CN117729164A - Dynamic bandwidth allocation system for four-port gigabit network card - Google Patents
Dynamic bandwidth allocation system for four-port gigabit network card Download PDFInfo
- Publication number
- CN117729164A CN117729164A CN202410177712.4A CN202410177712A CN117729164A CN 117729164 A CN117729164 A CN 117729164A CN 202410177712 A CN202410177712 A CN 202410177712A CN 117729164 A CN117729164 A CN 117729164A
- Authority
- CN
- China
- Prior art keywords
- network
- module
- bandwidth
- flow
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000005457 optimization Methods 0.000 claims abstract description 131
- 238000012544 monitoring process Methods 0.000 claims abstract description 110
- 238000011156 evaluation Methods 0.000 claims abstract description 71
- 238000004088 simulation Methods 0.000 claims abstract description 61
- 230000004044 response Effects 0.000 claims abstract description 42
- 230000010354 integration Effects 0.000 claims abstract description 29
- 238000004445 quantitative analysis Methods 0.000 claims abstract description 12
- 238000012300 Sequence Analysis Methods 0.000 claims abstract description 7
- 238000000034 method Methods 0.000 claims description 98
- 238000004458 analytical method Methods 0.000 claims description 88
- 238000005516 engineering process Methods 0.000 claims description 74
- 238000004422 calculation algorithm Methods 0.000 claims description 64
- 238000013468 resource allocation Methods 0.000 claims description 44
- 230000008569 process Effects 0.000 claims description 34
- 238000013507 mapping Methods 0.000 claims description 31
- 238000011217 control strategy Methods 0.000 claims description 30
- 230000006399 behavior Effects 0.000 claims description 26
- 230000008859 change Effects 0.000 claims description 23
- 238000003066 decision tree Methods 0.000 claims description 21
- 238000005206 flow analysis Methods 0.000 claims description 20
- 230000001932 seasonal effect Effects 0.000 claims description 19
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 14
- 238000007726 management method Methods 0.000 claims description 12
- 230000000694 effects Effects 0.000 claims description 11
- 238000013139 quantization Methods 0.000 claims description 11
- 238000007405 data analysis Methods 0.000 claims description 10
- 238000000354 decomposition reaction Methods 0.000 claims description 10
- 238000003012 network analysis Methods 0.000 claims description 9
- 230000003044 adaptive effect Effects 0.000 claims description 8
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 230000006872 improvement Effects 0.000 claims description 6
- 230000000737 periodic effect Effects 0.000 claims description 6
- 238000010223 real-time analysis Methods 0.000 claims description 6
- 230000004927 fusion Effects 0.000 claims description 5
- 230000007246 mechanism Effects 0.000 claims description 5
- 238000003909 pattern recognition Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 5
- 238000009499 grossing Methods 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 230000010485 coping Effects 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 claims description 3
- 238000013459 approach Methods 0.000 abstract description 4
- 230000005540 biological transmission Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 10
- 230000007774 longterm Effects 0.000 description 4
- 230000026676 system process Effects 0.000 description 4
- 230000003247 decreasing effect Effects 0.000 description 3
- 230000001934 delay Effects 0.000 description 3
- 238000010921 in-depth analysis Methods 0.000 description 3
- 238000007619 statistical method Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000015556 catabolic process Effects 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 235000003642 hunger Nutrition 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000000611 regression analysis Methods 0.000 description 2
- 230000037351 starvation Effects 0.000 description 2
- 238000010998 test method Methods 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 241001123248 Arma Species 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000008713 feedback mechanism Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention relates to the technical field of dynamic bandwidth allocation, in particular to a four-port gigabit network card dynamic bandwidth allocation system which comprises a prediction optimization decision module, a hierarchical dynamic adjustment module, a network simulation and evaluation module, an intelligent logic control module, a quantitative analysis and adjustment module, a geometric space optimization module, a dynamic bandwidth allocation module and a system integration and monitoring module. According to the invention, by applying the layered time window bandwidth allocation strategy, the system can efficiently adjust bandwidth allocation on different time scales (millisecond level, second level and minute level), and the response speed and accuracy to the sudden flow fluctuation are greatly improved. This approach optimizes network performance and resource utilization, especially when dealing with fine-grained traffic dynamics in high-speed network environments. The application of the time sequence analysis optimization bandwidth pre-allocation strategy enables the system to accurately predict the network usage mode and adjust bandwidth allocation in advance to adapt to predicted traffic changes.
Description
Technical Field
The invention relates to the technical field of dynamic bandwidth allocation, in particular to a four-port gigabit network card dynamic bandwidth allocation system.
Background
Dynamic bandwidth allocation techniques are part of the field of network communications, and are particularly concerned with the efficient management of network resources. This technique allows the network system to intelligently adjust and allocate bandwidth resources based on current traffic demands and network conditions. The key point is to maximize network efficiency, reduce congestion, and guarantee priority and quality of key data transmission. Such techniques are widely used in data centers, enterprise networks, and internet service providers to accommodate changing network requirements and optimize user experience.
A dynamic bandwidth allocation system for four gigabit network cards is a device designed for high-speed networks and has four gigabit rate network ports. The main purpose of this system is to improve the utilization and transmission efficiency of network resources by dynamically allocating bandwidth. Under the condition of dense or unbalanced network traffic, the bandwidth allocation of each port can be flexibly adjusted, the stability and reliability of the network performance are ensured, and the influence on the performance of the whole network due to overload of a certain port is avoided.
Although the prior art has achieved significant success in the area of dynamic bandwidth allocation, there are significant shortcomings in handling complex traffic dynamics in high-speed network environments. Firstly, the bandwidth management capability of the existing system on a fine granularity time scale is limited, and sudden microsecond flow fluctuation is difficult to efficiently process, so that the problem of network congestion or insufficient utilization of bandwidth resources occurs in a short time. Second, while conventional techniques are capable of handling general traffic patterns, they have limited effectiveness in predicting complex network usage patterns and long-term traffic trends, which limits their ability to pre-allocate bandwidth resources. Furthermore, conventional techniques also exhibit deficiencies in predicting bandwidth demands and identifying potential bottlenecks using discrete event simulations, which are difficult to adequately predict and accommodate complex changes in the network. The uncertainty and ambiguity handling of the network environment is also not fine enough in the decision process of bandwidth management, which limits the flexibility and accuracy of bandwidth allocation policies. Finally, the conventional technology also has limitation in global optimization configuration of bandwidth resources, and particularly in the aspect of converting the bandwidth allocation problem into a more efficient geometric space optimization problem, a large improvement space is still reserved.
Disclosure of Invention
The invention aims to solve the defects in the prior art, and provides a four-port gigabit network card dynamic bandwidth allocation system.
In order to achieve the above purpose, the present invention adopts the following technical scheme: the system comprises a prediction optimization decision module, a hierarchical dynamic adjustment module, a network simulation and evaluation module, an intelligent logic control module, a quantitative analysis and adjustment module, a geometric space optimization module, a dynamic bandwidth allocation module and a system integration and monitoring module;
the prediction optimization decision module predicts and analyzes future network loads by adopting an autoregressive moving average model and a seasonal decomposition trend, a seasonal and residual model based on historical and real-time network load data, and performs preliminary optimization allocation on bandwidth resources by using a linear programming method to generate an optimized bandwidth allocation plan;
the hierarchical dynamic adjustment module is used for coping with sudden flow changes by adopting a dynamic threshold adjustment algorithm based on an optimized bandwidth allocation plan, smoothing continuous flow trends by adopting a weighted moving average method, and performing bandwidth adjustment on multiple time scales to generate a hierarchical adjustment strategy;
The network simulation and evaluation module simulates a network flow scene by adopting a queue theoretical model and a network flow prediction algorithm based on current network configuration and historical flow data, and comprises the steps of modeling the network flow, predicting future bandwidth demands, identifying potential network bottlenecks and performance bottlenecks, and generating a network behavior simulation evaluation result;
the intelligent logic control module carries out intelligent evaluation on the network condition by adopting a fuzzy logic controller based on the network behavior simulation evaluation result, adjusts bandwidth allocation by utilizing a self-adaptive control strategy technology, automatically adjusts control parameters according to a preset performance target by continuously monitoring the change of the network condition, and generates a logic control strategy;
the quantitative analysis and adjustment module adopts decision tree classification based on a logic control strategy, utilizes decision trees to identify multiple types of flow modes by analyzing historical data and real-time data, utilizes a probability network analysis method, evaluates the influence of multiple decisions on bandwidth allocation by using a probability model, and performs quantitative analysis on the flow modes and bandwidth utilization data to generate a quantitative adjustment scheme;
the geometric space optimization module performs optimization of the bandwidth allocation scheme by adopting a convex optimization method based on a quantization adjustment scheme, maps the bandwidth allocation problem into a geometric space by utilizing a space mapping algorithm, and re-analyzes and optimizes the bandwidth allocation by virtue of a space relation and geometric characteristics to generate a geometric optimization result;
The dynamic bandwidth allocation module adopts a dynamic resource allocation algorithm to analyze real-time network data and forecast future flow trend based on a hierarchical adjustment strategy and a geometric optimization result, dynamically adjusts bandwidth allocation, and matches the change of network requirements to generate a self-adaptive bandwidth allocation scheme;
the system integration and monitoring module continuously tracks performance indexes of the system, including bandwidth utilization rate, flow distribution and response time, and analyzes the system performance in real time to generate a system comprehensive state overview based on an optimized bandwidth allocation plan, a hierarchical adjustment strategy, a network behavior simulation evaluation result, a logic control strategy, a quantitative adjustment scheme, a geometric optimization result and a self-adaptive bandwidth allocation scheme by adopting a comprehensive performance analysis method and a performance monitoring technology.
The invention improves, the bandwidth allocation plan comprises network load peak prediction data, daily flow pattern analysis and current bandwidth demand prediction, the hierarchical adjustment strategy comprises a sudden flow emergency response scheme, a stable flow maintenance strategy and a conventional flow balance adjustment, the network behavior simulation evaluation result comprises network bottleneck recognition, performance bottleneck analysis and flow allocation efficiency evaluation, the logic control strategy comprises an adaptive flow response scheme, a bandwidth allocation dynamic adjustment strategy and network stability guarantee measures, the quantitative adjustment scheme comprises a flow pattern optimization strategy, a bandwidth utilization improvement measure and a resource allocation efficiency optimization scheme, the geometric optimization result comprises a space resource allocation map, an optimized network topology structure and a flow allocation geometric model, the adaptive bandwidth allocation scheme comprises a real-time flow response strategy, a predictive resource allocation plan and a network load balance scheme, and the system comprehensive state overview comprises a performance index comprehensive analysis, a bandwidth utilization comprehensive result and a response time optimization overview.
