Continuous education informatization management system based on cloud platform

Abstract

The invention relates to the technical field of informatization management, and discloses a continuous education informatization management system based on a cloud platform, a honeycomb unit topology is constructed, each honeycomb unit is adjacent to six units, course identifiers are mapped to G course groups through a Hash algorithm and modulo operation, and each course group is divided into G groups; dividing request moments according to a sliding time window and a time slice width to generate a time slice identifier, and distributing a course selection request to a unique honeycomb unit; and adaptively adjusting the processing capacity of the honeycomb unit based on the historical request mean value, the standard deviation and the preset lower limit, controlling the request processing rhythm through a token bucket mechanism, and transferring the overflow request to an adjacent unit according to a polling mode when the queue exceeds a preset threshold value. And after the flood peak processing is finished, updating the historical request parameters according to the processing data to optimize the subsequent capacity configuration. The problems of space-time traffic imbalance and low resource allocation efficiency of a traditional system are solved, and space-time decentralized processing, dynamic load balancing and high-concurrency stability improvement of course selection requests are achieved.

Description

Technical Field

[0001] The present invention relates to the field of information management technology, and more specifically, to a cloud platform-based continuing education information management system. Background Art

[0002] The publication number is CN118279100B, which is an intelligent course selection method and system for research and study education. It relates to an intelligent course selection method and system for research and study education, which obtains the course duration of research and study courses obtained by users of the research and study education platform, obtains the course selection ratio of users of the research and study education platform, obtains the course selection recommendation value through the peak number of people and the course selection ratio, and matches the user time with the research and study courses that are suitable for the course according to the course selection recommendation value. For course projects that require research and study practice, this invention can effectively improve the efficiency of school course selection, improve the efficiency of acceptance and management of research and study bases, improve the floor efficiency of research and study bases, make full and reasonable use of the venues of research and study practice bases, and avoid the gathering of research and study practice activities.

[0003] However, in the field of continuing education information management, the above-mentioned applications face the dual challenges of unbalanced flow in the spatial and temporal dimensions and inefficient resource allocation when selecting courses, resulting in the following problems:

[0004] Continuing education course selection often leads to a sudden increase in instantaneous traffic due to concentrated visits by students. The traditional single-queue queuing model cannot disperse the pressure caused by differences in time and course types.

[0005] Requests for popular course groups are concentrated on a single server node. Traditional load balancing technology can only distribute traffic among physical nodes and cannot perform logical diffusion based on course type and time dimensions. Summary of the Invention

[0006] The present invention provides a cloud-based continuing education information management system to solve the technical problems raised in the background technology.

[0007] The present invention provides a cloud platform-based continuing education information management system, including: a honeycomb unit topology; wherein the honeycomb unit C is adjacent to the honeycomb unit set N in the honeycomb unit topology. c ,|N c |=6;

[0008] The first management module 101 is used to set the number of course groups to G and the time slice width to Δτ;

[0009] At time t, the course identifier C of the course selection request is id and the request time Q t After combination, it is mapped to honeycomb unit c, which includes processing capacity C c , and make adaptive adjustments based on historical request parameters;

[0010] In response to the number of course selection requests on the target cloud platform within a unit time being greater than or equal to a preset request threshold, the target cloud platform triggers a peak breaking process, including:

[0011] Step 1: Receive the course selection request and extract the course identifier C of the course selection request id Perform the first mapping process to determine the course group g; set the course selection request time Q t Perform a second mapping process to determine the time slice identifier τ; assign the course selection request to the honeycomb unit C based on the course group g and the time slice identifier τ of the course selection request;

[0012] Step 2: Based on the processing capacity C of honeycomb unit C c and the time slice width Δτ, determine the token generation rate μ of the honeycomb unit c c , and trigger the token bucket mechanism as follows:

[0013] Based on the token generation rate μ c Determine the number of tokens in honeycomb unit c as T c (Q t );

[0014] If T c (Q t )≥1, the course selection request is processed by honeycomb unit c and 1 token is consumed synchronously; otherwise, the course selection request is added to the queue Q of honeycomb unit c. w (Q t );

