Content prefetching method based on node service flow portrait

Abstract

The invention relates to the technical field of content distribution networks, and discloses a content prefetching method based on a node service flow portrait, computer equipment, a computer readable storage medium and a computer program product in order to solve the problem that the existing prefetching technology is low in content prefetching efficiency. The method comprises the following steps: constructing a node service traffic portrait; based on the node service flow portrait, a dynamic demand model of the node service content is established, and the dynamic demand model of the node service content is used for predicting a dynamic demand vector of the node to the service content at any moment in the next access period; generating a node service content prefetching list according to the dynamic demand vector of the node at each moment to the service content in the next access period; and performing content prefetching according to the node service content prefetching list. By adopting the method, the demand prediction accuracy and the content prefetching efficiency of the service content of the node can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of content distribution networks, and more specifically, to a content pre-fetching method based on node business traffic profile, a computer device, a computer-readable storage medium, and a computer program product. Background Art

[0002] Focusing on the demand for high-traffic business distribution at nodes, and addressing the difficult issues such as high concurrent pressure at central nodes, low forwarding efficiency at backbone nodes, and low transmission rate at edge nodes, the content delivery network (CDN) is currently mainly used to cache business content on cache nodes (also called CDN nodes) close to users, so that node users can access content nearby, and introduce content pre-fetching based on CDN cache technology, which alleviates the traffic pressure on nodes to a certain extent. However, the dynamic changes in node business needs and the uncertainty of node business traffic result in poor cache pre-fetching effects, making it difficult to fundamentally reduce node business traffic pressure and ensure node user business experience.

[0003] Existing caching technology (i.e. CDN caching technology) mainly caches business content to cache service devices (i.e. cache nodes) that are close to the node, and forwards business requests to the nearest cache node for response through network request redirection, thereby shortening business access time and alleviating node access pressure. Existing pre-fetching technology usually pre-fetches the content of specific URLs manually or in a scheduled manner based on the user's interest type or the popularity of cached content to ensure the specific access needs of node users.

[0004] However, the existing caching technology and prefetching technology still have the following defects: the existing caching technology cannot solve the network pressure caused by new user requests or new business content access; the existing prefetching technology usually only considers the interest type of node users or the popularity of cached content, ignores the correlation between network topology and resource distribution, ignores network bandwidth or network business traffic, and is a resource acquisition method that artificially predicts future user access needs. It has problems such as low prefetching resource efficiency, unreasonable prefetching time period, and poor matching of prefetched content. Summary of the invention

[0005] In order to solve the above problems, the present invention provides a content pre-fetching method, computer equipment, computer-readable storage medium and computer program product based on node business traffic portrait, which will improve the accuracy of demand prediction for node business content and content pre-fetching efficiency.

[0006] To achieve the above object, according to a first aspect of the present invention, a content pre-fetching method based on a node service traffic profile is provided, the method comprising:

[0007] Build a node business traffic profile;

[0008] Based on the node service traffic profile, a dynamic demand model for node service content is established. The dynamic demand model for node service content is used to predict the dynamic demand vector of the node for service content at any time in the next access cycle.

[0009] Generate a node service content pre-fetch list based on the dynamic demand vector of the node for service content at each time in the next access cycle;

[0010] Content is pre-fetched according to the node service content pre-fetch list.

[0011] Furthermore, a node business traffic profile is constructed, including collecting node network traffic and extracting node business information, where the node business information includes traffic source IP address, destination IP address, source port number, destination port number and transport layer protocol type; based on the node business information, access statistics are performed on the node business to obtain the business content and node business rules that the node users are most interested in; based on the business content and node business rules that the node users are most interested in, a node business traffic profile is constructed.

[0012] Furthermore, based on the node business traffic portrait, a dynamic demand model for the node business content is established, including determining the node's demand for business content based on the node business traffic distribution and duration in the node business traffic portrait; using the forgetting curve to calculate the weight of the historical access cycle; using an incremental method to describe the process of the node's demand for business content changing with the node business traffic; and establishing a dynamic demand model for the node business content based on the weight of the historical access cycle and the process of the node's demand for business content changing with the node business traffic.

[0013] Furthermore, based on the distribution and duration of the node business traffic in the node business traffic portrait, the node's demand for business content is determined, including if a certain business traffic of the node does not reappear for a long time after appearing or appears intermittently with a long period, it means that the node's demand for business content is lower than the demand threshold or the demand is decreasing and the rate of decrease is getting faster; if a certain business traffic of the node lasts for a duration greater than the set duration threshold, it means that the node's demand for business content is higher than the demand threshold.