The invention is improved in that the prediction optimization decision module comprises a load prediction sub-module, a decision making sub-module and a resource allocation sub-module;
the load prediction sub-module is used for predicting and analyzing future network load by adopting an autoregressive moving average model and analyzing autocorrelation and partial autocorrelation of time sequence data based on historical and real-time network load data, and generating a network load prediction analysis result by decomposing the time sequence data into trend, seasonal and residual components and understanding the periodic change of the data by using a seasonal trend decomposition technology;
the decision making submodule optimizes bandwidth resource allocation by establishing an objective function and constraint conditions based on a network load prediction analysis result, solves the problems of resource conflict and allocation efficiency, adjusts resource allocation by using a resource optimization algorithm and generates a formulated bandwidth strategy scheme;
the resource allocation sub-module adopts a dynamic resource allocation technology based on a formulated bandwidth policy scheme, adjusts the resource allocation policy by monitoring network flow and bandwidth requirements in real time, and matches real-time changes of network loads to generate an optimized bandwidth allocation plan.
The invention is improved in that the hierarchical dynamic adjustment module comprises a burst response sub-module, a trend matching sub-module and a strategy implementation sub-module;
the burst response submodule responds to the flow peak value and reacts to the burst flow change by adopting a real-time data analysis algorithm and monitoring and analyzing the real-time comparison of the network flow and a preset threshold value to generate a burst flow response strategy;
the trend matching sub-module calculates smooth flow trend by weighting analysis of historical flow data based on a sudden flow response strategy by adopting a statistical time sequence analysis method, and comprises the steps of weighted average calculation of the historical flow data and generation of trend lines, predicting and matching continuous flow change, and generating a flow trend matching scheme;
the strategy implementation submodule adopts a multi-layer network management strategy based on a flow trend matching scheme, and dynamically adjusts the bandwidth by comprehensively analyzing data of multiple time scales and flow changes, wherein the dynamic adjustment comprises continuous monitoring of the state of each layer of network and bandwidth adjustment based on monitoring data, and generates a hierarchical adjustment strategy.
The invention is improved in that the network simulation and evaluation module comprises a flow simulation sub-module, a network performance evaluation sub-module and a bottleneck identification sub-module;
The flow simulation submodule adopts a queuing theory model based on the current network configuration and historical flow data, simulates arrival and processing processes of various flows in a network by establishing a multi-type queuing model, and predicts future bandwidth requirements by analyzing trend and fluctuation of past flow data in combination with a historical data analysis technology to generate a network flow simulation result;
the network performance evaluation submodule adopts a network performance analysis technology based on the network flow simulation result, calculates and analyzes performance indexes of the simulation data, including delay, bandwidth utilization rate and packet loss rate, analyzes network performance in multiple aspects, and generates a network performance evaluation result;
the bottleneck recognition sub-module is used for recognizing and analyzing abnormal efficiency areas in simulation and evaluation data by adopting a bottleneck analysis technology based on network performance evaluation results, revealing bottleneck problems in a network, including insufficient bandwidth and equipment overload, and generating network behavior simulation evaluation results.
The intelligent logic control module comprises a real-time monitoring sub-module, a decision logic sub-module and an adjustment execution sub-module;
the real-time monitoring sub-module is based on a network behavior simulation evaluation result, adopts a flow analysis and monitoring algorithm, and monitors the current state of the network by analyzing a plurality of indexes of the network flow, including speed, density and data packet type, including capturing network data in real time and analyzing the characteristics and variation trend of the data flow to generate a network flow analysis result;
The decision logic submodule adopts a fuzzy logic algorithm based on the network flow analysis result, and performs bandwidth allocation decision under the network environment by processing fuzzy input data, and comprises the steps of establishing a fuzzy rule set, performing fuzzy processing on input variables, performing fuzzy reasoning, making a bandwidth allocation strategy and generating a bandwidth allocation decision scheme;
the adjustment execution submodule adopts a self-adaptive control technology based on a bandwidth allocation decision scheme, adjusts network configuration to match real-time network conditions by dynamically analyzing network load and bandwidth demands, and comprises the steps of continuously monitoring network performance indexes, adjusting control parameters according to performance feedback, implementing dynamic bandwidth allocation and generating a logic control strategy.
The invention is improved in that the quantitative analysis and adjustment module comprises a flow analysis sub-module, a logic reasoning sub-module and a scheme optimization sub-module;
the flow analysis submodule adopts a decision tree analysis method based on a logic control strategy, analyzes historical and real-time data by constructing a decision tree model, and identifies multiple types of network flow modes, and comprises the steps of analyzing data characteristics, classifying the flow data according to the importance of the multiple characteristics, distinguishing and identifying multiple modes of network flow, and generating a flow mode identification result;
The logic reasoning sub-module is based on the flow pattern recognition result, adopts a probability network analysis technology, constructs a Bayesian network model by applying Bayesian network and probability inference, performs probability calculation on the flow pattern, evaluates the influence on bandwidth allocation by using probability data, and generates a probability influence analysis result of the flow pattern;
the scheme optimizing submodule analyzes the instant bandwidth use condition and the predicted flow demand by adopting a linear programming and heuristic search method based on the probability influence analysis result of the flow mode, analyzes and adjusts the bandwidth allocation scheme, and matches various flow modes and network demands to generate a quantized adjustment scheme.
The invention is improved in that the geometric space optimization module comprises a space mapping sub-module, an optimization calculation sub-module and an allocation implementation sub-module;
the space mapping submodule adopts a multidimensional space mapping method based on a quantization adjustment scheme, and performs geometric analysis operation by expressing the bandwidth allocation problem in a multidimensional space, wherein the geometric analysis operation comprises the steps of converting parameters and constraints of bandwidth allocation into points and lines in a geometric space by applying a space coordinate transformation technology, and generating a geometric space mapping model;
The optimization calculation submodule is based on a geometric space mapping model, adopts a convex optimization technology, performs bandwidth allocation solution optimizing operation in a geometric space through a linear programming and nonlinear optimization method, and comprises the steps of identifying convex sets and convex functions, applying a mathematical optimization algorithm to solve, and generating a bandwidth optimization scheme in the geometric space;
the allocation implementation submodule adopts a parameter mapping and configuration application method based on a bandwidth optimization scheme in a geometric space, converts a convex optimization result into a network configuration parameter through a network parameter conversion technology, adjusts a bandwidth setting matching allocation strategy of network equipment and generates a geometric optimization result.
The invention is improved in that the dynamic bandwidth allocation module comprises a real-time monitoring sub-module, a bandwidth adjusting sub-module and an execution feedback sub-module;
the real-time monitoring submodule adopts a network flow real-time monitoring technology based on a hierarchical adjustment strategy and a geometric optimization result, analyzes real-time network data through a data packet analysis and flow rate measurement algorithm, and comprises the steps of capturing a network data packet in real time, analyzing the size, the speed and the flow direction of the network data packet, quantitatively evaluating the overall trend of the network flow and generating a network flow real-time analysis result;
The bandwidth adjustment submodule adopts a dynamic bandwidth adjustment mechanism based on a network flow real-time analysis result, and adjusts bandwidth configuration according to real-time monitoring data and flow prediction through a bandwidth demand prediction and resource allocation algorithm to generate a bandwidth adjustment scheme;
the execution feedback submodule adopts a network performance feedback technology based on a bandwidth adjustment scheme, evaluates the network condition after bandwidth adjustment through an effect monitoring and data analysis tool, and comprises the steps of collecting data of bandwidth use, flow distribution and delay, analyzing whether the adjusted effect accords with a preset performance target, and generating a self-adaptive bandwidth allocation scheme.
The invention improves that the system integration and monitoring module comprises a function integration sub-module, a state monitoring sub-module and a system performance monitoring sub-module;
the function integration submodule integrates functions and data by adopting a data fusion and strategy coordination technology based on an optimized bandwidth allocation plan, a hierarchy adjustment strategy, a network behavior simulation evaluation result, a logic control strategy, a quantitative adjustment scheme, a geometric optimization result and a self-adaptive bandwidth allocation scheme, and performs summarization analysis to generate a system function integration result;
The state monitoring submodule monitors the running state of the system, including the bandwidth utilization rate and the flow distribution, by adopting a real-time state monitoring technology and a dynamic data tracking and analyzing method based on the system function integration result to generate a system real-time state monitoring result;
the system performance monitoring submodule is used for generating a system comprehensive state overview by adopting a comprehensive performance analysis technology and adopting a performance index evaluation and optimization strategy to continuously track and evaluate the performance index of the system, including response time and bandwidth efficiency.
Compared with the prior art, the invention has the advantages and positive effects that:
in the invention, by applying the layered time window bandwidth allocation strategy, the system can efficiently adjust bandwidth allocation on different time scales (millisecond level, second level and minute level), thereby greatly improving the response speed and accuracy to the sudden flow fluctuation. This approach optimizes network performance and resource utilization, especially when dealing with fine-grained traffic dynamics in high-speed network environments. The application of the time sequence analysis optimization bandwidth pre-allocation strategy enables the system to accurately predict the network usage mode and adjust bandwidth allocation in advance to adapt to predicted traffic changes. This not only improves the pre-allocation efficiency of bandwidth resources, but also optimizes the bandwidth configuration of critical services, thereby guaranteeing network performance in peak hours. Discrete event analog driven bandwidth optimization further improves the ability to predict bandwidth demands and identify potential bottlenecks. The system can pre-adjust the bandwidth allocation before the actual load arrives, thereby optimizing overall network performance. The design of the double-layer fuzzy logic controller also significantly improves the evaluation capability of the system to the network state and the adaptability of bandwidth allocation. The method effectively handles the uncertainty and the ambiguity of the network environment and provides more flexible and accurate bandwidth management. The application of bandwidth allocation algorithms optimized by quantitative logical reasoning and geometric constraint brings new perspectives and solutions to the bandwidth allocation problem. The combination of the technologies ensures that the system shows extremely high accuracy and adaptability in the aspect of bandwidth resource management, effectively improves the utilization rate and transmission efficiency of network resources, reduces the phenomenon of network congestion, and simultaneously ensures the priority and quality of key data transmission. Through the innovative strategies, the system can be more flexibly adapted to the continuously changing network demands, and the user experience is obviously optimized.