[0015] If the queue Q w (Q t )>θC c , then the overflow course selection request is transferred to the adjacent honeycomb unit set N in a round-robin manner c For any honeycomb unit n, θ is a preset constant;

[0016] Step 3: Based on the queue Q w (O t ) and the token generation rate μ c Returns the waiting time for course selection request;

[0017] The second management module 102 is configured to stop the peak breaking process on the target cloud platform in response to the number of course selection requests on the target cloud platform being less than a preset request threshold within a unit time, including:

[0018] Obtain the course selection request processing data of honeycomb unit c during the flood peak breaking processing period, and update the historical request parameters of honeycomb unit c.

[0019] Further, receive the course selection request and extract the course identifier C of the course selection requestid Performing a first mapping process to determine the course group g includes:

[0020] Mark the course as C id Obtain the corresponding hash value through the hash algorithm;

[0021] Perform a modulo calculation on the hash value and add one to the modulo result to obtain the course group g; where 1≤g≤G.

[0022] Furthermore, the request time Q of the course selection request is t Performing a second mapping process to determine the time slice identifier τ includes:

[0023] Generate a timeline starting from the moment when the target cloud platform triggers the flood peak breaking process;

[0024] Initialize the sliding time window H with a sliding step size of H;

[0025] The time axis is divided by sliding the time window H to determine the request time Q t Relative time in the sliding time window H;

[0026] Based on the ratio of the relative time and the time slice width Δτ, the time slice identifier τ of the course selection request is obtained.

[0027] Furthermore, based on the course group g and the time slice identifier τ of the course selection request, the course selection request is assigned to the honeycomb unit c, including:

[0028] Construct binary coordinates (g, τ) based on the course group g and time slice identifier τ of the course selection request;

[0029] A unique honeycomb cell is assigned to each binary coordinate to assign the course selection request to the honeycomb cell c; the number of honeycomb cells c is

[0030] Furthermore, according to the processing capacity C of the honeycomb unit c c and the time slice width Δτ, determine the token generation rate μ of the honeycomb unit c c ,

[0031] Furthermore, based on the token generation rate μ c Determine the number of honeycomb unit C tokens to be T c ,include:

[0032] T c (Q t )=F(T c (Q t-1 ))+μ c ×(Q t -Qt-1 )

[0033] F(T c (Q t-1 ))=T c (Q t-1 )-U

[0034] Among them, T c (Q t ) represents the honeycomb unit C at the request time Q t The number of tokens, T c (Q t-1 ) represents the honeycomb unit c at the request time Q t-1 The number of tokens, F(T c (Q t-1 )) represents the honeycomb unit c at the request time Q t-1 To Q t The number of remaining tokens, U, represents the number of tokens in the honeycomb unit c at the request time Q t-1 To Q t The number of course enrollment requests processed.

[0035] Furthermore, the overflow course selection request is transferred to the adjacent honeycomb unit set N in a round-robin manner. c Any honeycomb unit n includes:

[0036] Query the remaining processing capacity of the honeycomb unit n adjacent to the honeycomb unit c in the preset order, n∈N c ;

[0037] If the remaining processing capacity of honeycomb unit n is greater than the preset processing capacity, the overflowed course selection requests will be transferred to honeycomb unit n for processing;

[0038] The remaining processing capacity is based on the honeycomb unit n at the request time Q t The number of tokens and the queue Q n (Q t ) is determined by the difference.

[0039] Furthermore, based on the queue Q w (Q t ) and the token generation rate μ c Returns the waiting time T for course selection request w ,

[0040] Furthermore, the course selection request processing data of honeycomb unit c during the flood peak breaking processing period is obtained, and the historical request parameters of honeycomb unit c are updated. The historical request parameters include: historical request mean and historical request standard deviation;

[0041] The ratio of the number and value of course selection requests processed by honeycomb unit c in each peak breaking processing period in the historical time period to the duration and value of each peak breaking processing period in the historical time period is taken as the historical request mean.

