and the second value, and the second difference is the difference between the port physical rate and the first value. Compare the scaled credit value with the upper and lower bounds of the credit value corresponding to the second value. If the value exceeds the range, the boundary value is taken; otherwise, the original value is kept, and the new credit status value is obtained.

4. The method according to claim 2 or 3, characterized in that, The method further includes: Based on the credit status value before the update, the first value, and the second value, determine the time penalty introduced due to performing the atomic state-parameter reconstruction operation; Based on the time penalty and the steady-state service curve determined based on the second value, the transient minimum service curve of the CBS queue after the global update time is determined.

5. The method according to claim 4, characterized in that, The determination of the time penalty introduced by performing the atomic state-parameter reconstruction operation based on the credit status value before the update, the first value, and the second value includes: When a strategy of resetting the credit status value to zero is adopted, the time penalty is equal to the positive part of the credit status value before the update divided by the difference between the port physical rate and the first value. When a scaling strategy is used, the time penalty is equal to the portion of the scaling credit value that exceeds the credit upper bound corresponding to the second value divided by the difference between the port physical rate and the second value. If the scaling credit value does not exceed the credit upper bound, the time penalty is zero.

6. The method according to claim 4, characterized in that, The method further includes: Based on the time-varying server model spanning the global update time, the arrival curve constraint corresponding to the CBS queue, and the transient minimum service curve, the worst-case upper bound of the delay of the CBS queue during the parameter reconfiguration transient process is determined.

7. The method according to claim 6, characterized in that, Determining the worst-case upper bound of the delay for the CBS queue during the parameter reconfiguration transient process includes: If the second value is greater than the first value and a preset condition is met, the worst-case latency upper bound is determined by the queue emptying rate calculated based on the second value. The preset condition includes that the traffic burst that causes the maximum latency arrives mainly after the global update time.

8. The method according to claim 1, characterized in that, The method further includes: Assess the additional burst impact of the atomic state-parameter reconstruction operation on other priority traffic on the port to which the CBS queue belongs; When the credit status value before the update is negative, if a strategy of resetting the credit status value to zero is adopted, the upper bound of the additional burst impact is the ratio of the port physical rate multiplied by the absolute value of the credit status value before the update to the first value.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 8.

Images