A method, system, medium, product, and equipment for processing traffic data.

By sorting the total traffic within historical billing periods and lowering the baseline billing line, the problem of business teams struggling to quantify the gap between scheduling schemes and theoretical optimalities was solved. This enabled the optimization of idle bandwidth and cumulative peak-shaving traffic, providing quantifiable optimization targets and reducing billing costs.

CN122891385BActive Publication Date: 2026-11-17HANGZHOU YOUYUN TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202611364312.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2026-09-04
Publication Date
2026-11-17
Estimated Expiration
2046-09-04

AI Technical Summary

Technical Problem

Business teams struggle to quantify the gap between historical scheduling schemes and theoretical optimal solutions, making it difficult to define quantitative targets for the next round of optimization, determine whether the current strategy is approaching its limit, and miss opportunities to further reduce costs.

Method used

By sorting the total traffic within the historical billing period and dividing it into free and billing windows, a baseline billing line is determined. The baseline billing line is then lowered and peak-shaving traffic is accumulated until the accumulated peak-shaving traffic equals or exceeds the idle bandwidth, thus obtaining the ideal billing line and guiding the optimization of traffic scheduling strategies.

Benefits of technology

It optimizes idle bandwidth and cumulative peak-shaving traffic, makes full use of the free window, provides quantifiable optimization targets, guides the improvement of traffic scheduling strategies, and reduces billing costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122891385B_ABST
    Figure CN122891385B_ABST
Patent Text Reader

Abstract

This application discloses a traffic data processing method, system, medium, product, and device. The method includes: sorting the total traffic of services at multiple sampling times within a historical billing period; dividing each sampling time into a free window or a billing window based on the sorting; determining a baseline billing line based on the billing window, whereby the baseline billing line characterizes the total traffic of services used as the basis for billing within the billing window; determining the cumulative unused idle bandwidth of the system within the free window; adjusting the baseline billing line at least once, and accumulating the peak-shaving traffic in each round to obtain the cumulative peak-shaving traffic; the peak-shaving traffic is the difference between the first excess and the second excess; if the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth, an ideal billing line is obtained. The ideal billing line obtained by adjusting the baseline billing line in this application can provide optimization guidance for traffic scheduling strategies.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of mobile communication technology, and in particular to a traffic data processing method, system, medium, product, and device. Background Technology

[0002] With the development of the Internet, services such as CDN, short videos, and live streaming rely on large-scale network traffic. This traffic is usually carried by multiple physical or logical ports provided by operators or data centers. Services are contracted for bandwidth per port and use various billing methods, among which 95% billing is widely used.

[0003] The 95 billing method works as follows: within a billing cycle, the bandwidth of each port is sampled every 5 minutes. After removing the highest 5% of sample points, the remaining maximum value is the bandwidth that needs to be billed. This means that each port has a "free peak shaving" window of about 5%. As long as the instantaneous traffic peaks can be squeezed into each window across time and port, and the peaks of each port can be staggered and fall into their respective culprit sets, the overall billing bandwidth will be significantly reduced compared to when there is no scheduling. Therefore, the business team will continuously iterate the scheduling strategy to make the billing bandwidth approach the theoretical lower limit.

[0004] However, in actual operations, business teams often rely on experience to iterate scheduling strategies, making it difficult to quantify the gap between historical scheduling schemes and theoretical optimal ones. This makes it impossible to clearly define the quantitative target for the next round of optimization, and also makes it difficult to judge whether the current strategy has approached its limit, which may lead to missing the opportunity to further reduce costs. Summary of the Invention

[0005] In view of the above-mentioned defects or deficiencies in the prior art, it is desirable to provide a traffic data processing method, system, medium, product and device that can make idle bandwidth close to the cumulative peak-shaving traffic by lowering the baseline billing line, thereby making full use of the idle bandwidth of the free window, obtaining an ideal billing value, and evaluating the traffic scheduling strategy of the historical billing period based on the ideal billing value, thereby providing guidance for the optimization of traffic scheduling strategy.