[0042] The standard deviation of the number of course selection requests processed by honeycomb unit c in each flood peak breaking processing time period within the historical time period and the historical request mean is taken as the historical request standard deviation σ.

[0043] Furthermore, the processing capacity C of the honeycomb unit c is adaptively adjusted based on the historical request parameters. c ,include:

[0044]

[0045] Among them, ∈ represents the minimum processing capacity of the honeycomb unit c, α and β represent capacity weight coefficients, and the sum of α and β is 1, and neither is 0.

[0046] The beneficial effects of the present invention are as follows: by constructing a honeycomb unit topology, the course selection requests are binary mapped in the course grouping and time slice dimensions, the processing rate is dynamically controlled in combination with the token bucket mechanism, and the processing capacity is adaptively adjusted based on historical data, breaking through the bottleneck of the traditional system's time-space traffic imbalance and inefficient resource allocation. The present application achieves uniform dispersion of course selection requests in the time-space dimension, avoiding the overload of a single node caused by the concentration of traffic in popular courses and peak hours; by polling and transferring overflow requests through adjacent units, a local load balancing mechanism is constructed, which improves the stability of the system in high-concurrency scenarios; the processing capacity is dynamically optimized using historical request parameters, so that resource allocation is accurately matched with the actual load, significantly improving the cloud platform's processing efficiency and resource utilization for continuing education course selection scenarios, and effectively solving the limitations of the traditional single queue model and physical node load balancing. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a module diagram of a cloud platform-based continuing education information management system of the present invention;

[0048] In the figure: a first management module 101 and a second management module 102. DETAILED DESCRIPTION

[0049] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed solely to enable those skilled in the art to better understand and implement the subject matter described herein, and that the functions and arrangements of the elements discussed may be varied without departing from the scope of this specification. Various examples may omit, substitute, or add various processes or components as needed. In addition, features described with respect to some examples may also be combined in other examples.

[0050] like Figure 1 As shown, a cloud platform-based continuing education information management system includes: a honeycomb unit topology; wherein the honeycomb unit c is adjacent to the honeycomb unit set N in the honeycomb unit topology. c ,|N c |=6;

[0051] The first management module 101 is used to set the number of course groups to G and the time slice width to Δτ;

[0052] At time t, the course identifier C of the course selection request is id and the request time Q t After combination, it is mapped to honeycomb unit c, which includes processing capacity C c , and make adaptive adjustments based on historical request parameters;

[0053] In response to the number of course selection requests on the target cloud platform within a unit time being greater than or equal to a preset request threshold, the target cloud platform triggers a peak breaking process, including:

[0054] Step 1: Receive the course selection request and extract the course identifier C of the course selection request id Perform the first mapping process to determine the course group g; set the course selection request time Q t Perform a second mapping process to determine the time slice identifier τ; assign the course selection request to the honeycomb unit c based on the course group g and the time slice identifier τ of the course selection request;

[0055] Step 2: Based on the processing capacity C of honeycomb unit c c and the time slice width Δτ, determine the token generation rate μ of the honeycomb unit c c , and trigger the token bucket mechanism as follows:

[0056] Based on the token generation rate μ c Determine the number of tokens in honeycomb unit c as T c (Q t );

[0057] If T c (Q t )≥1, the course selection request is processed by honeycomb unit c and 1 token is consumed synchronously; otherwise, the course selection request is added to the queue Q of honeycomb unit c. w (Q t );

[0058] If the queue Q w (Q t )>θC c , then the overflow course selection request is transferred to the adjacent honeycomb unit set N in a round-robin manner cFor any honeycomb unit n, θ is a preset constant;

[0059] Step 3: Based on the queue Q w (Q t ) and the token generation rate μ c Returns the waiting time for course selection request;

[0060] The second management module 102 is configured to stop the peak breaking process on the target cloud platform in response to the number of course selection requests on the target cloud platform being less than a preset request threshold within a unit time, including:

[0061] Obtain the course selection request processing data of honeycomb unit c during the flood peak breaking processing period, and update the historical request parameters of honeycomb unit c.