Drawings
Fig. 1 is a block diagram of a dynamic bandwidth allocation system for a four-port gigabit network card according to the present invention;
fig. 2 is a system frame diagram of a dynamic bandwidth allocation system for four gigabit network cards according to the present invention;
fig. 3 is a schematic diagram of a prediction optimization decision module in a four-port gigabit network card dynamic bandwidth allocation system according to the present invention;
fig. 4 is a schematic diagram of a level dynamic adjustment module in a dynamic bandwidth allocation system of a four-port gigabit network card according to the present invention;
FIG. 5 is a schematic diagram of a network simulation and evaluation module in a four-port gigabit network card dynamic bandwidth allocation system in accordance with the present invention;
fig. 6 is a schematic diagram of an intelligent logic control module in a four-port gigabit network card dynamic bandwidth allocation system according to the present invention;
fig. 7 is a schematic diagram of a quantization analysis and adjustment module in a four-port gigabit network card dynamic bandwidth allocation system according to the present invention;
fig. 8 is a schematic diagram of a geometric space optimization module in a dynamic bandwidth allocation system of a four-port gigabit network card according to the present invention;
fig. 9 is a schematic diagram of a dynamic bandwidth allocation module in a dynamic bandwidth allocation system of a four-port gigabit network card according to the present invention;
fig. 10 is a schematic diagram of a system integration and monitoring module in a four-port gigabit network card dynamic bandwidth allocation system according to the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
In the description of the present invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," etc. indicate or are based on the orientation or positional relationship shown in the drawings, merely to facilitate description of the invention and simplify description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the invention. Furthermore, in the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
Examples
Referring to fig. 1, the present invention provides a technical solution: the system comprises a prediction optimization decision module, a hierarchical dynamic adjustment module, a network simulation and evaluation module, an intelligent logic control module, a quantitative analysis and adjustment module, a geometric space optimization module, a dynamic bandwidth allocation module and a system integration and monitoring module;
The prediction optimization decision module predicts and analyzes future network loads by adopting an autoregressive moving average model and a seasonal decomposition trend, a seasonal and residual model based on historical and real-time network load data, and performs preliminary optimization allocation on bandwidth resources by using a linear programming method to generate an optimized bandwidth allocation plan;
the hierarchical dynamic adjustment module is used for coping with sudden flow changes by adopting a dynamic threshold adjustment algorithm based on an optimized bandwidth allocation plan, smoothing continuous flow trends by adopting a weighted moving average method, and performing bandwidth adjustment on multiple time scales to generate a hierarchical adjustment strategy;
the network simulation and evaluation module simulates a network flow scene by adopting a queue theoretical model and a network flow prediction algorithm based on current network configuration and historical flow data, and comprises the steps of modeling the network flow, predicting future bandwidth demands, identifying potential network bottlenecks and performance bottlenecks, and generating a network behavior simulation evaluation result;
the intelligent logic control module carries out intelligent evaluation on the network condition by adopting a fuzzy logic controller based on the network behavior simulation evaluation result, adjusts bandwidth allocation by utilizing a self-adaptive control strategy technology, automatically adjusts control parameters according to a preset performance target and generates a logic control strategy by continuously monitoring the change of the network condition;
The quantitative analysis and adjustment module adopts decision tree classification based on a logic control strategy, utilizes decision tree to identify multiple types of flow modes by analyzing historical data and real-time data, utilizes a probability network analysis method, evaluates the influence of multiple decisions on bandwidth allocation by a probability model, and performs quantitative analysis on the flow modes and bandwidth utilization data to generate a quantitative adjustment scheme;
the geometric space optimization module is used for optimizing a bandwidth allocation scheme by adopting a convex optimization method based on a quantization adjustment scheme, mapping the bandwidth allocation problem into a geometric space by utilizing a space mapping algorithm, and re-analyzing and optimizing the bandwidth allocation by virtue of a space relation and geometric characteristics to generate a geometric optimization result;
the dynamic bandwidth allocation module analyzes real-time network data and predicts future flow trend by adopting a dynamic resource allocation algorithm based on a hierarchical adjustment strategy and a geometric optimization result, dynamically adjusts bandwidth allocation, and matches the change of network requirements to generate a self-adaptive bandwidth allocation scheme;
the system integration and monitoring module continuously tracks performance indexes of the system, including bandwidth utilization rate, flow distribution and response time, and analyzes the system performance in real time to generate a system comprehensive state overview based on an optimized bandwidth allocation plan, a hierarchical adjustment strategy, a network behavior simulation evaluation result, a logic control strategy, a quantitative adjustment scheme, a geometric optimization result and a self-adaptive bandwidth allocation scheme by adopting a comprehensive performance analysis method and a performance monitoring technology.
The optimized bandwidth allocation plan comprises network load peak prediction data, daily flow pattern analysis and current bandwidth demand prediction, the hierarchical adjustment strategy comprises a sudden flow emergency response scheme, a stable flow maintenance strategy and a conventional flow balance adjustment, the network behavior simulation evaluation result comprises network bottleneck recognition, performance bottleneck analysis and flow allocation efficiency evaluation, the logic control strategy comprises an adaptive flow response scheme, a bandwidth allocation dynamic adjustment strategy and network stability guarantee measures, the quantitative adjustment scheme comprises a flow pattern optimization strategy, a bandwidth utilization improvement measure and a resource allocation efficiency optimization scheme, the geometric optimization result comprises a space resource allocation map, an optimized network topology structure and a flow allocation geometric model, the adaptive bandwidth allocation scheme comprises a real-time flow response strategy, a predictive resource allocation plan and a network load balance scheme, and the system comprehensive state overview comprises a performance index comprehensive analysis, a bandwidth utilization comprehensive result and a response time optimization overview.
In the predictive optimization decision module, the system processes historical and real-time network load data through an autoregressive moving average model and seasonal decomposition. The autoregressive moving average model analyzes the autocorrelation and partial autocorrelation of time series data to predict future network loads. Seasonal decomposition then decomposes the time series data into trend, seasonal and residual components, helping to understand the periodic variation of the data. The bandwidth resources are then optimally allocated using a linear programming approach, which involves establishing objective functions and constraints to resolve resource conflicts and improve allocation efficiency, ultimately yielding an optimized bandwidth allocation plan.
And in the hierarchy dynamic adjustment module, a dynamic threshold adjustment algorithm and a weighted moving average method are adopted to adjust the bandwidth on multiple time scales. The dynamic threshold adjustment algorithm rapidly responds to sudden flow changes by monitoring the comparison of network flow and a set threshold in real time. The weighted moving average method predicts and adapts to the continuous flow change through the weighted analysis of the historical flow data, so as to generate a hierarchical adjustment strategy and optimize the network performance.
The network simulation and evaluation module simulates a network traffic scenario based on current network configuration and historical traffic data using a queue theoretical model and a network traffic prediction algorithm. The queue theoretical model establishes a multi-type queuing model to simulate the arrival and processing processes of different flows. The network flow prediction algorithm analyzes the historical flow data, predicts future bandwidth demands, identifies potential network bottlenecks and performance bottlenecks, and generates network behavior simulation evaluation results. The process not only simulates the actual condition of network traffic, but also provides deep insight into future network states and helps to make more accurate bandwidth allocation decisions.
The intelligent logic control module performs intelligent analysis on the network state based on the network behavior simulation evaluation result by using the fuzzy logic controller. The fuzzy logic controller processes fuzzy input data, establishes a fuzzy rule set, performs fuzzification processing on input variables, and performs fuzzy reasoning. And combining with the self-adaptive control strategy technology, the module automatically adjusts the control parameters of bandwidth allocation according to the real-time network state to generate a logic control strategy. This process increases the processing power for network environment uncertainty, making bandwidth allocation more flexible and accurate.
The quantization analysis and adjustment module carries out deep analysis on the logic control strategy through a decision tree classification and probability network analysis method. The decision tree model analyzes the historical and real-time data and identifies a plurality of network traffic patterns. The probability network analysis rule evaluates the influence of different decisions on bandwidth allocation and generates a quantization adjustment scheme. The method ensures that the bandwidth allocation strategy is more scientific and accurate, and optimizes the overall allocation efficiency of resources.
The geometric space optimization module adopts a convex optimization method and a space mapping algorithm to map the bandwidth allocation problem into the geometric space. The algorithm converts the bandwidth allocation parameters and constraints into points and lines in the geometric space by using a space coordinate transformation technology, and searches an optimal solution in the geometric space by using a mathematical optimization method to generate a geometric optimization result. This approach provides a new perspective to bandwidth allocation, so that the process of bandwidth allocation is not only based on network data, but also relies on geometric principles, thereby more efficiently optimizing resource utilization.
The dynamic bandwidth allocation module combines the hierarchical adjustment strategy and the geometric optimization result and applies a dynamic resource allocation algorithm. This module analyzes real-time network data and predicts future traffic trends to dynamically adjust bandwidth allocation to match changes in network demand. And the dynamic resource allocation algorithm adjusts bandwidth configuration according to the real-time monitoring data and the traffic prediction to generate an adaptive bandwidth allocation scheme. This procedure ensures optimal performance of the network under different conditions while maintaining efficient utilization of resources.
And the system integration and monitoring module integrates the data and strategies generated by all modules, and adopts a data fusion and strategy coordination technology to integrate functions and data. The module continuously tracks and evaluates the performance indexes of the system, including bandwidth utilization, flow distribution and response time, through a comprehensive performance analysis method and a performance monitoring technology. This process generates a comprehensive system state overview that provides the administrator with comprehensive network state information that helps them make more efficient management decisions. Through such comprehensive monitoring and analysis, the system is able to ensure that optimal performance levels are achieved under a variety of network environments.
Referring to fig. 2 and 3, the prediction optimization decision module includes a load prediction sub-module, a decision making sub-module, and a resource allocation sub-module;
the load prediction sub-module is used for predicting and analyzing future network load by adopting an autoregressive moving average model and analyzing autocorrelation and partial autocorrelation of time sequence data based on historical and real-time network load data, and generating a network load prediction analysis result by decomposing the time sequence data into trend, seasonal and residual components and understanding the periodic change of the data by using a seasonal trend decomposition technology;
The decision-making submodule optimizes bandwidth resource allocation by establishing an objective function and constraint conditions based on a network load prediction analysis result, solves the problems of resource conflict and allocation efficiency, adjusts resource allocation by using a resource optimization algorithm and generates a formulated bandwidth strategy scheme;
the resource allocation sub-module adopts a dynamic resource allocation technology based on a formulated bandwidth policy scheme, adjusts the resource allocation policy by monitoring network flow and bandwidth requirements in real time, and matches real-time changes of network loads to generate an optimized bandwidth allocation plan.
In the load prediction sub-module, the system first collects historical and real-time network load data, wherein the data comprises information such as network traffic size, data packet type and quantity at each time point. These time series data are then processed by an autoregressive moving average model. The model first analyzes the autocorrelation of the data, i.e., the effect of past data on future data, and the partial autocorrelation, i.e., the inherent correlation of the data after removal of the intervening interference effects. Next, the time series data is decomposed into three components using seasonal trend decomposition techniques: trend, seasonal, and residual components. The trend portion reveals the long-term trend of the data, the seasonal portion reveals periodic fluctuations, and the residual portion contains irregular fluctuations. This process helps understand the periodic variation of the data and generates network load predictive analysis results, such as predicting network traffic trends over a period of time in the future.
The decision making sub-module optimizes bandwidth resource allocation by adopting a linear programming method based on the analysis result of the load prediction sub-module. In this process, an objective function is first constructed, such as maximizing bandwidth utilization or minimizing delay. Constraints are then defined, including bandwidth capacity limitations, quality of service requirements, and the like. The linear programming algorithm determines the optimal resource allocation strategy by solving these objective functions and constraints. This process involves iterative adjustment of the variables to achieve an optimal solution. The resource optimization algorithm further adjusts the resource configuration to generate a formulated bandwidth policy scheme, such as allocating a particular bandwidth size for different services or applications.