[0006] In a first aspect, this application provides a traffic data processing method, applied to a traffic data processing system with network traffic billing functionality, the method comprising: The total traffic of services at multiple sampling times within the historical billing period is sorted. Based on the sorting, each sampling time is divided into a free window or a billing window. The baseline billing line is determined based on the billing window. The total traffic of services is the total traffic of services actually processed by the system. The baseline billing line is used to characterize the total traffic of services used as the basis for billing within the billing window. Determine the total unused idle bandwidth within the free window; The baseline billing line is lowered at least once, and the peak-shaving traffic in each round is accumulated to obtain the cumulative peak-shaving traffic. The peak-shaving traffic is the difference between the first excess and the second excess. The first excess is the sum of the differences between the total traffic of each service exceeding the baseline billing line after the current round of reduction and the baseline billing line after the reduction in the billing window. The second excess is the sum of the differences between the total traffic of each service exceeding the baseline billing line before the current round of reduction and the baseline billing line before the reduction in the billing window. If the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth, an ideal billing line is obtained.

[0007] Secondly, this application also provides a traffic data processing system, the system comprising: The partitioning module is used to sort the total traffic of services at multiple sampling times within the historical billing period. Based on the sorting, each sampling time is divided into a free window or a billing window, and a baseline billing line is determined based on the billing window. The total traffic of services is the total traffic of services actually processed by the system. The baseline billing line is used to characterize the total traffic of services used as the basis for billing within the billing window. The idle module is used to determine the system's cumulative unused idle bandwidth within the free window; The adjustment module is used to adjust the baseline billing line at least once and accumulate the peak-shaving traffic in each round to obtain the cumulative peak-shaving traffic. The peak-shaving traffic is the difference between the first excess and the second excess. The first excess is the sum of the differences between the total traffic of each service exceeding the baseline billing line after the current round of adjustment and the baseline billing line after the adjustment in the billing window. The second excess is the sum of the differences between the total traffic of each service exceeding the baseline billing line before the current round of adjustment and the baseline billing line before the adjustment in the billing window. The optimization module is used to obtain the ideal billing line when the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth.

[0008] Thirdly, this application also provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the method of any one of the first aspects above.

[0009] Fourthly, this application also provides a computer program product containing instructions that, when executed, perform any of the methods described in the first aspect above.

[0010] Fifthly, this application also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the method of any one of the first aspects above. Attached Figure Description

[0011] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart illustrating the method of this application in one embodiment; Figure 2 This is a schematic diagram illustrating the descending order of the total traffic at each sampling time in one embodiment of the present application; Figure 3 In one embodiment, this application fully aligns the traffic peaks of each port within the historical billing period to obtain a schematic diagram of the baseline billing line; Figure 4 This is a schematic diagram illustrating how the baseline billing line is lowered in one embodiment to make full use of idle bandwidth. Detailed Implementation

[0012] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0013] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The application will now be described in detail with reference to the accompanying drawings and embodiments. Furthermore, the term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The terms "first" and "second," etc., in the specification and claims of the embodiments of this application are used to distinguish different objects, not to describe a specific order of objects.

[0014] With the development of the Internet, services such as CDN, short videos, and live streaming rely on large-scale network traffic. This traffic is usually carried by multiple physical or logical ports provided by operators or data centers. Services are contracted for bandwidth per port and use various billing methods, among which 95% billing (percentile billing) is widely used.

[0015] The 95 billing system samples bandwidth for each port at a fixed granularity (e.g., every 5 minutes) within the billing cycle. After removing the top 5% of samples, the maximum value among the remaining samples is used as the billing bandwidth for that cycle. Intuitively, each port has a sampling period of about 5% within the cycle that can be considered a "peak-shaving, non-billing" window. If traffic is reasonably scheduled across the timeline and multiple ports, instantaneous peaks can be shifted into these windows for each port, thereby reducing the billing bandwidth. In multi-port scenarios, if the peaks for each port are staggered in time and fall into their respective removeable sampling sets, the approximately 5% "free" period of each port can be utilized simultaneously, making the overall cost lower than the simple summation of the port usage without scheduling. Therefore, the service continuously iterates the scheduling strategy.

[0016] However, businesses often struggle to directly answer how good the current strategy is, how much room there is for improvement relative to the theoretical optimum, and what quantitative metrics should be used as the target for the next round of optimization. Therefore, given the historical total traffic patterns and constraints on each port, it is necessary to calculate the theoretically achievable 95% billing level through optimal scheduling, and then compare it with the actual strategy output.