[0062] Preferably, the honeycomb units are divided based on the logic of OpenStack;

[0063] OpenStack is an open-source cloud computing management platform that implements the logical division of resources through the collaboration of multiple components, as follows:

[0064] Computing resource partitioning (Nova component): Using Nova's InstanceGroup feature, you can control the processing capacity of a honeycomb unit by assigning different vCPU, memory, and other resources to instance groups. For example, a binary coordinate (g, τ) can be used as an instance group to allocate resources.

[0065] Time Slice Mapping (Task Scheduling): Combined with scheduled task tools such as Cron, the cycle is divided by time slice width (preferably 5 seconds). Within each cycle, the Neutron component routes request traffic to the corresponding instance group based on the course group g, achieving a combined mapping of time slice identifiers and course groups to form a logical honeycomb unit c.

[0066] Use Neutron to create an independent virtual network for each course group.

[0067] Deploy the router Router_{g=10} in Neutron, and connect the instance groups of different time slices under the same course group g to Router_{g=10} through port mapping.

[0068] For adjacent cellular units n, a logical connection is established through router cascade or distributed routing mechanism across virtual networks. For example, Router_{g=10} and Router_{g=11} are connected through port forwarding rules, allowing the adjacent topology rules to be met. c Traffic interaction of |=6 ensures request transfer between honeycomb unit n and honeycomb unit c.

[0069] In one embodiment of the present invention, a course selection request is received, and a course identifier C of the course selection request is extracted. id Performing a first mapping process to determine the course group g includes:

[0070] Mark the course as C id Obtain the corresponding hash value through the hash algorithm;

[0071] Perform a modulo calculation on the hash value and add one to the modulo result to obtain the course group g; where 1≤g≤G.

[0072] It should be noted that for each course selection request, the course ID is first extracted. The course ID is then calculated using a hash algorithm. This algorithm can convert input data of any length into a fixed-length hash value, which is a numerical result. The characteristics of the hash algorithm (randomness and uniform distribution) ensure that the hash values ​​obtained from different course IDs are evenly and discretely distributed, laying the foundation for balanced grouping in the subsequent process.

[0073] A modulo operation is performed on the hash value, with the modulus being G (the number of course groups set), and the modulo result ranging from 0 to G-1. Then, an increment operation is performed on the modulo result, so that the final result ranges from 1 to G, thereby determining the course group g.

[0074] For example, G=10, and the hash value of a course identifier obtained by the hash algorithm is 23. 23 modulo 10 is 3, and then adding one to it is g=4, so the course is assigned to the 4th course group.

[0075] In one embodiment of the present invention, the request time Q of the course selection request is t Performing a second mapping process to determine the time slice identifier τ includes:

[0076] Generate a timeline starting from the moment when the target cloud platform triggers the flood peak breaking process;

[0077] Initialize the sliding time window H with a sliding step size of H;

[0078] The time axis is divided by sliding the time window H to determine the request time Q t Relative time in the sliding time window H;

[0079] Based on the ratio of the relative time and the time slice width Δτ, the time slice identifier τ of the course selection request is obtained.

[0080] It should be noted that the timeline starts at the moment when the target cloud platform triggers the peak-breaking process. For example, if the platform triggers the peak-breaking process at 10:00:00 on May 15, 2025, this moment will be the starting point of the timeline (T0). All subsequent course selection requests will be measured based on this starting point, forming a relative time system that provides a benchmark for unified request processing time.

[0081] Initialize a fixed-length sliding time window H, and the sliding step is equal to H. For example, if H = 300 seconds, the first time window is [T0, T 300 ], the second one is [T 300 , T 600 ], and so on. This setting ensures that the timeline is divided into continuous and non-overlapping window segments, which is convenient for segmented processing of the requested moments.

[0082] Compare the request time with the start of the timeline to determine the sliding time window into which the request time falls. Calculate its relative time within that window. Divide the relative time by the time slice width and round it up to the integer to obtain the time slice identifier.