[0014] Furthermore, based on the dynamic demand vectors of the nodes for the business content at each moment in the next access cycle, a node business content pre-fetch list is generated, including defining the priority of the business content based on the dynamic demand vectors of the nodes for the business content at each moment in the next access cycle; and adding each business content to the initial pre-fetch list according to the size of the priority to generate a node business content pre-fetch list.

[0015] Furthermore, according to the priority level, each business content is added to the initial pre-fetch list to generate a node business content pre-fetch list, including calculating the maximum value of the node's demand for business content to obtain the highest priority; adding the highest priority content to the initial pre-fetch list; when the number of a certain business content in the initial pre-fetch list reaches the upper limit, the demand level of the node represented by the dynamic demand vector for the business content is set to 0, and the priority of the remaining business content is calculated; repeat the above process until the demand level of the node represented by the dynamic demand vector for each business content is 0, or the length of the initial pre-fetch list reaches the maximum, and a node business content pre-fetch list is generated.

[0016] Furthermore, content is pre-fetched according to the node business content pre-fetch list, including defining the demand value of the content by predicting the frequency of the node business content being accessed in the next cycle; if the storage space of the node is sufficient, the content is pre-fetched in sequence according to the node business content pre-fetch list; if the storage space of the node is insufficient, then when the demand value of the cached content is less than the demand value of the content to be pre-fetched, the cached content is deleted and the content to be pre-fetched is pre-fetched from the node business content pre-fetch list, otherwise, the pre-fetching of the content to be pre-fetched is not performed; when performing the pre-fetching operation, if the business content to be pre-fetched exists in multiple nodes, a comprehensive judgment is made based on the number of node business requests for the business content to be pre-fetched, the node storage capacity and the node traffic situation, and the node with the largest pre-fetch coefficient is selected to perform the pre-fetching operation.

[0017] According to a second aspect of the present invention, there is also provided a computer device, which includes a memory, a processor and a computer program stored in the memory, and the processor executes the computer program to implement the steps of any one of the above methods.

[0018] According to a third aspect of the present invention, there is also provided a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0019] According to a fourth aspect of the present invention, there is also provided a computer program product, comprising a computer program, which implements the steps of any of the above methods when executed by a processor.

[0020] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0021] (1) The present invention provides a content pre-fetching method based on node business traffic portrait, by constructing a node business traffic portrait; based on the node business traffic portrait, a dynamic demand model of node business content is established, the model can predict the dynamic demand vector of the node for business content at any time in the next access cycle, and generate a node business content pre-fetching list according to the dynamic demand vector of the node for business content at each time in the next access cycle, and perform content pre-fetching according to the node business content pre-fetching list, thereby realizing automatic pre-fetching of node business content. Compared with the existing pre-fetching technology that mainly adopts manual or timed methods to perform a priori pre-fetching of the content of a specific URL, the present invention performs automatic content pre-fetching based on the node business content pre-fetching list, avoiding the waste of human resources, and the pre-fetching process is more efficient, thereby achieving the purpose of improving the efficiency of pre-fetching content.

[0022] (2) The present invention provides a content pre-fetching method based on a node business traffic portrait, which collects node network traffic, performs accurate statistical analysis on the business information covered by the node, and constructs an accurate node business traffic portrait.

[0023] (3) The present invention provides a content pre-fetching method based on node business traffic portrait, which maps the node's demand for business content through the node business traffic portrait, calculates the weight of the historical access cycle using the forgetting curve, and uses an incremental method to describe the change process of the node's demand for business content as the node's business traffic changes, thereby establishing a dynamic demand model for node business content. Compared with the existing prediction and recommendation system that treats historical access cycle information equally and lacks quantitative analysis of timeliness, the present invention predicts the demand for node business content dynamically over time, and has a higher accuracy rate for predicting the demand for node business content.

[0024] (4) The present invention provides a content pre-fetching method based on node business traffic portrait. By pre-fetching business content in a timely manner according to priority during the node traffic valley period, combined with the demand prediction results for the node's business content, the present invention achieves the effect of "peak shaving and valley flattening" by pre-fetching content in advance before the user actually requests the content, saving distribution overhead, reducing service delay, and eliminating the decline in service quality caused by dynamic changes in the network. Compared with existing pre-fetching technologies that often ignore the consideration of node network bandwidth, node traffic load and other conditions, especially the large fluctuations in the edge node network, which can easily cause network congestion and seriously affect network quality and user experience, the present invention maximizes the network resource utilization efficiency and business content pre-fetching efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 A flowchart of a content pre-fetching method based on a node service traffic profile provided in an embodiment of the present application;

[0027] Figure 2 A flowchart of the steps of establishing a dynamic demand model for node service content provided in an embodiment of the present application;

[0028] Figure 3 A schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0030] The terms "first", "second", "third", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0031] like Figure 1 As shown, a content pre-fetching method based on node business traffic portrait is provided, which can be executed by a node cache server or by other terminal devices that communicate with the node through a network. Among them, the terminal device can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, etc. The node cache server can be an independent server, or a server cluster composed of multiple servers can be used to implement it. Taking the application of this method to a node cache server as an example, it includes the following steps:

[0032] Step 101, obtain the node business traffic profile.