[0017] Based on this, embodiments of this application provide a traffic data processing method, system, medium, product, and device that can perform peak shaving on the total traffic of each service in the billing window by lowering the baseline billing line, so as to make full use of the idle bandwidth of the free window, thereby calculating the ideal billing line, and providing guidance for optimizing traffic scheduling strategies based on the ideal billing line.

[0018] In a first aspect, this application provides a traffic data processing method, applied to a traffic data processing system with network traffic billing functionality, the method comprising: S1. Sort the total traffic of services at multiple sampling times within the historical billing period. Based on the sorting, divide each sampling time into a free window or a billing window, and determine the baseline billing line based on the billing window. The total traffic of services is the total traffic of services actually processed by the system. The baseline billing line is used to characterize the total traffic of services used as the basis for billing within the billing window. Specifically, the billing rule of this application can be the 95 billing rule; the system has multiple 95 billing ports, and at each sampling time, the total service traffic of the system is the sum of the service traffic actually processed by the multiple 95 billing ports at that sampling time.

[0019] In the 95 billing rule, during each billing cycle, for each billing port, the sampling time with the highest business processing traffic is usually excluded from the billing. These sampling times are called the free window for that port. Therefore, in the 95 billing rule, the key to cost control is to ensure that as much "high business processing traffic" as possible falls within the free window, and that the port bandwidth limit corresponding to the free window can be fully utilized.

[0020] Traffic scheduling strategy refers to the rule of independently implementing 95% billing on each port (that is, after removing the highest 5% of sampling points of each port, the remaining maximum value is taken as the billing bandwidth of that port), so as to reasonably distribute the total traffic of services at each sampling time to each port and make full use of the bandwidth limit of each port; by staggering the peak periods of different ports, the peak of each port should fall within its own 5% free window as much as possible, thereby reducing the sum of the billing bandwidth of each port, that is, the overall cost of multi-port.

[0021] It is worth mentioning that optimal scheduling is to minimize the overall 95% billing value while satisfying the upper and lower limits of each port at each time and the conservation of traffic. The 95% billing traffic under optimal scheduling is referred to as the ideal billing line in this application.

[0022] Because the 95 billing rules only allow each port to remove about 5% of the highest traffic, and the total traffic included in the billing is determined by the maximum value of the remaining portion, in order to minimize billing, it is necessary to prioritize using the free window to carry the highest segment of the total traffic and make full use of the bandwidth limit of each port within the free window.

[0023] Therefore, in order to calculate the ideal billing line in this application, the following two conditions must be met: First, peak traffic should be shaving, that is, the traffic consuming the free window of each port should come from the segment with the highest total business traffic as much as possible, so as to avoid wasting the free quota on relatively low peaks; Second, the free window bandwidth should be fully utilized, that is, each port should try to fully utilize the bandwidth limit allowed at that moment when consuming free points, so as to avoid waste due to the free quota not being fully utilized.

[0024] Based on this, one possible implementation is to sort the total traffic of services at multiple sampling times within a historical billing period, and then allocate each sampling time to a free window or a billing window based on the sorting, specifically as follows: The total traffic volume at multiple sampling times within the historical billing period is sorted from high to low. For the total traffic of services ranked at the top of the preset number of positions, the sampling time of the total traffic is allocated to the free window; The remaining sampling time is allocated to the billing window.

[0025] Based on the two conditions mentioned above, this application treats the n ports of the system as a single aggregated port and assumes that the peak traffic of each port is perfectly aligned, resulting in a simplified model. Under this simplified model, the free windows of all ports completely overlap; therefore, the top 5% of the total business traffic is the union of the free traffic moments of each port.

[0026] For example, suppose the system has n 95 billing ports, and the total traffic is sampled at fixed intervals during the historical billing period to obtain the sequence Flow = {v1, v2, …, vm (m is the sampling time).

[0027] The sum of the service traffic at each port at the t-th sampling time should equal the total service traffic v of the system at that time. t The top 5% of total business traffic will be used as the free window.

[0028] One possible implementation involves determining the baseline billing line based on the billing window, specifically as follows: The highest total business traffic within the billing window is used as the baseline for billing.