[0083] In one embodiment of the present invention, the course selection request is assigned to the honeycomb unit C based on the course group g and the time slice identifier τ of the course selection request, including:

[0084] Construct binary coordinates (g, τ) based on the course group g and time slice identifier τ of the course selection request;

[0085] A unique honeycomb unit is assigned to each binary coordinate to assign the course selection request to the honeycomb unit c; the number of honeycomb units c is

[0086] It should be noted that the course grouping based on the course selection request is hashed, modulo-added, and combined with the time slice identifier to construct a binary coordinate. This coordinate is a unique identifier. For example, if g is 5 and the time slice identifier is 10, the binary coordinate is (5, 10), which reflects information about both the course grouping dimension (spatial dimension) and the time slice dimension (temporal dimension).

[0087] Each binary coordinate is assigned a unique honeycomb cell. Each course registration request is processed in a specific honeycomb cell. For example, the binary coordinate (5, 10) corresponds to honeycomb cell c. All course registration requests with the binary coordinates (5, 10) are processed in honeycomb cell c.

[0088] The number of honeycomb cells is in, τ represents the number of time slices obtained by dividing the sliding time window H by the time slice width Δτ, and G represents the number of course groups. The product of the two represents the number of combinations in the spatiotemporal dimension, that is, each course group corresponds to an independent honeycomb unit in each time slice.

[0089] In one embodiment of the present invention, according to the processing capacity C of the honeycomb unit c c and the time slice width Δτ, determine the token generation rate μ of the honeycomb unit c c ,

[0090] It should be noted that μ c Indicates: the token generation rate of honeycomb unit c, that is, the number of tokens generated per unit time (such as per second), which is used to control the processing rhythm of course selection requests.

[0091] C c Indicates: The processing capacity of honeycomb unit c refers to the maximum number of course selection requests that the unit can process within a time slice width Δτ, which reflects the processing power of the unit.

[0092] Δτ: Time slice width, which is the time segment length preset by the system.

[0093] Determine the token generation rate. The token bucket mechanism uses this rate to generate tokens, consuming one token for each course registration request. If insufficient tokens are available, the request will be queued. This prevents unit overload caused by rapid request processing, ensures system stability and processing efficiency in high-concurrency scenarios, and ensures that the processing capacity of the honeycomb unit c matches the actual request traffic.

[0094] In one embodiment of the present invention, based on the token generation rate μ c Determine the number of honeycomb unit C tokens to be T c ,include:

[0095] T c (Q t )=F(T c (Q t-1 ))+μ c ×(Q t -Q t-1 )

[0096] F(T c (Q t-1 ))=T c (Q t-1 )-U

[0097] Among them, T c (Q t ) represents the honeycomb unit C at the request time Q tThe number of tokens, T c (Q t-1 ) represents the honeycomb unit c at the request time Q t-1 The number of tokens, F(T c (Q t-1 )) represents the honeycomb unit c at the request time Q t-1 To Q t The number of remaining tokens, U, represents the number of tokens in the honeycomb unit c at the request time Q t-1 To Q t The number of course enrollment requests processed.

[0098] In one embodiment of the present invention, overflow course selection requests are transferred to the adjacent honeycomb unit set N in a round-robin manner. c Any honeycomb unit n includes:

[0099] Query the remaining processing capacity of the honeycomb unit n adjacent to the honeycomb unit c in the preset order, n∈N c ;

[0100] If the remaining processing capacity of honeycomb unit n is greater than the preset processing capacity, the overflowed course selection requests will be transferred to honeycomb unit n for processing;

[0101] The remaining processing capacity is based on the honeycomb unit n at the request time Q t The number of tokens and the queue Q n (Q t ) is determined by the difference.

[0102] It should be noted that the system checks the remaining processing capacity of the honeycomb unit n adjacent to the honeycomb unit c in sequence according to a preset order (custom order).

[0103] By polling neighboring cells to determine their remaining processing capacity and determining whether to transfer overflow requests based on the number of tokens and remaining queue space, the system achieves load balancing. This prevents a single cell from being overloaded by too many requests and fully utilizes the idle processing capacity of neighboring cells, improving the stability and processing efficiency of the entire system in high-concurrency scenarios. This ensures that course selection requests are processed promptly and effectively, optimizing user experience and resource utilization.