The resource allocation submodule adopts a dynamic resource allocation technology to adapt to network load change in real time on the basis of a formulated bandwidth strategy scheme. By continuously monitoring network traffic and bandwidth demands, the sub-modules dynamically adjust resource allocation policies in response to real-time changes in network load. In the process, the submodule collects network traffic data such as data packet size, sending and receiving rates and the like in real time and performs comparison analysis with a preset bandwidth policy scheme. The dynamic resource allocation technique automatically adjusts the bandwidth configuration according to the difference between the real-time data and the predicted data. For example, if a surge in traffic for a certain network segment is monitored in real time, the system will automatically increase the bandwidth quota for that segment; conversely, if the traffic is below the predicted value, the system decreases the bandwidth quota for the segment. Such dynamic adjustment ensures efficient utilization of bandwidth resources, reduces resource waste, and improves overall performance of the network. Finally, the sub-module generates an optimized bandwidth allocation plan detailing the bandwidth quota of each network segment and its allocation logic, providing clear bandwidth management guidance for the network administrator.
Assume that a four-port gigabit network card dynamic bandwidth allocation system is managing an enterprise network. In the load prediction sub-module, the historical and real-time data show that 9 to 11 points in the morning are peaks of network traffic, and the average traffic is 500 Mbps. Using the ARMA model and seasonal trend decomposition technique, it is predicted that the average flow for the same time period in the next week will increase to 550 Mbps. In the decision making sub-module, the linear programming method optimizes bandwidth allocation on the premise of ensuring service quality, and improves the bandwidth quota of the main server by 20% in the peak period. The resource allocation sub-module monitors that the actual flow reaches 570 Mbps in a certain day, and immediately adjusts the bandwidth configuration of the corresponding server to ensure that the network is unobstructed. These operations generate detailed network traffic analysis reports and dynamically adjusted bandwidth allocation plans. The plan includes not only specific traffic prediction values, such as average traffic of 9 to 11 points per day, but also detailed descriptions of bandwidth allocation to various network segments, such as increasing the bandwidth of a particular server from the original 400 Mbps to 480 Mbps during peak hours. In addition, the plan also includes emergency response measures to network flow fluctuation, such as a strategy for automatically increasing bandwidth quota when the flow exceeds the expected value. The implementation of the strategies and plans effectively improves the overall performance and stability of the network, ensures the data transmission rate and the network service quality of the key time period, and simultaneously provides clear guidance and decision support for network administrators.
Referring to fig. 2 and fig. 4, the hierarchical dynamic adjustment module includes a burst response sub-module, a trend matching sub-module, and a policy implementation sub-module;
the burst response submodule responds to the flow peak value and reacts to the burst flow change by adopting a real-time data analysis algorithm and monitoring and analyzing the real-time comparison of the network flow and a preset threshold value to generate a burst flow response strategy;
the trend matching sub-module calculates smooth flow trend by weighting analysis of historical flow data based on a sudden flow response strategy by adopting a statistical time sequence analysis method, and comprises the steps of weighting average calculation of the historical flow data and generation of trend lines, predicting and matching continuous flow change, and generating a flow trend matching scheme;
the strategy implementation submodule adopts a multi-layer network management strategy based on a flow trend matching scheme, and dynamically adjusts the bandwidth by comprehensively analyzing data of multiple time scales and flow changes, wherein the dynamic adjustment comprises continuous monitoring of the state of each layer of network and bandwidth adjustment based on monitoring data, and generates a hierarchical adjustment strategy.
In the burst response sub-module, the system employs a real-time data analysis algorithm, which involves collecting and processing network traffic data in real-time, including the size, number, transmission and reception times, etc. of the data packets. These data are used to compare in real time with preset flow thresholds. When network traffic rises rapidly and exceeds a threshold, the submodule immediately initiates a bursty traffic response mechanism. The specific process comprises the steps of calculating the increasing rate of the network flow in real time and comparing the increasing rate with the historical data to judge whether the flow increase belongs to a normal fluctuation range. If an abnormal growth is determined, the system will automatically initiate an emergency bandwidth allocation policy, such as temporarily increasing the bandwidth quota of the affected area. This process ensures the stability and continuity of the network in the event of an emergency traffic event, avoiding potential service interruption or performance degradation.
The trend matching sub-module is used for analyzing the long-term trend of the network traffic based on a statistical time sequence analysis method. The sub-module first collects historical data of network traffic over a period of time and then calculates a smooth trend of the traffic by a weighted moving average method. In this process, the system gives higher weight to recent data based on the time stamp of the traffic data, thereby more accurately reflecting the current network conditions. Based on these trend data, the sub-module can predict changes in traffic over a future period of time and adjust the bandwidth allocation policy accordingly, such as increasing the bandwidth quota of the corresponding region in advance over a period of time in which traffic is predicted to increase. The method enables the network to better adapt to the long-term trend of flow change, and optimizes the use efficiency of bandwidth resources.
The strategy implementation sub-module is responsible for integrating the burst response and the output of the trend matching sub-module to form a multi-level network management strategy. This process involves collecting and analyzing data for different time scales and traffic variations, and then dynamically adjusting the bandwidth based on these data. Specifically, the sub-module evaluates the current network state based on information obtained from the burst response and trend matching sub-module and predicts traffic changes over a short period of time. This includes knowledge of bandwidth usage, traffic distribution, and potential bottlenecks at each network layer. According to the information, the submodule dynamically adjusts the bandwidth through a multi-layer network management strategy. For example, if a network layer is predicted to face a peak in traffic, the policy enforcement sub-module will increase the bandwidth quota for that layer in advance while decreasing the bandwidth quota for other more idle network layers to maintain the balance of the overall network. This process involves continuous monitoring of the state of each layer of the network and bandwidth adjustments based on the monitored data and the predictions to ensure optimal performance of the network at different levels.
It is assumed that the internal network of an enterprise is managed by a four-port gigabit network card dynamic bandwidth allocation system. On a typical workday, the burst response sub-module of the system monitors that a department suddenly begins to transmit a large amount of data, resulting in a dramatic increase in network traffic for that department. The real-time data shows that the traffic of this department increases from 200 Mbps per second to 500 Mbps in a short time, rapidly exceeding the set threshold of 400 Mbps. The system immediately initiates an emergency bandwidth allocation to temporarily boost the department's bandwidth quota to 600 Mbps to handle bursty traffic. At the same time, the trend matching submodule analyzes the traffic data of the past weeks, and discovers that the network traffic of the department in afternoon every Wednesday is increased. Thus, the submodule suggests that the department's bandwidth quota be increased in advance every three afternoons. The strategy implementation submodule synthesizes the information, correspondingly adjusts the bandwidth, and not only is the emergency of the current day dealt with, but also the preparation is made for the similar situation in the future. In this process, the system generates detailed network traffic reports and bandwidth adjustment plans, including detailed data of traffic peaks, time and scale of bandwidth adjustments, and predictions of future traffic trends. For example, it is indicated in the report that the department's bandwidth quota will be increased from the original 400 Mbps to 550 Mbps in the three afternoon per week to cope with the expected traffic growth.
Referring to fig. 2 and 5, the network simulation and evaluation module includes a flow simulation sub-module, a network performance evaluation sub-module, and a bottleneck recognition sub-module;
the flow simulation sub-module is based on the current network configuration and historical flow data, adopts a queuing theory model, simulates the arrival and processing processes of various flows in a network by establishing a multi-type queuing model, and predicts the future bandwidth requirements by analyzing the trend and fluctuation of the past flow data in combination with a historical data analysis technology to generate a network flow simulation result;
the network performance evaluation sub-module adopts a network performance analysis technology based on the network flow simulation result, calculates and analyzes performance indexes of the simulation data, including delay, bandwidth utilization rate and packet loss rate, and analyzes network performance in multiple aspects to generate a network performance evaluation result;
the bottleneck recognition sub-module adopts a bottleneck analysis technology based on the network performance evaluation result, and reveals bottleneck problems in the network, including insufficient bandwidth and equipment overload, by recognizing and analyzing abnormal efficiency areas in simulation and evaluation data, so as to generate a network behavior simulation evaluation result.
In the traffic simulation sub-module, the system processes current network configuration and historical traffic data using a queuing theory model. The data format includes the traffic size, packet type, arrival and departure times, etc. of each node of the network. The queuing theory model establishes the arrival and processing processes of various flows and simulates the behaviors of various flows in the network. In practice, the system first builds multiple queuing models, e.g., setting up different queues for different types of services or applications, based on network configuration and historical data. Then, by modeling the flow changes in these queues, the system predicts future bandwidth demands, such as predicting the flow peaks over a particular period of time. This process covers trend and fluctuation analysis of historical flow data, and statistical and probabilistic calculations of such data to predict future flow patterns. Ultimately, the submodules generate network traffic simulation results, such as predicted traffic charts and reports, which help network administrators learn about future facing traffic pressures, thereby better planning bandwidth resources.
The network performance evaluation sub-module adopts a network performance analysis technology based on the flow simulation result. This technique involves detailed performance index calculation and analysis of the simulated network data. For example, the submodule calculates key performance indicators such as delay time, bandwidth utilization, packet loss rate and the like in the analog data. These calculations are based on network theory and statistical methods, such as calculating delays by analyzing the transmission time of the data packets, or determining packet loss rates by comparing the number of data packets sent and received. The performance evaluation results include not only these calculated quantitative indicators, but also an in-depth analysis of the network state behind these indicators. Finally, the sub-module generated network performance assessment report provides a comprehensive network performance overview to the network administrator, including which aspects perform well and which aspects need improvement.
The bottleneck recognition sub-module adopts a bottleneck analysis technology based on the network performance evaluation result. The submodule first identifies areas of abnormal efficiency in the simulation and evaluation data, such as sudden delay increases or bandwidth utilization decreases. This process involves pattern recognition and trend analysis of the data to find bottleneck causes such as bandwidth starvation or device overload. The sub-module identifies areas that lead to performance degradation by analyzing each portion of the network in detail. For example, if the bandwidth utilization of a certain network node continues to exceed 90% and the utilization of surrounding nodes is low, the submodule may identify this node as a potential bottleneck. The sub-modules will then conduct an in-depth analysis of these potential bottlenecks, including assessing their impact on overall network performance, and proposing solutions. Eventually, the network behavior simulation evaluation results generated by the submodule provide detailed information about the network bottleneck location, reasons and solution strategies for the network administrator.