[0029] S2. Determine the total unused idle bandwidth within the free window; One possible implementation involves determining the system's accumulated unused idle bandwidth within the free window, specifically: Determine the difference between the system's total bandwidth limit and the actual total traffic processed at each sampling moment within the free window; The system accumulates the total unused idle bandwidth by summing up the multiple differences within the free window.

[0030] The following explains the total bandwidth limit: Assume the system has n billed ports (95), and the bandwidth limit for the i-th port is h. i The total bandwidth limit can be expressed as H = h1 + h2 + ... + h n .

[0031] It is worth mentioning that, "h i It can be a time-varying constraint interval or a constant upper limit.

[0032] S3. Lower the baseline billing line at least once, and accumulate the peak-shaving traffic in each round to obtain the cumulative peak-shaving traffic; the peak-shaving traffic is the difference between the first excess and the second excess; the first excess is the sum of the differences between the total traffic of each service exceeding the baseline billing line after the current round of reduction and the baseline billing line after the reduction in the billing window; the second excess is the sum of the differences between the total traffic of each service exceeding the baseline billing line before the current round of reduction and the baseline billing line before the reduction in the billing window; Peak-shaving traffic can also be understood as, in order to reach the new baseline billing line, the traffic that needs to be further reduced from the total traffic of each service exceeding the new baseline billing line within the billing window in the current round compared to the previous round; for easier understanding, combined with... Figure 4 Peak shaving flow rate explanation: In Figure 4In this context, the peak shaving flow rate in each round can be viewed as the change in area of ​​B after the baseline billing line is lowered from its current position. The first excess is essentially the area of ​​B after the current reduction, and the second excess is essentially the area of ​​B before the current reduction. The difference between the two is the change in area of ​​B after the current reduction.

[0033] One possible implementation is to lower the baseline billing line in each round based on a preset step size.

[0034] Another possible implementation, S3, is as follows: The total traffic of each service in the billing window is sorted from high to low to obtain the first sequence; Starting with the second-ranked total traffic in the first sequence, the total traffic of each service in the first sequence will be used as the target value for the reduction of the current benchmark billing line in this round. Based on the target value for each round of reduction, the peak shaving flow rate for this round is obtained. The peak shaving flow rates for each round are accumulated to obtain the cumulative peak shaving flow rate.

[0035] S4. If the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth, the ideal billing line is obtained.

[0036] The following is an example of the downsizing process of S3: Assumptions: There are 3 sampling points in the free window (top 5%), and the total bandwidth limit of the port is H=100.

[0037] Total business traffic during the free period: 90, 85, 80; The free window's idle bandwidth A=(100-90)+(100-85)+(100-80)=10+15+20=45.

[0038] The total traffic volume at each sampling time of the charging window is sorted from high to low to obtain the first sequence A1: 75, 70, 65, 60, 55... The specific process of lowering the baseline billing line is as follows: Round 1: The target value for lowering the base billing line is 70, which is the second value in the first sequence.

[0039] The first excess in this round = the sum of the differences between the flow rate above 70 (i.e., 75) in A1 and 70 = 5; The second excess in this round = the sum of the differences between the flow rate above 75 in A1 and 75 = 0; Peak shaving flow rate in this round = first excess amount Second excess = 5 0 = 5; The cumulative peak shaving flow rate for this round is 5 < 45, continue.

[0040] Round Two: The target value for lowering the base billing line is 65, which is the third value in the first sequence.

[0041] The baseline rate before this round of adjustments was 70. The second excess in this round = the difference between the flow rate above 70 (i.e., 75) in A1 and 70 = 5; After this round of adjustments, the base rate is 65. The first excess in this round = the difference between the flow rate above 65 (i.e., 75, 70) in A1 and 65. The sum of the values ​​= (75-65) + (70-65) = 10 + 5 = 15; Peak shaving flow rate for this round = 15 5 = 10; The cumulative peak shaving flow rate for this round is 5 + 10 = 15 < 45, so continue.

[0042] Third round: The target value for lowering the base billing line is 60, which is the fourth value in the first sequence.