[0104] In one embodiment of the present invention, based on the queue Q w (Q t ) and the token generation rate μ c Returns the waiting time T for course selection request w ,

[0105] By calculating and returning the estimated waiting time for course selection requests in real time, users are provided with clear feedback to keep them informed of the progress of the request processing.

[0106] In one embodiment of the present invention, the course selection request processing data of the honeycomb unit c in the peak breaking processing time period is obtained, and the historical request parameters of the honeycomb unit c are updated. The historical request parameters include: the historical request mean and the historical request standard deviation;

[0107] The ratio of the number and value of course selection requests processed by honeycomb unit c in each peak breaking processing period in the historical time period to the duration and value of each peak breaking processing period in the historical time period is taken as the historical request mean.

[0108] The standard deviation of the number of course selection requests processed by honeycomb unit c in each flood peak breaking processing time period within the historical time period and the historical request mean is taken as the historical request standard deviation σ.

[0109] In one embodiment of the present invention, the processing capacity C of the honeycomb unit c is adaptively adjusted based on historical request parameters. c ,include:

[0110]

[0111] Among them, ∈ represents the minimum processing capacity of the honeycomb unit c, α and β represent capacity weight coefficients, and the sum of α and β is 1, and neither is 0.

[0112] It should be noted that the minimum value of the processing capacity of the honeycomb unit c is a preset guarantee value. For example, if ∈ = 50, to ensure that C c No less than 50 to avoid request backlogs caused by insufficient processing capacity.

[0113] By integrating the historical load mean and fluctuation, C c Can be dynamically adjusted according to actual request patterns. In the popular course grouping, Increase, improve C c To cope with high traffic; in unpopular groups, C c Not less than 50 to ensure basic processing capacity.

[0114] The above describes the embodiments of this embodiment, but this embodiment is not limited to the above specific implementation methods. The above specific implementation methods are merely illustrative and not restrictive. Ordinary technicians in this field can also make many forms based on the inspiration of this embodiment, all of which are protected by this embodiment.

Claims

1. A cloud-based continuing education information management system, characterized in that: include: Honeycomb cell topology; wherein, honeycomb cell c is adjacent to honeycomb cell set N in the honeycomb cell topology c ,|N c |=6; The first management module 101 is used to set the number of course groups to G and the time slice width to Δτ; At time t, the course identifier C of the course selection request is id and the request time Q t After combination, it is mapped to honeycomb unit c, which includes processing capacity C c , and make adaptive adjustments based on historical request parameters; In response to the number of course selection requests on the target cloud platform within a unit time being greater than or equal to a preset request threshold, the target cloud platform triggers a peak breaking process, including: Step 1: Receive the course selection request and extract the course identifier C of the course selection request id Perform the first mapping process to determine the course group g; set the course selection request time Q t Perform a second mapping process to determine the time slice identifier τ; assign the course selection request to the honeycomb unit c based on the course group g and the time slice identifier τ of the course selection request; Step 2: Based on the processing capacity C of honeycomb unit c c and the time slice width Δτ, determine the token generation rate μ of the honeycomb unit c c , and trigger the token bucket mechanism as follows: Based on the token generation rate μ c Determine the number of tokens in honeycomb unit c as T c (Q t ); If T c (Q t )≥1, the course selection request is processed by honeycomb unit c and 1 token is consumed synchronously; otherwise, the course selection request is added to the queue Q of honeycomb unit c. w (Q t ); If the queue Q w (Q t )>θC c , then the overflow course selection request is transferred to the adjacent honeycomb unit set N in a round-robin manner c For any honeycomb unit n, θ is a preset constant; Step 3: Based on the queue Q w (Q t ) and the token generation rate μ c Returns the waiting time for course selection request; The second management module 102 is configured to stop the peak breaking process on the target cloud platform in response to the number of course selection requests on the target cloud platform being less than a preset request threshold within a unit time, including: Obtain the course selection request processing data of honeycomb unit c during the flood peak breaking processing period, and update the historical request parameters of honeycomb unit c.