Assume that the network of one data center is managed by a four-port gigabit network card dynamic bandwidth allocation system. The traffic simulation submodule collects traffic data of the past month, including the peak, duration and type of traffic per day. And predicting network traffic in the future week by using a queuing theory model, and indicating that the period from 2 pm to 4 pm is a period in which traffic peaks appear. The network performance evaluation sub-module then analyzes these simulation results to find that during the predicted peak hours, the network delay increases by 10% and the bandwidth utilization reaches 80% of the maximum. The bottleneck recognition submodule further analyzes and discovers that the bandwidth utilization rate of the core switch of the data center in the peak time is continuously over 95 percent, and the core switch becomes a bottleneck of the network. Based on these analyses, the system proposes a strategy to increase the bandwidth quota of the core switch in advance during peak hours to mitigate delays and avoid packet loss. Implementation of these policies ultimately helps the data center optimize network performance, ensuring smooth operation during critical traffic periods. The generated detailed report and analysis chart not only provides deep network performance and bottleneck analysis for network administrators, but also provides valuable data support for them to formulate future network optimization strategies.
Referring to fig. 2 and 6, the intelligent logic control module includes a real-time monitoring sub-module, a decision logic sub-module, and an adjustment execution sub-module;
the real-time monitoring sub-module monitors the current state of the network by analyzing a plurality of indexes of the network flow, including speed, density and data packet type, including capturing network data in real time and analyzing the characteristics and variation trend of the data flow by adopting a flow analysis and monitoring algorithm based on the network behavior simulation evaluation result to generate a network flow analysis result;
the decision logic submodule adopts a fuzzy logic algorithm based on the network flow analysis result, and performs bandwidth allocation decision under the network environment by processing fuzzy input data, and comprises the steps of establishing a fuzzy rule set, performing fuzzy processing on input variables, performing fuzzy reasoning, making a bandwidth allocation strategy and generating a bandwidth allocation decision scheme;
the adjusting execution submodule adopts a self-adaptive control technology based on a bandwidth allocation decision scheme, adjusts network configuration to match real-time network conditions by dynamically analyzing network load and bandwidth demands, and comprises the steps of continuously monitoring network performance indexes, adjusting control parameters according to performance feedback, implementing dynamic bandwidth allocation and generating a logic control strategy.
In the real-time monitoring sub-module, the system captures and analyzes network data in real-time using a traffic analysis and monitoring algorithm. The data format handled by the sub-module includes the speed, density, packet type, etc. of the network traffic. By analyzing the data, the sub-module can monitor the current state of the network in real time, such as detecting peaks in network traffic and identifying abnormal patterns in the data stream. In the implementation process, the submodule firstly monitors the network flow continuously according to preset parameters such as the size of a data packet or the transmission time interval. Statistical analysis is then performed on the collected data, such as calculating an average flow rate or identifying a trend in flow density. This process helps the sub-module to learn about network conditions in real time and discover problems in time, such as a sudden increase in traffic. Finally, the network traffic analysis results generated by the submodules provide real-time network monitoring reports for network administrators, including traffic statistics and alarms of potential problems.
And the decision logic sub-module performs bandwidth allocation decision by adopting a fuzzy logic algorithm based on the network flow analysis result. Fuzzy logic algorithms are particularly suited to handling fuzzy or incomplete input data such as traffic fluctuations or temporary network congestion. The sub-modules first build a set of fuzzy rule sets based on network performance metrics such as traffic speed or delay time. The incoming network data is then obfuscated, such as by dividing packet speeds into "slow", "medium", and "fast". And the submodule evaluates the current network condition through fuzzy reasoning and formulates a corresponding bandwidth allocation strategy. For example, if network traffic is "fast" and delay is "low," the bandwidth quota is increased. Such a decision process makes bandwidth allocation more flexible and accurate. The bandwidth allocation decision scheme generated by the submodule provides a basis for dynamically adjusting bandwidth allocation.
The adjustment execution submodule is implemented by adopting an adaptive control technology based on a bandwidth allocation decision scheme. This sub-module adjusts the network configuration to accommodate real-time network conditions by dynamically analyzing network load and bandwidth requirements. During execution, the sub-modules continuously monitor key performance indicators such as real-time bandwidth usage, network delay, and packet loss rate. Based on these monitoring data and the bandwidth allocation scheme received from the decision logic submodule, the submodule dynamically adjusts the configuration of the network device, such as increasing or decreasing the bandwidth quota of a particular area or service. The self-adaptive control strategy enables the network to flexibly cope with the flow change and ensures the optimization of the network performance. Finally, the logical control strategy generated by the submodule provides a detailed bandwidth adjustment guideline for the network administrator, including which areas need to be increased in bandwidth, which can be decreased, and the corresponding adjustment times and magnitudes.
Assuming a network environment of a large enterprise, the network is managed by a four-port gigabit network card dynamic bandwidth allocation system. The enterprise network is subject to significant traffic surges during certain hours, such as lunch time (12:00 to 13:00) and peak hours of work hours (17:00 to 18:00). For example, real-time monitoring data shows that during lunch hours, network traffic jumps from an average of 2GB per second to 5GB per second, while peak off-peak traffic reaches even 7GB per second. The real-time monitoring sub-module captures information such as speed, density and data packet type of the flow in real time by continuously monitoring the network state. During the analysis, the sub-module uses statistical analysis methods, such as calculating the average speed of the flow or identifying the trend of the flow density. This process enables the sub-module to discover peaks and abnormal patterns of network traffic in time, such as a sudden increase in traffic during lunch hours. The decision logic submodule adopts a fuzzy logic algorithm to carry out bandwidth allocation decision based on the real-time monitoring result. In this example, the module recognizes the traffic surge during lunch time and peak hours of work as a "high traffic" state and formulates a bandwidth allocation scheme based thereon. For example, the decision logic submodule decides to automatically increase the bandwidth quota for critical traffic areas during these peak hours to cope with traffic surge situations. The adjustment execution submodule dynamically adjusts the network configuration according to the bandwidth allocation scheme of the decision logic submodule. In the implementation process, the submodule continuously monitors key performance indexes of the network and dynamically adjusts the configuration of the network equipment according to the real-time data. For example, according to the scheme, lunch time and rush hour automatically boost the bandwidth of the data center and the main office area from 1Gbps to 1.5Gbps, and reduce the bandwidth of these areas to 0.8Gbps during the evening and weekend hours.
Referring to fig. 2 and 7, the quantization analysis and adjustment module includes a flow analysis sub-module, a logic reasoning sub-module, and a scheme optimization sub-module;
the flow analysis submodule adopts a decision tree analysis method based on a logic control strategy, analyzes historical and real-time data by constructing a decision tree model, and identifies multiple types of network flow modes, including analyzing data characteristics, classifying the flow data according to the importance of the multiple characteristics, distinguishing and identifying multiple modes of network flow, and generating a flow mode identification result;
the logic reasoning sub-module is based on the flow pattern recognition result, adopts a probability network analysis technology, builds a Bayesian network model by applying Bayesian network and probability inference, performs probability calculation on the flow pattern, evaluates the influence on bandwidth allocation by using probability data, and generates a probability influence analysis result of the flow pattern;
the scheme optimization submodule analyzes the instant bandwidth use condition and the predicted flow demand by adopting a linear programming and heuristic search method based on the probability influence analysis result of the flow mode, analyzes and adjusts the bandwidth allocation scheme, and matches various flow modes and network demands to generate a quantized adjustment scheme.
In the flow analysis sub-module, the system processes historical and real-time data by utilizing a decision tree analysis method based on a logic control strategy. The data format includes information such as traffic size, duration, source and destination. The decision tree model identifies different types of network traffic patterns by analyzing these data features. In practice, a decision tree model is first created based on historical data, which includes a plurality of decision points, each of which represents a characteristic of the traffic data, such as traffic size or packet type. Then, the system traverses the decision tree according to the real-time data, and classifies and identifies the current flow mode. For example, if the size of a certain data stream exceeds a certain threshold and lasts longer, the system classifies it as a large file transfer. By the method, the sub-module can accurately identify and classify network traffic and generate traffic pattern identification results such as a traffic type distribution map. These results are critical to understanding the behavior patterns of network traffic and to developing corresponding bandwidth adjustment policies.
The logic reasoning sub-module adopts a probability network analysis technology, in particular a Bayesian network and probability inference, based on the traffic pattern recognition result. A bayesian network is a graphical model for expressing the probability relationships between variables. In this sub-module, the system first builds a bayesian network model in which nodes represent different traffic patterns and edges represent the probabilistic relationships between these patterns. Then, by applying probabilistic inference, the system calculates the probability of occurrence of various traffic patterns and evaluates the impact of those patterns on bandwidth allocation. For example, the system calculates the probability of impact of the video traffic increase on the total bandwidth demand. The probability calculation results help network administrators understand the potential influence of different flow modes on network performance, and scientific basis is provided for bandwidth allocation decisions.
And the scheme optimizing sub-module optimizes the bandwidth allocation scheme by adopting a linear programming and heuristic search method based on the probability influence analysis result of the flow mode. Linear programming is used to determine an optimal bandwidth allocation scheme by establishing an objective function (e.g., maximizing bandwidth utilization) and constraint (e.g., not exceeding a total bandwidth limit). Heuristic searches are then used to find the best solution among a variety of bandwidth allocation schemes. By the method, the submodule can analyze the instant bandwidth use condition and the predicted flow demand, and adjust the bandwidth allocation scheme according to the instant bandwidth use condition and the predicted flow demand so as to match various flow modes and network demands. Finally, the quantization adjustment scheme generated by the sub-module provides a detailed bandwidth allocation strategy, including specific bandwidth allocation amounts and adjustment times. These policies enable network administrators to flexibly adjust bandwidth based on actual and predicted network conditions, thereby optimizing network performance and improving resource utilization.
Assume a company providing online video services, the network of which is managed by a four-port gigabit network card dynamic bandwidth allocation system. The corporate network environment, particularly during peak evening hours, is often challenged with bandwidth starvation. The flow analysis sub-module first performs in-depth analysis on historical and real-time data of the company network. Assuming historical data shows that video traffic for the evening hours (e.g., 20:00 to 23:00) averages 8GB per second, accounting for 75% of the total bandwidth. The submodule identifies a mode of rapid increase of video flow at night by using a decision tree analysis method. By analyzing data for different time periods, such as traffic size, duration, data source and destination, etc., the module successfully delineates traffic features and patterns over a particular period. The logic reasoning sub-module uses Bayesian network technology to conduct deep analysis on the traffic patterns. By building a bayesian network model, the module evaluates the effect of evening video traffic increases on overall bandwidth demand. For example, the model predicts that during a similar evening rush hour in the future, the proportion of video traffic to bandwidth rises above 80%. The scheme optimization submodule utilizes a linear programming and heuristic search method to provide an effective bandwidth allocation strategy. Assuming the linear programming model suggests that the bandwidth of a particular server is increased by 20% during the peak evening hours to cope with the increase in video traffic. Heuristic searches then find the best solution among multiple bandwidth allocation schemes, i.e., increasing the bandwidth of a particular server from 1Gbps to 1.2Gbps during the evening hours between 20:00 and 23:00.