[0033] Exemplarily, the node cache server collects node network traffic (i.e., the overall traffic of the node network) on a daily basis, and extracts node business information including identifiers such as the traffic source IP address, destination IP address, source port number, destination port number, and transport layer protocol type; performs access statistics on the node business according to the access cycle to obtain the business content that the node users are most interested in, and summarizes the node business rules to construct a node business traffic portrait within the operating time. The node business traffic is the traffic generated by a specific business of the node.

[0034] Step 102: Based on the node business traffic profile, a dynamic demand model of the node business content is established.

[0035] The dynamic demand model of the node service content is used to predict the dynamic demand vector of the node for the service content at any time in the next access cycle.

[0036] For example, the node cache server performs dynamic demand modeling of the node service content based on the node service traffic profile. By analyzing factors such as the node service traffic distribution and duration in the node service traffic profile, the node's demand for the service and the change process of the demand are mapped.

[0037] Considering the demand for node services, we need to conduct a superposition analysis of multiple cycles, calculate the weights of historical access cycles based on the forgetting curve, and use an incremental method to describe the process of how the node's demand for service content changes with traffic, and predict the dynamic demand vector of the node for service content at a certain moment in the next access cycle.

[0038] Step 103: Generate a node service content pre-fetch list according to the dynamic demand vector of the node for service content at each time in the next access cycle.

[0039] Exemplarily, the node cache server defines the priority of the service content according to the dynamic demand vector of the node for the service content at each moment in the next access cycle, and adds each service content to the initial pre-fetch list according to the priority to obtain the final node service content pre-fetch list. In the process of adding each service content to the initial pre-fetch list, in order to prevent overloading of service content with high demand, the calculation method of the service content priority will also be changed according to the proportion of added service content.

[0040] Step 104: pre-fetch content from the node service content pre-fetch list.

[0041] Exemplarily, the node cache server defines the demand value of the content by predicting the frequency of access to the node service content in the next cycle, and performs a pre-fetching operation by comparing the demand value of the cached content with the demand value of the content to be pre-fetched.

[0042] If the node has sufficient storage space, the content is pre-fetched in sequence according to the node service content pre-fetch list; if the node has insufficient storage space, the cached content is deleted and the content to be pre-fetched is pre-fetched from the node service content pre-fetch list when the demand value of the cached content is less than the demand value of the content to be pre-fetched; otherwise, the content to be pre-fetched is not pre-fetched; in addition, when performing the pre-fetch operation, if the service content to be pre-fetched exists in multiple nodes, a comprehensive judgment is made based on the number of node service requests for the service content to be pre-fetched, the node storage capacity and the node traffic situation, and the node with the largest pre-fetch coefficient is selected to perform the pre-fetch operation. The pre-fetch coefficient is a measure of the pre-fetch operation performed by multiple nodes. The larger the pre-fetch coefficient, the higher the success rate of the pre-fetch operation.

[0043] The pre-fetching operation needs to be based on the node network traffic situation. The node traffic trough is found based on the node business traffic profile. When the traffic is small or low, it is executed according to the node business content pre-fetch list.

[0044] In the above-mentioned content prefetching method based on node business traffic portrait, a node business traffic portrait is constructed; based on the node business traffic portrait, a dynamic demand model for node business content is established, and the model can predict the dynamic demand vector of the node for business content at any time in the next access cycle, and generate a node business content prefetching list according to the dynamic demand vector of the node for business content at each time in the next access cycle, and prefetch the content on the list, thereby realizing automatic prefetching of node business content and achieving the purpose of improving the efficiency of prefetching content.

[0045] In one embodiment, Figure 2 As shown, the above step 102, based on the node service traffic profile, establishes a dynamic demand model for the node service content, including the following steps:

[0046] Step 201, based on the node service traffic distribution and duration in the node service traffic profile, determine the node's demand for service content;

[0047] Step 202, using the forgetting curve to calculate the weight of the historical access cycle;

[0048] Step 203, using an incremental method to describe the process in which the node's demand for service content changes with the node's service traffic;

[0049] Step 204 : Based on the weight of the historical access cycle and the change process of the node's demand for service content along with the node's service traffic, a dynamic demand model for the node's service content is established.