2. A cloud-based continuing education information management system according to claim 1, characterized in that: Receive the course selection request and extract the course identifier C of the course selection request id Performing a first mapping process to determine the course group g includes: Mark the course as C id Obtain the corresponding hash value through the hash algorithm; Perform a modulo calculation on the hash value and add one to the modulo result to obtain the course group g; where 1≤g≤G.

3. The cloud-based continuing education information management system according to claim 2, characterized in that: The time of requesting course selection is Q t Performing a second mapping process to determine the time slice identifier τ includes: Generate a timeline starting from the moment when the target cloud platform triggers the flood peak breaking process; Initialize the sliding time window H with a sliding step size of H; The time axis is divided by sliding the time window H to determine the request time Q t Relative time in the sliding time window H; Based on the ratio of the relative time and the time slice width Δτ, the time slice identifier τ of the course selection request is obtained.

4. The cloud-based continuing education information management system according to claim 3 is characterized in that: Based on the course group g and time slice identifier τ of the course request, the course request is assigned to the honeycomb unit c, including: Construct binary coordinates (g, τ) based on the course group g and time slice identifier τ of the course selection request; A unique honeycomb unit is assigned to each binary coordinate to assign the course selection request to the honeycomb unit c; the number of honeycomb units c is 5. The cloud-based continuing education information management system according to claim 4 is characterized in that: According to the processing capacity C of honeycomb unit c c and the time slice width Δτ, determine the token generation rate μ of the honeycomb unit c c , 6. The cloud-based continuing education information management system according to claim 5, characterized in that: Based on the token generation rate μ c Determine the number of tokens in honeycomb unit c as T c ,include: T c (Q t )=F(T c (Q t-1 ))+μ c ×(Q t -Q t-1 ) F(T c (Q t-1 ))=T c (Q t-1 )-U Among them, T c (Q t ) represents the honeycomb unit c at the request time Q t The number of tokens, T c (Q t-1 ) represents the honeycomb unit c at the request time Q t-1 The number of tokens, F(T c (Q t-1 )) represents the honeycomb unit c at the request time Q t-1 To Q t The number of remaining tokens, U, represents the number of tokens in the honeycomb unit c at the request time Q t-1 To Q t The number of course enrollment requests processed.

7. The cloud-based continuing education information management system according to claim 6, characterized in that: Transfer the overflow course selection requests to the adjacent honeycomb unit set N in a round-robin manner c Any honeycomb unit n includes: Query the remaining processing capacity of the honeycomb unit n adjacent to the honeycomb unit c in the preset order, n∈N c ; If the remaining processing capacity of honeycomb unit n is greater than the preset processing capacity, the overflowed course selection requests will be transferred to honeycomb unit n for processing; The remaining processing capacity is based on the honeycomb unit n at the request time Q t The number of tokens and the queue Q n (Q t ) is determined by the difference.

8. The cloud-based continuing education information management system according to claim 7, characterized in that: Based on the queue Q w (Q t ) and the token generation rate μ c Returns the waiting time T for course selection request w , 9. The cloud-based continuing education information management system according to claim 8, characterized in that: Obtain the course selection request processing data of honeycomb unit c during the flood peak breaking processing period, and update the historical request parameters of honeycomb unit c. The historical request parameters include: historical request mean and historical request standard deviation; The ratio of the number and value of course selection requests processed by honeycomb unit c in each peak breaking processing period in the historical time period to the duration and value of each peak breaking processing period in the historical time period is taken as the historical request mean. The standard deviation of the number of course selection requests processed by honeycomb unit c in each flood peak breaking processing time period within the historical time period and the historical request mean is taken as the historical request standard deviation σ.

10. The cloud platform-based continuing education information management system according to claim 9, characterized in that: Adaptively adjust the processing capacity C of honeycomb unit c based on historical request parameters c ,include: Among them, ∈ represents the minimum processing capacity of the honeycomb unit c, α and β represent capacity weight coefficients, and the sum of α and β is 1, and neither is 0.

Images