Referring to fig. 2 and 8, the geometric space optimization module includes a space mapping sub-module, an optimization calculation sub-module, and an allocation implementation sub-module;
the space mapping submodule adopts a multidimensional space mapping method based on a quantization adjustment scheme, performs geometric analysis operation by expressing the bandwidth allocation problem in a multidimensional space, and comprises the steps of converting parameters and constraints of bandwidth allocation into points and lines in the geometric space by applying a space coordinate transformation technology to generate a geometric space mapping model;
the optimization calculation submodule carries out the optimizing operation of the bandwidth allocation solution in the geometric space by adopting a convex optimization technology and a linear programming and nonlinear optimization method based on the geometric space mapping model, and comprises the steps of identifying convex sets and convex functions, applying a mathematical optimization algorithm to solve, and generating a bandwidth optimization scheme in the geometric space;
the distribution implementation submodule adopts a parameter mapping and configuration application method based on a bandwidth optimization scheme in a geometric space, converts a convex optimization result into a network configuration parameter through a network parameter conversion technology, adjusts a bandwidth setting matching distribution strategy of network equipment, and generates a geometric optimization result.
In the spatial mapping sub-module, the bandwidth allocation problem is converted into an expression in geometric space by a multidimensional spatial mapping method. The data format includes information such as bandwidth requirements, network loading characteristics, and available resources, which are mapped to points and lines in geometric space. In the execution process, the submodule firstly uses a space coordinate transformation technology to convert parameters (such as bandwidth size and demand duration) and constraints (such as resource limitation and priority rule) of bandwidth allocation into elements in a geometric space. For example, a bandwidth requirement may be represented as a point in geometric space, while a resource constraint may be represented as a line or curved surface. By this mapping, the bandwidth allocation problem is visualized as a geometric space model, which provides an intuitive and efficient way of expressing the optimization calculation. Finally, the generated geometric space mapping model provides a geometric view of the bandwidth allocation problem, so that the optimization process is more visual and easy to understand.
The optimization calculation sub-module adopts a convex optimization technology based on the geometric space mapping model. In this sub-module, the system first identifies a convex set and convex function to define a feasible solution space for the bandwidth allocation problem. Then, the optimal solution is searched in the geometric space through a linear programming and nonlinear optimization method. For example, the sub-module may find a point (representing a bandwidth allocation scheme) that is both within the defined convex set and maximizes a certain performance index (e.g., overall bandwidth utilization). This convex optimization method allows the system to find the optimal bandwidth allocation scheme while ensuring constraints. Finally, the bandwidth optimization scheme generated by the submodule provides a mathematical solution for bandwidth allocation on the basis of guaranteeing network performance and resource utilization efficiency.
The allocation implementation sub-module is responsible for converting the bandwidth optimization scheme in the geometry space into the actual network configuration parameters. In this sub-module, the convex optimization results (e.g., specific points or paths in geometric space) are first converted to specific network configuration parameters, such as bandwidth settings for a specific device or link, using a parameter mapping technique. This process involves parsing the geometry optimization results to extract critical configuration information such as bandwidth allocation and time duration. The submodule then applies these parameters to the network devices by configuring the application method, adjusting the bandwidth settings of the respective devices to accommodate the allocation policy. For example, if the geometry optimization results indicate that the bandwidth of a certain server link should be increased to meet high traffic demands, the sub-module will adjust the corresponding router or switch configuration to ensure that the bandwidth setting of the link reflects this change. The real-time adjustment ensures that the network configuration can quickly respond to the decision of the optimization module, thereby improving the overall performance and efficiency of the network. In this way, the allocation enforcement submodule ensures that the optimization results in the geometric space can be effectively converted into actual network operations, thereby realizing bridging between theoretical optimization and actual application.
Assume a data center network of a nationwide company, which is managed by a four-gigabit network card dynamic bandwidth allocation system. During peak traffic periods, network traffic demands proliferate. Consider the specific case of a data center network during peak traffic hours. Assuming that the network traffic demand data shows an average 40% increase in bandwidth demand for cross-country videoconferencing during late peak periods, this is embodied as an increase from 200 Mbps to 280 Mbps. At this time, the data received by the spatial mapping sub-module includes: current bandwidth usage is 200 Mbps, demand for 80 Mbps increases, and the total bandwidth capacity of the network is 500 Mbps. The space mapping sub-module maps the data into geometric space, e.g., representing the current bandwidth usage as a point within a three-dimensional space bounded by bandwidth capacity. The increased bandwidth requirement is then represented as a vector from the current point to the target point (280 Mbps). The optimization computation sub-module analyzes these points and vectors in geometric space, using convex optimization techniques to determine whether bandwidth can be increased without violating resource constraints. For example, the module may calculate whether a path from 200 Mbps to 280 Mbps would cross a defined bandwidth capacity limit boundary (500 Mbps). If the path is reasonable, the system determines that this bandwidth increase is feasible. The allocation enforcement sub-module translates the convex optimization results into operations of the actual network configuration, such as instructing the network device to increase the bandwidth setting of a particular link from 200 Mbps to 280 Mbps. This adjustment is immediately applied to the data center network, ensuring that the national video conference gets the necessary network resources during peak periods, maintaining a smooth conference experience.
Referring to fig. 2 and 9, the dynamic bandwidth allocation module includes a real-time monitoring sub-module, a bandwidth adjustment sub-module, and an execution feedback sub-module;
the real-time monitoring submodule analyzes real-time network data by adopting a network flow real-time monitoring technology and through a data packet analysis and flow rate measurement algorithm based on a hierarchical adjustment strategy and a geometric optimization result, and comprises the steps of capturing a network data packet in real time, analyzing the size, the speed and the flow direction of the network data packet, quantitatively evaluating the overall trend of the network flow and generating a network flow real-time analysis result;
the bandwidth adjustment submodule adopts a dynamic bandwidth adjustment mechanism based on a network flow real-time analysis result, and adjusts bandwidth configuration according to real-time monitoring data and flow prediction by a bandwidth demand prediction and resource allocation algorithm to generate a bandwidth adjustment scheme;
the feedback sub-module is executed, based on a bandwidth adjustment scheme, a network performance feedback technology is adopted, the network condition after bandwidth adjustment is evaluated through an effect monitoring and data analysis tool, the method comprises the steps of collecting data of bandwidth use, flow distribution and delay, analyzing whether the effect of adjustment accords with a preset performance target, and generating a self-adaptive bandwidth allocation scheme.
In the real-time monitoring sub-module, the system processes real-time network data by using a data packet analysis and flow rate measurement algorithm through a network flow real-time monitoring technology. Specifically, the submodule captures detailed information of each data packet including its size, transmission speed and transmission direction. Therefore, the real-time monitoring sub-module captures the network data packet in real time by adopting a flow analysis algorithm, such as a flow analysis tool, such as a Wireshark or a similar network monitoring tool, and comprehensively analyzes the characteristics and the change trend of the data stream. The results of the analysis include overall traffic trends of the network, such as peaks, valleys, and average traffic over a period of time. These results are critical to understanding the current state of the network and predicting future traffic trends.
In the bandwidth adjustment sub-module, based on the real-time monitoring data and the hierarchical adjustment strategy, a dynamic bandwidth adjustment mechanism is employed, including bandwidth demand prediction and resource allocation algorithms, such as linear or nonlinear prediction models, to estimate future bandwidth demands and adjust the bandwidth configuration accordingly. This process involves analyzing real-time monitoring data, such as traffic peaks and trends, and predicting future bandwidth demands based on these data. In this way, the sub-modules can generate bandwidth adjustment schemes for future demands and adjust bandwidth configurations in the network in real time to optimize network performance.
The execution feedback submodule adopts a network performance feedback technology to evaluate the effect of bandwidth adjustment. This involves using performance monitoring tools, such as network performance monitoring systems, to collect bandwidth usage, traffic distribution, and network delays. Through analysis of the data, the sub-module can evaluate the effect of bandwidth adjustment, such as whether network congestion is successfully relieved, data transmission efficiency is improved, and the like. These evaluation results help to further optimize the bandwidth allocation strategy, ensuring that network performance continues to meet the preset goals.
Assume an enterprise network environment in which a four-port gigabit network card dynamic bandwidth allocation system is deployed. The network environment includes a plurality of departments, each department having different network traffic characteristics and bandwidth requirements. To simulate this environment, the following data items and simulated values were set: the real-time monitoring sub-module monitors a dramatic increase in network traffic, particularly traffic to financial and human resources departments, during peak hours of the workday (e.g., 9 a.m. to 11 a.m.). The size of the monitored data packet ranges from 100 bytes to 1500 bytes, and the transmission speed is about 800Mbps to 1Gbps. The bandwidth adjustment sub-module predicts that the bandwidth requirements of these two departments will increase by 20% in the next hour based on the real-time monitoring data. It adjusts the bandwidth configuration, thus providing additional bandwidth resources for both departments, while slightly reducing bandwidth allocation from other low load departments. After the feedback sub-module adjustment is performed, the system collects and analyzes bandwidth usage data and network performance metrics. The result shows that after the bandwidth is adjusted, the network delay of financial and human resource departments is reduced by 15%, the packet loss rate is reduced by 5%, and the adjustment is proved to effectively improve the network performance.
Referring to fig. 2 and 10, the system integration and monitoring module includes a function integration sub-module, a status monitoring sub-module, and a system performance monitoring sub-module;
the function integration submodule integrates functions and data by adopting a data fusion and strategy coordination technology based on an optimized bandwidth allocation plan, a hierarchical adjustment strategy, a network behavior simulation evaluation result, a logic control strategy, a quantitative adjustment scheme, a geometric optimization result and a self-adaptive bandwidth allocation scheme, and performs summarization analysis to generate a system function integration result;
the state monitoring submodule monitors the running state of the system, including the bandwidth utilization rate and the flow distribution, by adopting a real-time state monitoring technology and a dynamic data tracking and analyzing method based on the system function integration result to generate a system real-time state monitoring result;
the system performance monitoring submodule is used for generating a system comprehensive state overview by continuously tracking and evaluating the performance indexes of the system, including response time and bandwidth efficiency, by adopting a comprehensive performance analysis technology and a performance index evaluation and optimization strategy based on the system real-time state monitoring result.
In the function integration submodule, the data exists in the form of a multidimensional data set, and various factors such as an optimized bandwidth allocation plan, a hierarchical adjustment strategy and the like are covered. Firstly, a data fusion technology is applied, and the technology integrates information of different data sources through weighted average and correlation analysis, so that the comprehensiveness and accuracy of data are improved. And then, adopting a strategy coordination technology, wherein the technology is based on a decision tree and a neural network algorithm, and comprehensively evaluates and optimizes various strategies to ensure effective coordination among the strategies. In the integration process, the quantitative adjustment scheme is processed by a linear programming method to achieve the optimization of resource allocation. The geometric optimization result depends on geometric algorithms such as convex hulls and minimum bounding rectangle algorithms to ensure the space efficiency of data layout. The self-adaptive bandwidth allocation scheme utilizes a dynamic programming algorithm to dynamically adjust the bandwidth allocation strategy according to the real-time flow and the historical data. The integrated result is output in a report form, the function integrated condition of the system is recorded in detail, and a foundation is provided for subsequent monitoring and optimization.