[0043] The baseline for this round of price reduction was 65. The second excess in this round = the sum of the differences between the flow rates above 65 (i.e., 75, 70) in A1 and 65 = (75-65) + (70-65) = 10 + 5 = 15; After this round of reductions, the base rate is 60. The first excess in this round = the sum of the differences between the flow rates above 60 in A1 (i.e., 75, 70, 65) and 60 = (75-60) + (70-60) + (65-60) = 15 + 10 + 5 = 30; Peak shaving flow rate for this round = 30 15 = 15; The cumulative peak shaving flow rate for this round is 15 + 15 = 30 < 45, so continue.

[0044] Round 4: The reduction is based on the fifth value in the first sequence—55—as the baseline for the tariff reduction. Target value.

[0045] The baseline for this round of price reduction was 60. The second excess in this round = the flow rate above 60 in A1 (i.e., 75, 70, 65) and 60 The sum of the differences = (75-60) + (70-60) + (65-60) = 15 + 10 + 5 = 30; After this round of reductions, the baseline rate is 55. The first excess in this round = the flow rate above 55 in A1 (i.e., 75, 70, 65, 60). The difference between 75 and 55 is (75-55)+(70-55)+(65-55)+(60-55)=20+15+10+5=50; Peak shaving flow rate for this round = 50 30 = 20; The cumulative peak shaving flow rate for this round is 30 + 20 = 50 ≥ 45, so stop.

[0046] S4. If the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth, the ideal billing line is obtained.

[0047] One possible implementation is to obtain the ideal billing line if the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth, specifically as follows: If the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth, the baseline billing line of the previous round immediately preceding the current round will be used as the ideal billing line.

[0048] In the previous example, the baseline billing line obtained in the third round—60—was taken as the ideal billing line.

[0049] The following combination Figures 2-4 The principle of obtaining an ideal billing line when the cumulative peak-shaving traffic is greater than or equal to the idle bandwidth is explained: It should be noted that this application is used to calculate the ideal billing line retrospectively based on the total service traffic at all sampling moments of the historical billing period after the end of the entire historical billing period, rather than to make predictions before or during the traffic occurrence. Therefore, the proof of the ideal billing line calculation principle is also based on retrospective deduction of complete historical data, and does not involve the prediction of future traffic or real-time scheduling decisions.

[0050] First, construct a simplified model, such as Figure 2 As shown, specifically, multiple ports are regarded as an aggregated port with a total bandwidth of H, and each port consumes its own free points at the same time. In this simplified model, there is no need for cross-port time scheduling.

[0051] like Figure 3 As shown, after sorting the sequence flow in descending order of sampled values, about 5% of the highest sampling times are divided into "free windows", and the remaining sampling times are billing windows. The maximum value of the total traffic in the billing window is the baseline billing line. It is worth mentioning that the baseline billing line refers to the natural 95 billing value obtained under the premise that the peaks of each port are completely aligned and the peaks are not actively staggered to different time periods.

[0052] Figure 3The obtained baseline billing line meets condition 1: the free window corresponds to the highest 5% of the total global service traffic, but may not meet condition 2: the total bandwidth H of each port corresponding to the free window is not fully utilized at each sampling time, and there is idle bandwidth. Figure 3 This is reflected in the presence of unfilled free space (gray area).

[0053] Secondly, implement multi-port peak shifting: each port does not need to consume free points simultaneously at the same sampling time; by shifting peak times, a portion of the free quota can be moved to later periods with continued peak traffic, thereby improving bandwidth utilization on free points. For example... Figure 4 As shown, it can be seen from Figure 4 The baseline billing line in the diagram is shifted downwards to obtain area B in the figure below. When the area of ​​area B is greater than or equal to the area of ​​area A, the bandwidth space of each port at the free point is fully utilized, which corresponds to a visual depiction of the ideal billing line.

[0054] Finally, regarding why the ideal billing line is obtained when region A ≥ region B, the specifics are as follows: Figure 4 The top 5% of sampled values ​​correspond to the highest free traffic window. For billing purposes, this can be seen as the majority of free points for each port falling within these high-traffic sampling times. Area A represents the portion of idle bandwidth remaining in the free allowance if all ports consume free traffic during these 5% of sampling times, even if the bandwidth limit is not fully utilized. Since peak-shaving for multiple ports doesn't require strictly limiting all free points to the top 5% of total traffic, some unused free capacity can be transferred to sampling points outside the top 5% but still at a relatively high level for continued peak shaving. Area B then represents the portion of actual traffic exceeding the new billing line at other billing sampling points after the billing line shifts downwards. This traffic needs to be transferred to the idle bandwidth within the free window through scheduling.