[0050] In this embodiment, the demand prediction of the node's business content by the dynamic demand model of the node's business content changes dynamically over time, which can improve the accuracy of the demand prediction of the node's business content.

[0051] In one embodiment, a content pre-fetching method based on a node service traffic profile includes the following steps:

[0052] (1) Build a node business traffic profile.

[0053] Node network traffic is collected on a daily basis, access statistics are performed on node services according to access cycles, node business rules are summarized, and a node business traffic profile is constructed.

[0054] (2) Modeling node business dynamic demand based on node business traffic profile

[0055] The node's service demand is mapped through the node business traffic portrait, mainly considering the following factors: 1) If a certain business traffic of the node does not appear again for a long time after it appears, or is intermittent and has a long period, it means that the demand for the business content is low (that is, lower than a certain set demand threshold) or the demand is decreasing rapidly; if a certain business traffic of the node is large for a long time, or the traffic is small but lasts for a long time (that is, the duration is greater than a certain set duration threshold), it means that the demand for the business content is high (that is, higher than the set demand threshold); 2) The node's demand for business content is the superposition of the influence of multiple periodic access situations; 3) The access duration of the node to the business content is a continuous downward process.

[0056] The forgetting curve is used to calculate the weight of the historical access cycle, and the incremental method is used to describe the change process of the node's demand for business content with the traffic. The dynamic demand vector of the node for business content at a certain moment in the next access cycle is predicted. The demand vector reflects the overall demand situation of the node's business.

[0057] (3) Generate a node service content pre-fetch list

[0058] 1) Define the priority of the content according to the node's demand for the service content: calculate the priority of the content with the highest demand in the node demand vector;

[0059] 2) Add the content to the pre-fetch list according to the priority of the content: Add the content with the highest priority to the pre-fetch list according to the priority of the content. In the process of adding content to the pre-fetch list, in order to prevent overload of content with high demand, the calculation method of content priority will be changed according to the proportion of added content. For example, when the number of a certain business content in the list reaches the upper limit, the demand level of the corresponding business content in the demand vector used to calculate the priority is set to 0, and then the priority calculation is performed on the remaining business content;

[0060] 3) Repeat the above process until the demand levels of the service contents in the demand vector used to calculate the priority are all 0, or the length of the pre-fetch list reaches the maximum, thereby forming a final node service content pre-fetch list.

[0061] (4) Pre-fetching node business content

[0062] The node cache server performs pre-fetching operations based on the node network traffic and the node service content pre-fetching list when the traffic is small or low. The detailed process is as follows:

[0063] 1) The node cache server counts the node user access records every period to obtain the demand level of each business content, and calculates the dynamic demand vector of the node for the business content;

[0064] 2) The node cache server calculates the priority of the content according to the dynamic demand vector and obtains a node service content pre-fetch list according to the priority;

[0065] 3) The node cache server performs the pre-fetching operation according to the node business content pre-fetching list, and defines the demand value of the content by predicting the frequency of node content access in the next cycle. During the pre-fetching process, if the node has sufficient storage space, the content is pre-fetched in sequence according to the node business content pre-fetching list; if the node has insufficient storage space, but the demand value of the cached content is less than the demand value of the content to be pre-fetched, the cached content is deleted, and the content to be pre-fetched is pre-fetched from the node business content pre-fetching list; otherwise, the pre-fetching operation is stopped;

[0066] 4) When the node cache server performs a pre-fetch operation, if the intended pre-fetched business content exists in multiple nodes, a comprehensive judgment is made based on the number of node business requests for the intended pre-fetched business content, the node storage capacity, and the node traffic situation, and the intended pre-fetched business node with the largest (i.e. optimal) pre-fetch coefficient is selected to perform pre-fetching of the business content. The pre-fetched content may come from multiple different nodes, and the pre-fetch coefficient is a measure of the pre-fetch operation performed by multiple nodes. The larger the pre-fetch coefficient, the higher the success rate of the pre-fetch operation.

[0067] After testing, the content pre-fetching method based on the node business traffic portrait provided in this embodiment has an accuracy rate of predicting the demand for node business content of no less than 80%, and the node business content is automatically pre-fetched to avoid manual operation, and the content pre-fetching efficiency is improved by no less than 50%.