The data processed by the state monitoring sub-module is presented in a time sequence format and mainly reflects the running state of the system. The sub-module uses a real-time state monitoring technique, wherein dynamic data tracking adopts a sliding window algorithm to continuously update the latest state of the system, and an analysis method is based on time sequence analysis, such as an autoregressive moving average model, so as to identify and predict the change trend of the state of the system. The monitoring of bandwidth utilization and traffic distribution is performed by capturing network traffic data and performing a frequency domain analysis on the data using a traffic analysis algorithm, such as fourier transform, to accurately assess the usage of the network. The generated real-time status monitoring results are presented in the form of charts and index reports, providing a real-time view of system performance.
The data utilized by the system performance monitoring submodule is also based on time series, focusing on performance index evaluation. The module adopts a comprehensive performance analysis technology, wherein performance index evaluation adopts a benchmark test method to quantitatively evaluate the response time and bandwidth efficiency of the system. The optimization strategy section then combines the historical data with the current performance data, and applies regression analysis and trend prediction algorithms, such as moving average or exponential smoothing algorithms, to identify performance bottlenecks and improvement points. In addition, the sub-module also comprises a set of feedback mechanism, and the decision support system is used for automatically adjusting relevant parameters according to the performance evaluation result so as to optimize the system performance. The generated comprehensive system state overview appears in the form of comprehensive reports, provides comprehensive evaluation of the overall system performance, and guides the future optimization direction.
Assume an enterprise network environment in which a four-port gigabit network card dynamic bandwidth allocation system is deployed. During peak hours, the network traffic reaches 10GB per second, the bandwidth occupancy rate is 80%, and the service request times are 1000 times per second. At this time, the function integration sub-module of the system integration and monitoring module will first intervene. In the function integration sub-module, the policy coordination technique will analyze the network traffic data. And analyzing according to the flow distribution of different time periods by using a decision tree algorithm, and predicting the future flow trend by using a neural network algorithm based on the current data. Based on these analysis results, the module will automatically adjust the bandwidth allocation policy to optimize network resource usage. The state monitoring sub-module intervenes through a real-time state monitoring technology. The module captures the change of the network state in real time by utilizing a sliding window algorithm, and analyzes the flow fluctuation by an autoregressive moving average model. These analyses help monitor and demonstrate real-time bandwidth utilization, ensuring that network traffic is effectively managed during peak hours. The system performance monitoring submodule adopts a comprehensive performance analysis technology. By benchmark test methods, the module quantitatively evaluates the response time and bandwidth efficiency of the network. Meanwhile, performance data are analyzed by using regression analysis and trend prediction algorithms, and performance bottlenecks are identified. Based on these analyses, the decision support system will automatically adjust the network parameters to further optimize performance.
The present invention is not limited to the above embodiments, and any equivalent embodiments which can be changed or modified by the technical disclosure described above can be applied to other fields, but any simple modification, equivalent changes and modification made to the above embodiments according to the technical matter of the present invention will still fall within the scope of the technical disclosure.
Claims (10)
1. A four-port kilomega network card dynamic bandwidth allocation system is characterized in that: the system comprises a prediction optimization decision module, a hierarchical dynamic adjustment module, a network simulation and evaluation module, an intelligent logic control module, a quantitative analysis and adjustment module, a geometric space optimization module, a dynamic bandwidth allocation module and a system integration and monitoring module;
the prediction optimization decision module predicts and analyzes future network loads by adopting an autoregressive moving average model and a seasonal decomposition trend, a seasonal and residual model based on historical and real-time network load data, and performs preliminary optimization allocation on bandwidth resources by using a linear programming method to generate an optimized bandwidth allocation plan;
The hierarchical dynamic adjustment module is used for coping with sudden flow changes by adopting a dynamic threshold adjustment algorithm based on an optimized bandwidth allocation plan, smoothing continuous flow trends by adopting a weighted moving average method, and performing bandwidth adjustment on multiple time scales to generate a hierarchical adjustment strategy;
the network simulation and evaluation module simulates a network flow scene by adopting a queue theoretical model and a network flow prediction algorithm based on current network configuration and historical flow data, and comprises the steps of modeling the network flow, predicting future bandwidth demands, identifying potential network bottlenecks and performance bottlenecks, and generating a network behavior simulation evaluation result;
the intelligent logic control module carries out intelligent evaluation on the network condition by adopting a fuzzy logic controller based on the network behavior simulation evaluation result, adjusts bandwidth allocation by utilizing a self-adaptive control strategy technology, automatically adjusts control parameters according to a preset performance target by continuously monitoring the change of the network condition, and generates a logic control strategy;
the quantitative analysis and adjustment module adopts decision tree classification based on a logic control strategy, utilizes decision trees to identify multiple types of flow modes by analyzing historical data and real-time data, utilizes a probability network analysis method, evaluates the influence of multiple decisions on bandwidth allocation by using a probability model, and performs quantitative analysis on the flow modes and bandwidth utilization data to generate a quantitative adjustment scheme;
The geometric space optimization module performs optimization of the bandwidth allocation scheme by adopting a convex optimization method based on a quantization adjustment scheme, maps the bandwidth allocation problem into a geometric space by utilizing a space mapping algorithm, and re-analyzes and optimizes the bandwidth allocation by virtue of a space relation and geometric characteristics to generate a geometric optimization result;
the dynamic bandwidth allocation module adopts a dynamic resource allocation algorithm to analyze real-time network data and forecast future flow trend based on a hierarchical adjustment strategy and a geometric optimization result, dynamically adjusts bandwidth allocation, and matches the change of network requirements to generate a self-adaptive bandwidth allocation scheme;
the system integration and monitoring module continuously tracks performance indexes of the system, including bandwidth utilization rate, flow distribution and response time, and analyzes the system performance in real time to generate a system comprehensive state overview based on an optimized bandwidth allocation plan, a hierarchical adjustment strategy, a network behavior simulation evaluation result, a logic control strategy, a quantitative adjustment scheme, a geometric optimization result and a self-adaptive bandwidth allocation scheme by adopting a comprehensive performance analysis method and a performance monitoring technology.
2. The four-port gigabit network card dynamic bandwidth allocation system of claim 1, wherein: the optimized bandwidth allocation plan comprises network load peak prediction data, daily flow pattern analysis and current bandwidth demand prediction, the hierarchical adjustment strategy comprises a sudden flow emergency response scheme, a stable flow maintenance strategy and a conventional flow balance adjustment, the network behavior simulation evaluation result comprises network bottleneck identification, performance bottleneck analysis and flow allocation efficiency evaluation, the logic control strategy comprises an adaptive flow response scheme, a bandwidth allocation dynamic adjustment strategy and network stability guarantee measures, the quantitative adjustment scheme comprises a flow pattern optimization strategy, a bandwidth utilization improvement measure and a resource allocation efficiency optimization scheme, the geometric optimization result comprises a space resource allocation map, an optimized network topology structure and a flow allocation geometric model, the adaptive bandwidth allocation scheme comprises a real-time flow response strategy, a predictive resource allocation plan and a network load balance scheme, and the system comprehensive state overview comprises a performance index comprehensive analysis, a bandwidth utilization comprehensive result and a response time optimization summary.
3. The four-port gigabit network card dynamic bandwidth allocation system of claim 1, wherein: the prediction optimization decision module comprises a load prediction sub-module, a decision making sub-module and a resource allocation sub-module;
the load prediction sub-module is used for predicting and analyzing future network load by adopting an autoregressive moving average model and analyzing autocorrelation and partial autocorrelation of time sequence data based on historical and real-time network load data, and generating a network load prediction analysis result by decomposing the time sequence data into trend, seasonal and residual components and understanding the periodic change of the data by using a seasonal trend decomposition technology;
the decision making submodule optimizes bandwidth resource allocation by establishing an objective function and constraint conditions based on a network load prediction analysis result, solves the problems of resource conflict and allocation efficiency, adjusts resource allocation by using a resource optimization algorithm and generates a formulated bandwidth strategy scheme;
the resource allocation sub-module adopts a dynamic resource allocation technology based on a formulated bandwidth policy scheme, adjusts the resource allocation policy by monitoring network flow and bandwidth requirements in real time, and matches real-time changes of network loads to generate an optimized bandwidth allocation plan.
4. The four-port gigabit network card dynamic bandwidth allocation system of claim 1, wherein: the hierarchical dynamic adjustment module comprises a burst response sub-module, a trend matching sub-module and a strategy implementation sub-module;
the burst response submodule responds to the flow peak value and reacts to the burst flow change by adopting a real-time data analysis algorithm and monitoring and analyzing the real-time comparison of the network flow and a preset threshold value to generate a burst flow response strategy;
the trend matching sub-module calculates smooth flow trend by weighting analysis of historical flow data based on a sudden flow response strategy by adopting a statistical time sequence analysis method, and comprises the steps of weighted average calculation of the historical flow data and generation of trend lines, predicting and matching continuous flow change, and generating a flow trend matching scheme;
the strategy implementation submodule adopts a multi-layer network management strategy based on a flow trend matching scheme, and dynamically adjusts the bandwidth by comprehensively analyzing data of multiple time scales and flow changes, wherein the dynamic adjustment comprises continuous monitoring of the state of each layer of network and bandwidth adjustment based on monitoring data, and generates a hierarchical adjustment strategy.
5. The four-port gigabit network card dynamic bandwidth allocation system of claim 1, wherein: the network simulation and evaluation module comprises a flow simulation sub-module, a network performance evaluation sub-module and a bottleneck identification sub-module;
the flow simulation submodule adopts a queuing theory model based on the current network configuration and historical flow data, simulates arrival and processing processes of various flows in a network by establishing a multi-type queuing model, and predicts future bandwidth requirements by analyzing trend and fluctuation of past flow data in combination with a historical data analysis technology to generate a network flow simulation result;
the network performance evaluation submodule adopts a network performance analysis technology based on the network flow simulation result, calculates and analyzes performance indexes of the simulation data, including delay, bandwidth utilization rate and packet loss rate, analyzes network performance in multiple aspects, and generates a network performance evaluation result;
the bottleneck recognition sub-module is used for recognizing and analyzing abnormal efficiency areas in simulation and evaluation data by adopting a bottleneck analysis technology based on network performance evaluation results, revealing bottleneck problems in a network, including insufficient bandwidth and equipment overload, and generating network behavior simulation evaluation results.