[0055] When the areas of region A and region B are equal (ideal case): the remaining slots in the first 5% can just take over the excess represented by region B, and the ports under the free window are neither idle nor overloaded; corresponding to the previous conditions 1 (peak shaving to target high traffic) and 2 (use the maximum limit during free periods as much as possible) being met at the same time, the baseline billing line at this time is the ideal billing line.

[0056] Furthermore, when the cumulative peak-shaving traffic (i.e., the cumulative area of ​​region B) is greater than that of region A, it indicates that the cumulative peak-shaving traffic generated by the reduction of the base billing line is sufficient to cover all the idle bandwidth within the free window. At this point, the bandwidth space of the free window has been fully utilized. Although further reducing the billing line can generate more peak-shaving traffic, there is no corresponding idle capacity to accommodate it. Therefore, the ideal billing line is to take the billing line when region B first reaches or exceeds that of region A. This fully utilizes the free bandwidth and avoids excessive peak shaving that leads to an excessively low billing line and distortion.

[0057] The following is an example of the execution code for S3: enter H: Maximum bandwidth for multi-port aggregation; Flow={v1,v2,…,v m}: The total flow sequence of each sampling point within the billing period.

[0058] Preprocessing: Sort by traffic in descending order Sort the Flow values ​​from highest to lowest to get Flow_sorted, and denote Flow_sorted[i] as the i-th sorted value. i The flow rate of 0 (starting from 0) sampling points satisfies Flow_sorted[0]≥Flow_sorted[1] ≥…≥Flow_sorted[m 1).

[0059] Step 1: Calculate the area of ​​region A (unused bandwidth not fully utilized during the free period). The baseline billing line is determined by approximately the first 5% of sampling times. Within the free window comprised of these first 5% of sampling times, if the total service traffic at a given time is lower than the aggregation limit H, the difference represents the idle bandwidth at that time that is "with available free bandwidth but not fully utilized." The idle bandwidth at each sampling time is then summed up to define region A. index 95 ← / / The dividing index of the baseline billing line in the sorting sequence, i.e., the sorting number SpaceA ← 0 For i ← 0 to index 95 1 Execution SpaceA ← SpaceA + (H Flow_sorted[i]) Step 2: Gradually lower the baseline billing line, accumulating coverage in area B until area A is covered. Using Flow_sorted[index] 95is the reference billing line. Starting from the reference billing line, the sequence values in the descending sequence of total service traffic are adjusted downward. Then, in the order of the descending sequence of total service traffic, the reference billing line is sequentially updated to the next lower traffic value, and the descending sequence values are adjusted downward again. Each downward adjustment corresponds to an increment of region B.

[0060] Curr95 ← Flow_sorted[index 95 / / Current candidate 95th percentile billing line height SpaceB ← 0 / / Cumulative area of region B For i ← index 95 to m 1 do SpaceB ← SpaceB + (Curr95 - Flow_sorted[i]) × (i - index 95 ) If SpaceB ≥ SpaceA, then Exit the loop / / B is sufficient to accommodate A, stop pressing down Otherwise Curr95 ← Flow_sorted[i] / / The billing line follows up to the current sampling water level to continue the trial Output When the loop ends, Curr95 is the ideal billing line. If SpaceB is still less than SpaceA after traversing to the end of the sequence, the Curr95 at the end is taken.

[0061] The logic of this code is explained as follows: First, calculate the unused idle bandwidth (region A) in the free window, then simulate pressing the reference billing line down step by step, and accumulate the excess (region B) that needs to be accommodated through off-peak scheduling during the pressing process. When B is sufficient to cover A, the corresponding billing line height is the ideal billing line.

[0062] The present application can calculate the theoretically achievable minimum 95th percentile billing index for scheduling, that is, the ideal billing line, based on the sampling sequence and the constraints of each port. In actual online scheduling (during the process), since future traffic cannot be predicted and the complete traffic ranking within the billing period cannot be obtained, the real-time strategy usually cannot reach this post-hoc optimal value.

[0063] In a possible implementation, the present application further includes: optimizing the traffic scheduling strategy of historical billing periods based on the ideal billing line.

[0064] This application uses the ideal billing line as an evaluation benchmark after the end of a historical billing period: it measures the merits of actual traffic scheduling strategies, the gap between quantification and the theoretical lower limit, and guides the optimization direction for the next period. Specifically, given the total historical traffic, billing port set, and port constraints at each sampling time within a historical billing period, the ideal billing line (i.e., the sum of the 95% billing values ​​for each port) can be calculated based on this application. This line is then compared with the actual 95% billing results obtained after the traffic scheduling strategy was implemented within the same historical billing period. This allows for evaluation of the scheduling strategy's effectiveness, quantification of optimizable space, and a clear quantitative upper limit reference for strategy iteration, enabling more targeted optimization of scheduling and further reduction of traffic costs.

[0065] For example, if the actual billing result is significantly higher than the ideal billing line, it indicates that there is too much idle bandwidth within the free window or that peak periods are not effectively staggered. Further analysis of the actual traffic distribution at each sampling time reveals that if peak traffic for multiple ports is concentrated in the same time period, there is an issue of insufficient peak period staggering. If the actual traffic within the free window is significantly lower than the port capacity, it indicates that the free quota is not being fully utilized. Based on these analytical results, the shortcomings of the traffic scheduling strategy within the historical billing cycle can be identified, allowing for targeted optimization.

[0066] Secondly, this application also provides a traffic data processing system, the system comprising: The partitioning module is used to sort the total traffic of services at multiple sampling times within the historical billing period. Based on the sorting, each sampling time is divided into a free window or a billing window, and a baseline billing line is determined based on the billing window. The total traffic of services is the total traffic of services actually processed by the system. The baseline billing line is used to characterize the total traffic of services used as the basis for billing within the billing window. The idle module is used to determine the system's cumulative unused idle bandwidth within the free window; The adjustment module is used to adjust the baseline billing line at least once and accumulate the peak-shaving traffic in each round to obtain the cumulative peak-shaving traffic. The peak-shaving traffic is the difference between the first excess and the second excess. The first excess is the sum of the differences between the total traffic of each service exceeding the baseline billing line after the current round of adjustment and the baseline billing line after the adjustment in the billing window. The second excess is the sum of the differences between the total traffic of each service exceeding the baseline billing line before the current round of adjustment and the baseline billing line before the adjustment in the billing window. The optimization module is used to obtain the ideal billing line when the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth.

[0067] It should be noted that although the operations of the method of the present invention are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all of the operations shown must be performed to achieve the desired result. On the contrary, the steps depicted in the flowchart may be performed in a different order. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0068] On the other hand, this application also provides a computer-readable storage medium, which may be included in a computer device or exist independently without being assembled into the computer device. The aforementioned computer-readable storage medium stores one or more programs that, when used by one or more processors, execute the methods described in this application. For example, it may execute... Figure 1 The steps of the method shown are as follows.

[0069] In another aspect, this application also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements... Figure 1 The steps of the method shown are as follows.

[0070] On the other hand, this application also provides a computer-readable storage medium, which may be included in a computer device or exist independently without being assembled into the computer device. The aforementioned computer-readable storage medium stores one or more programs that, when used by one or more processors, execute the methods described in this application. For example, it may execute... Figure 1 The steps of the method shown are as follows.

[0071] This application provides a computer program product including instructions that, when executed, cause the method described in this application to be performed. For example, it can execute... Figure 1 The steps of the method shown are as follows.

[0072] Specifically, according to embodiments of this application, Figure 1 The described process can be implemented as a computer software program. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowchart.

[0073] On another front, this application also provides a computer-readable storage medium, which may be included in an electronic device or exist independently without being assembled into the electronic device. The aforementioned computer-readable storage medium stores one or more programs that, when used by one or more processors, execute the methods of this application. For example, it may execute... Figure 1 The steps of the method shown are as follows.

[0074] 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). 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.

[0075] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for processing traffic data, characterized in that, Applied to traffic data processing systems with network traffic billing functionality, the method includes: The total traffic of services at multiple sampling times within the historical billing period is sorted. Based on the sorting, each sampling time is divided into a free window or a billing window. A baseline billing line is determined based on the billing window. The total traffic of services is the total traffic of services actually processed by the system. The baseline billing line is used to characterize the total traffic of services used as the basis for billing within the billing window. Within the free window, the system accumulates unused idle bandwidth. The baseline billing line is lowered at least once, and the peak-shaving traffic in each round is accumulated to obtain the cumulative peak-shaving traffic; the peak-shaving traffic is the difference between the first excess and the second excess; the first excess is the sum of the differences between the total traffic of each service exceeding the baseline billing line after the current round of reduction and the baseline billing line after the reduction in the billing window; the second excess is the sum of the differences between the total traffic of each service exceeding the baseline billing line before the current round of reduction and the baseline billing line before the reduction in the billing window; If the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth, an ideal billing line is obtained; If the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth, an ideal billing line is obtained, specifically: If the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth, the baseline billing line obtained after the adjustment in the previous round immediately preceding the current round will be used as the ideal billing line.

2. The method according to claim 1, characterized in that, The total traffic of services at multiple sampling times within the historical billing period is sorted, and each sampling time is assigned to a free window or a billing window based on the sorting, specifically as follows: The total traffic volume at multiple sampling times within the historical billing period is sorted from high to low. For the total traffic of services ranked at the top of the preset number of positions, the sampling time of the total traffic of services is divided into the free window; The remaining sampling time is allocated to the billing window.

3. The method according to claim 2, characterized in that, The baseline billing line is determined based on the aforementioned billing window, specifically as follows: The maximum total traffic volume in the billing window is used as the baseline billing line.

4. The method according to claim 3, characterized in that, The baseline billing line is lowered at least once, and the peak-shaving traffic from each round is accumulated to obtain the cumulative peak-shaving traffic, specifically as follows: The total traffic of each service in the billing window is sorted from high to low to obtain the first sequence; Starting with the second-ranked total traffic of services in the first sequence, each total traffic of services in the first sequence is used as the target value for the reduction of the current benchmark billing line in this round. Based on the target value for each round of reduction, the peak shaving flow rate for this round is obtained. The peak shaving flow rates for each round are accumulated to obtain the cumulative peak shaving flow rate.

5. The method according to claim 1, characterized in that, Within the free window, the system's accumulated unused idle bandwidth is determined as follows: Determine the difference between the system's total bandwidth limit and the actual total traffic processed at each sampling time within the free window; The cumulative unused idle bandwidth of the system is obtained by summing up the multiple differences within the free window.

6. A traffic data processing system, characterized in that, include: The partitioning module is used to sort the total traffic of services at multiple sampling times within a historical billing period, and to divide each sampling time into a free window or a billing window based on the sorting, and to determine a baseline billing line based on the billing window. The total traffic of services is the total traffic of services actually processed by the system, and the baseline billing line is used to characterize the total traffic of services used as the basis for billing within the billing window. An idle module is used to determine the cumulative unused idle bandwidth of the system within the free window; The adjustment module is used to adjust the baseline billing line at least once and accumulate the peak-shaving traffic in each round to obtain the cumulative peak-shaving traffic; the peak-shaving traffic is the difference between the first excess amount and the second excess amount; the first excess amount is the sum of the differences between the total traffic of each service exceeding the baseline billing line after the current adjustment and the baseline billing line after the adjustment in the billing window; the second excess amount is the sum of the differences between the total traffic of each service exceeding the baseline billing line before the current adjustment and the baseline billing line before the adjustment in the billing window; The optimization module is used to obtain an ideal billing line when the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth; When the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth, an ideal billing line is obtained. Specifically, when the cumulative peak-shaving traffic in the current round is greater than or equal to the idle bandwidth, the baseline billing line obtained after the adjustment in the previous round immediately adjacent to the current round is taken as the ideal billing line.

7. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method of any one of claims 1 to 5.

8. A computer program product comprising instructions that, when executed, perform the method of any one of claims 1 to 5.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the method of any one of claims 1 to 5.

Citation Information

Patent Citations

  • Business range adjustment method and device, electronic equipment and storage medium

    CN114006821A

  • Operator traffic scheduling optimization method, medium and system

    CN119676172A