[0068] The present application also provides a computer device, whose internal structure diagram can be as follows: Figure 3As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a content pre-fetching method based on a node business traffic profile is implemented. The display unit of the computer device is used to form a visually visible image, and can be a display screen, a projection device or a virtual reality imaging device. The input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse, etc.

[0069] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0070] like Figure 3 As shown, the present application also provides a computer device, which includes a memory, a processor, and a computer program stored in the memory, and the processor executes the computer program to implement the steps in the above-mentioned method embodiments.

[0071] The present application also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a micro drive, and a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0072] The present application also provides a computer program product, including a computer program, which implements the steps in the above-mentioned method embodiments when executed by a processor.

[0073] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0074] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0075] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the specification and practicing the disclosure here, those skilled in the art will easily think of the implementation scheme of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field not recorded in the present disclosure. The description and examples are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

[0076] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0077] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A content pre-fetching method based on node service traffic profile, characterized in that: include: Build a node business traffic profile; Based on the node service traffic portrait, a dynamic demand model of the node service content is established, wherein the dynamic demand model of the node service content is used to predict the dynamic demand vector of the node for the service content at any time in the next access cycle; Generate a node service content pre-fetch list based on the dynamic demand vector of the node for service content at each time in the next access cycle; Content pre-fetching is performed according to the node service content pre-fetching list.

2. The method according to claim 1, characterized in that The step of constructing a node service traffic profile includes: Collect node network traffic and extract node service information, wherein the node service information includes traffic source IP address, sink IP address, source port number, sink port number and transport layer protocol type; According to the node service information, the access volume of the node service is counted to obtain the service content and node service rules that the node users are most interested in; Based on the business content that the node users are most interested in and the business rules of the node, a node business traffic profile is constructed.

3. The method according to claim 1, characterized in that The step of establishing a dynamic demand model for node service content based on the node service traffic profile includes: Determine the node's demand for service content based on the node's service traffic distribution and duration in the node's service traffic profile; Use the forgetting curve to calculate the weight of historical access cycles; An incremental approach is used to describe the process of changes in the node's demand for service content as the node's service traffic changes; Based on the weight of the historical access cycle and the change process of the node's demand for service content along with the node's service flow, a dynamic demand model for node service content is established.

4. The method according to claim 1, characterized in that The determining of the degree of demand of the node for the service content based on the node service traffic distribution and duration in the node service traffic profile includes: If a service flow of a node does not appear again for a long time after it appears, or appears intermittently with a long period, it means that the node's demand for the service content is lower than the demand threshold or the demand is decreasing and the decreasing speed is getting faster; If a service flow of a node lasts longer than a set duration threshold, it means that the node's demand for the service content is higher than the demand threshold.

5. The method according to claim 3, characterized in that The generating of a node service content pre-fetch list according to the dynamic demand vector of the node for service content at each time point in the next access cycle includes: Define the priority of business content according to the dynamic demand vector of business content at each node in the next access cycle; According to the priority level, each service content is added to the initial pre-fetch list to generate a node service content pre-fetch list.

6. The method according to claim 5, characterized in that The method of adding each service content to the initial pre-fetch list according to the priority level to generate a node service content pre-fetch list includes: The maximum value of the demand degree of the computing node for the business content is given the highest priority; Add the highest priority content to the initial prefetch list; When the number of a certain service content in the initial pre-fetch list reaches the upper limit, the demand degree of the node represented by the dynamic demand vector for the said service content is set to 0, and the priority of the remaining service content is calculated; The above process is repeated until the demand degree of each service content of the node represented by the dynamic demand vector is 0, or the length of the initial pre-fetch list reaches the maximum, and the node service content pre-fetch list is generated.

7. The method according to claim 5, characterized in that The pre-fetching of content according to the node service content pre-fetching list includes: Define the demand value of content by predicting the frequency of access to the node business content in the next cycle; If the node has sufficient storage space, the content is pre-fetched in sequence according to the node service content pre-fetch list; If the storage space of the node is insufficient, then if the demand value of the cached content is less than the demand value of the content to be pre-fetched, the cached content is deleted and the content to be pre-fetched is pre-fetched from the node service content pre-fetching list; otherwise, the pre-fetching of the content to be pre-fetched is not performed; When performing a prefetch operation, if the business content to be prefetched exists on multiple nodes, a comprehensive judgment is made based on the number of node business requests for the business content to be prefetched, the node storage capacity, and the node traffic situation, and the node with the largest prefetch coefficient is selected to perform the prefetch operation.

8. A computer device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the steps of the method according to any one of claims 1 to 7 when being executed by a processor.

Images