6. The four-port gigabit network card dynamic bandwidth allocation system of claim 1, wherein: the intelligent logic control module comprises a real-time monitoring sub-module, a decision logic sub-module and an adjustment execution sub-module;
the real-time monitoring sub-module is based on a network behavior simulation evaluation result, adopts a flow analysis and monitoring algorithm, and monitors the current state of the network by analyzing a plurality of indexes of the network flow, including speed, density and data packet type, including capturing network data in real time and analyzing the characteristics and variation trend of the data flow to generate a network flow analysis result;
the decision logic submodule adopts a fuzzy logic algorithm based on the network flow analysis result, and performs bandwidth allocation decision under the network environment by processing fuzzy input data, and comprises the steps of establishing a fuzzy rule set, performing fuzzy processing on input variables, performing fuzzy reasoning, making a bandwidth allocation strategy and generating a bandwidth allocation decision scheme;
the adjustment execution submodule adopts a self-adaptive control technology based on a bandwidth allocation decision scheme, adjusts network configuration to match real-time network conditions by dynamically analyzing network load and bandwidth demands, and comprises the steps of continuously monitoring network performance indexes, adjusting control parameters according to performance feedback, implementing dynamic bandwidth allocation and generating a logic control strategy.
7. The four-port gigabit network card dynamic bandwidth allocation system of claim 1, wherein: the quantitative analysis and adjustment module comprises a flow analysis sub-module, a logic reasoning sub-module and a scheme optimization sub-module;
the flow analysis submodule adopts a decision tree analysis method based on a logic control strategy, analyzes historical and real-time data by constructing a decision tree model, and identifies multiple types of network flow modes, and comprises the steps of analyzing data characteristics, classifying the flow data according to the importance of the multiple characteristics, distinguishing and identifying multiple modes of network flow, and generating a flow mode identification result;
the logic reasoning sub-module is based on the flow pattern recognition result, adopts a probability network analysis technology, constructs a Bayesian network model by applying Bayesian network and probability inference, performs probability calculation on the flow pattern, evaluates the influence on bandwidth allocation by using probability data, and generates a probability influence analysis result of the flow pattern;
the scheme optimizing submodule analyzes the instant bandwidth use condition and the predicted flow demand by adopting a linear programming and heuristic search method based on the probability influence analysis result of the flow mode, analyzes and adjusts the bandwidth allocation scheme, and matches various flow modes and network demands to generate a quantized adjustment scheme.
8. The four-port gigabit network card dynamic bandwidth allocation system of claim 1, wherein: the geometric space optimization module comprises a space mapping sub-module, an optimization calculation sub-module and an allocation implementation sub-module;
the space mapping submodule adopts a multidimensional space mapping method based on a quantization adjustment scheme, and performs geometric analysis operation by expressing the bandwidth allocation problem in a multidimensional space, wherein the geometric analysis operation comprises the steps of converting parameters and constraints of bandwidth allocation into points and lines in a geometric space by applying a space coordinate transformation technology, and generating a geometric space mapping model;
the optimization calculation submodule is based on a geometric space mapping model, adopts a convex optimization technology, performs bandwidth allocation solution optimizing operation in a geometric space through a linear programming and nonlinear optimization method, and comprises the steps of identifying convex sets and convex functions, applying a mathematical optimization algorithm to solve, and generating a bandwidth optimization scheme in the geometric space;
the allocation implementation submodule adopts a parameter mapping and configuration application method based on a bandwidth optimization scheme in a geometric space, converts a convex optimization result into a network configuration parameter through a network parameter conversion technology, adjusts a bandwidth setting matching allocation strategy of network equipment and generates a geometric optimization result.
9. The four-port gigabit network card dynamic bandwidth allocation system of claim 1, wherein: the dynamic bandwidth allocation module comprises a real-time monitoring sub-module, a bandwidth adjustment sub-module and an execution feedback sub-module;
the real-time monitoring submodule adopts a network flow real-time monitoring technology based on a hierarchical adjustment strategy and a geometric optimization result, analyzes real-time network data through a data packet analysis and flow rate measurement algorithm, and comprises the steps of capturing a network data packet in real time, analyzing the size, the speed and the flow direction of the network data packet, quantitatively evaluating the overall trend of the network flow and generating a network flow real-time analysis result;
the bandwidth adjustment submodule adopts a dynamic bandwidth adjustment mechanism based on a network flow real-time analysis result, and adjusts bandwidth configuration according to real-time monitoring data and flow prediction through a bandwidth demand prediction and resource allocation algorithm to generate a bandwidth adjustment scheme;
the execution feedback submodule adopts a network performance feedback technology based on a bandwidth adjustment scheme, evaluates the network condition after bandwidth adjustment through an effect monitoring and data analysis tool, and comprises the steps of collecting data of bandwidth use, flow distribution and delay, analyzing whether the adjusted effect accords with a preset performance target, and generating a self-adaptive bandwidth allocation scheme.
10. The four-port gigabit network card dynamic bandwidth allocation system of claim 1, wherein: the system integration and monitoring module comprises a function integration sub-module, a state monitoring sub-module and a system performance monitoring sub-module;
the function integration submodule integrates functions and data by adopting a data fusion and strategy coordination technology based on an optimized bandwidth allocation plan, a hierarchy adjustment strategy, a network behavior simulation evaluation result, a logic control strategy, a quantitative adjustment scheme, a geometric optimization result and a self-adaptive bandwidth allocation scheme, and performs summarization analysis to generate a system function integration result;
the state monitoring submodule monitors the running state of the system, including the bandwidth utilization rate and the flow distribution, by adopting a real-time state monitoring technology and a dynamic data tracking and analyzing method based on the system function integration result to generate a system real-time state monitoring result;
the system performance monitoring submodule is used for generating a system comprehensive state overview by adopting a comprehensive performance analysis technology and adopting a performance index evaluation and optimization strategy to continuously track and evaluate the performance index of the system, including response time and bandwidth efficiency.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410177712.4A CN117729164B (en) | 2024-02-08 | 2024-02-08 | Dynamic bandwidth allocation system for four-port gigabit network card |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410177712.4A CN117729164B (en) | 2024-02-08 | 2024-02-08 | Dynamic bandwidth allocation system for four-port gigabit network card |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117729164A true CN117729164A (en) | 2024-03-19 |
| CN117729164B CN117729164B (en) | 2024-04-12 |
Family
ID=90202013
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410177712.4A Active CN117729164B (en) | 2024-02-08 | 2024-02-08 | Dynamic bandwidth allocation system for four-port gigabit network card |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117729164B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060120282A1 (en) * | 2000-05-19 | 2006-06-08 | Carlson William S | Apparatus and methods for incorporating bandwidth forecasting and dynamic bandwidth allocation into a broadband communication system |
| CN101466124A (en) * | 2007-12-19 | 2009-06-24 | 中国移动通信集团公司 | Control method and system for network bandwidth |
| US20160182329A1 (en) * | 2014-12-19 | 2016-06-23 | Ciena Corporation | Bandwidth optimization systems and methods in networks |
| CN112202619A (en) * | 2020-10-12 | 2021-01-08 | 浪潮云信息技术股份公司 | Intelligent cloud computing network flow adjusting and optimizing system and method |
| CN112367276A (en) * | 2020-12-31 | 2021-02-12 | 南京群顶科技有限公司 | Network resource dynamic self-adaption method and system based on network flow priority |
| CN117395166A (en) * | 2023-12-11 | 2024-01-12 | 四海良田(天津)智能科技有限公司 | Intelligent agricultural management platform based on Internet of things |
| CN117527721A (en) * | 2023-11-24 | 2024-02-06 | 深圳Tcl新技术有限公司 | Content distribution network bandwidth control method, device, equipment and storage medium |
-
2024
- 2024-02-08 CN CN202410177712.4A patent/CN117729164B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060120282A1 (en) * | 2000-05-19 | 2006-06-08 | Carlson William S | Apparatus and methods for incorporating bandwidth forecasting and dynamic bandwidth allocation into a broadband communication system |
| CN101466124A (en) * | 2007-12-19 | 2009-06-24 | 中国移动通信集团公司 | Control method and system for network bandwidth |
| US20160182329A1 (en) * | 2014-12-19 | 2016-06-23 | Ciena Corporation | Bandwidth optimization systems and methods in networks |
| CN112202619A (en) * | 2020-10-12 | 2021-01-08 | 浪潮云信息技术股份公司 | Intelligent cloud computing network flow adjusting and optimizing system and method |
| CN112367276A (en) * | 2020-12-31 | 2021-02-12 | 南京群顶科技有限公司 | Network resource dynamic self-adaption method and system based on network flow priority |
| CN117527721A (en) * | 2023-11-24 | 2024-02-06 | 深圳Tcl新技术有限公司 | Content distribution network bandwidth control method, device, equipment and storage medium |
| CN117395166A (en) * | 2023-12-11 | 2024-01-12 | 四海良田(天津)智能科技有限公司 | Intelligent agricultural management platform based on Internet of things |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117729164B (en) | 2024-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108259367B (en) | Service-aware flow strategy customization method based on software defined network | |
| US7403988B1 (en) | Technique for autonomous network provisioning | |
| Chen et al. | Knowledge-based autonomous service provisioning in multi-domain elastic optical networks | |
| CN110891283A (en) | Small base station monitoring device and method based on edge calculation model | |
| CN113259355B (en) | Industrial Internet identification slice management system based on SDN | |
| CN106027288A (en) | Communication traffic prediction method for distribution line information monitoring service | |
| Kumar et al. | A comparative study of call admission control in mobile multimedia networks using soft computing | |
| CN117459457A (en) | Access method based on distributed core network | |
| CN114595970A (en) | Resource scheduling intelligent decision method and device, electronic equipment and storage medium | |
| CN117729164B (en) | Dynamic bandwidth allocation system for four-port gigabit network card | |
| CN116684472A (en) | Service deployment system and service deployment method for terminal-side computing network | |
| CN115604724A (en) | SRv6 method for evaluating risk of congestion of backbone network flow | |
| CN116109058A (en) | Substation inspection management method and device based on deep reinforcement learning | |
| Dronyuk et al. | High quality video traffic ateb-forecasting and fuzzy logic management | |
| Syal et al. | Automatic detection of network traffic anomalies and changes | |
| CN115495231A (en) | Dynamic resource scheduling method and system under complex scene of high concurrent tasks | |
| Ruiz et al. | End-to-end KPI analysis in converged fixed-mobile networks | |
| Piedra et al. | Study of the application of neural networks in internet traffic engineering | |
| CN111327494A (en) | Multi-domain SDN network traffic situation assessment method and system | |
| Sohail et al. | Simplifying network management with fuzzy logic | |
| Perepelkin et al. | Problem of Network Traffic Prediction in Multiprovider Cloud Infrastructures Based on Neural Networks Methods | |
| CN117560701B (en) | Remote management method and system for wireless router | |
| Hemalatha | Traffic control in 5g heterogenous network | |
| Suresh et al. | Performance Evaluation of Supervisory Machine Learning Algorithms for Cellular Networks Payload Enhancement in Downlink | |
| CN117896776B (en) | Data automatic retransmission method and related device for wireless ad hoc network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |