CN110858912A - Streaming media caching method and system, caching policy server and streaming service node - Google Patents
Streaming media caching method and system, caching policy server and streaming service node Download PDFInfo
- Publication number
- CN110858912A CN110858912A CN201810957204.2A CN201810957204A CN110858912A CN 110858912 A CN110858912 A CN 110858912A CN 201810957204 A CN201810957204 A CN 201810957204A CN 110858912 A CN110858912 A CN 110858912A
- Authority
- CN
- China
- Prior art keywords
- video
- hotspot
- cache
- streaming service
- service node
- 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.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/231—Content storage operation, e.g. caching movies for short term storage, replicating data over plural servers, prioritizing data for deletion
- H04N21/23106—Content storage operation, e.g. caching movies for short term storage, replicating data over plural servers, prioritizing data for deletion involving caching operations
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/235—Processing of additional data, e.g. scrambling of additional data or processing content descriptors
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/258—Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
- H04N21/25866—Management of end-user data
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/433—Content storage operation, e.g. storage operation in response to a pause request, caching operations
- H04N21/4331—Caching operations, e.g. of an advertisement for later insertion during playback
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/435—Processing of additional data, e.g. decrypting of additional data, reconstructing software from modules extracted from the transport stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/4508—Management of client data or end-user data
Abstract
The invention discloses a streaming media caching method and system, a caching strategy server and a streaming service node. The method comprises the following steps: determining the click probability of the video in a future preset time period according to historical correlation data of each streaming service node video; determining a hotspot attribute of the video according to the click probability of the video, wherein the hotspot attribute of the video is whether the video is a hotspot video or not; and issuing a replacement instruction to each stream service node according to the hotspot attributes of the video so as to facilitate cache replacement of each stream service node. According to the method and the device, the hot videos are determined based on the relevance prediction, so that the CDN streaming media cache can replace hot resources in time, and the cache hit rate is improved.
Description
Technical Field
The present invention relates to the field of data communication, and in particular, to a streaming media caching method and system, a caching policy server, and a streaming service node.
Background
The important reason why the CDN (Content Delivery Network) can accelerate video is to store hot video Content on a streaming media cache server in the CDN Network that a user can access nearby, and the streaming media cache server directly responds to a video playing request of the user.
The caching method applied by the streaming media caching server is an important factor for improving the hit rate of the streaming server and further improving the overall performance of the CDN, and the good caching method can ensure the video playing quality of a user.
Disclosure of Invention
The applicant found that: the cache method of the related technology mostly uses the click rate of the video in the historical time as an index for predicting the video heat, carries out the ordering of the video heat by counting the click number of the video in a certain historical time, and deletes the video content after the ordering of the heat from the streaming service cache node at regular time.
The related art caching method only considers the number of clicks of the video in a certain historical time, and ignores various relevant factors influencing the video click rate in practice, such as the correlation between the video click rate and the search rate, the correlation between the video click rate and the collection rate, and the correlation between the video click rate and the display time (for example, the click rate of the video content in spring festival is higher than usual). Because the number of clicks of a video at a certain time point may be affected by other related factors, a hot video predicted by a cache method of the related art has a large deviation from the actual value, and the cache hit rate is easily not up to the standard, thereby affecting the service quality of the CDN.
In view of at least one of the above technical problems, the present invention provides a streaming media caching method and system, a caching policy server, and a streaming service node, which improve the cache hit rate based on a correlation prediction model.
According to an aspect of the present invention, there is provided a streaming media caching method, including:
determining the click probability of the video in a future preset time period according to historical correlation data of each streaming service node video;
determining a hotspot attribute of the video according to the click probability of the video, wherein the hotspot attribute of the video is whether the video is a hotspot video or not;
and issuing a replacement instruction to each stream service node according to the hotspot attributes of the video so as to facilitate cache replacement of each stream service node.
In some embodiments of the invention, the historical relevance data includes at least two of an amount of video searches, an amount of video clicks, an amount of video collections, and a video presentation time.
In some embodiments of the present invention, the determining, according to the historical relevance data of the videos of the respective streaming service nodes, the click probability of the video in a predetermined period of time in the future includes:
and counting index values related to the video click rate in a plurality of time dimensions, and determining the click probability of the video in a future preset time period according to the index values.
In some embodiments of the invention, the indicator value comprises at least one of a click probability, a collection probability, and a search probability of the video in a plurality of time dimensions.
In some embodiments of the invention, the plurality of time dimensions comprises a plurality of time periods corresponding to predetermined time periods.
In some embodiments of the invention, the predetermined period of time is one day;
the multiple time dimensions include yesterday, last week of today, last month of today, last year of today.
In some embodiments of the present invention, the determining the click probability of the video in a predetermined time period in the future according to the index value comprises:
and inputting the index value into a preset relevance prediction model, and determining the click probability of the video in a future preset time period.
In some embodiments of the present invention, the determining the hotspot attribute of the video according to the click probability of the video includes:
for each video, judging whether the click probability of the video is greater than or equal to a preset threshold value;
if the click probability of the video is larger than or equal to a preset threshold value, judging that the video is a hot video;
and if the click probability of the video is smaller than a preset threshold value, judging that the video is a non-hotspot video.
According to another aspect of the present invention, there is provided a streaming media caching method, including:
receiving a replacement instruction issued by a cache policy server, wherein the cache policy server determines the click probability of a video in a future preset time period according to historical correlation data of the video of each streaming service node, determines the hotspot attribute of the video according to the click probability of the video, wherein the hotspot attribute of the video is whether the video is a hotspot video, and issues the replacement instruction to each streaming service node according to the hotspot attribute of the video;
and replacing the current stream service cache according to the replacement instruction.
In some embodiments of the present invention, the replacing, according to the replacement instruction, the current streaming service cache includes:
judging whether the video exists in a current streaming service cache or not under the condition that the video is a hot video;
maintaining the current streaming service cache unchanged under the condition that the hotspot video exists in the current streaming service cache;
and under the condition that the hotspot video does not exist in the current streaming service cache, injecting the hotspot video into the current streaming service cache.
In some embodiments of the present invention, the replacing, according to the replacement instruction, the current streaming service cache includes:
judging whether the video exists in a current streaming service cache or not under the condition that the video is a non-hotspot video;
deleting the non-hotspot video from the current streaming service cache under the condition that the non-hotspot video exists in the current streaming service cache;
and maintaining the current streaming service cache unchanged under the condition that the non-hotspot video does not exist in the current streaming service cache.
According to another aspect of the present invention, there is provided a cache policy server, including:
the click probability determining module is used for determining the click probability of the video in a future preset time period according to the historical correlation data of the videos of all the streaming service nodes;
the hotspot video determining module is used for determining hotspot attributes of the video according to the click probability of the video, wherein the hotspot attributes of the video are whether the video is a hotspot video or not;
and the replacement instruction issuing module is used for issuing a replacement instruction to each stream service node according to the hotspot attributes of the video so as to facilitate cache replacement of each stream service node.
In some embodiments of the present invention, the caching policy server is configured to perform operations for implementing the streaming media caching method according to any one of the above embodiments.
According to another aspect of the present invention, there is provided a cache policy server, including but not limited to:
a server memory to store instructions;
and the server processor is used for executing the instructions to enable the caching policy server to execute the operation of implementing the streaming media caching method according to any one of the embodiments.
According to another aspect of the present invention, there is provided a streaming service node, including:
the replacement instruction receiving module is used for receiving a replacement instruction issued by the cache policy server, wherein the cache policy server determines the click probability of a video in a future preset time period according to historical correlation data of videos of each streaming service node, determines the hotspot attribute of the video according to the click probability of the video, wherein the hotspot attribute of the video is whether the video is a hotspot video, and issues the replacement instruction to each streaming service node according to the hotspot attribute of the video;
and the cache replacement module is used for replacing the current stream service cache according to the replacement instruction.
In some embodiments of the present invention, the streaming service node is configured to perform an operation for implementing the streaming media caching method according to any one of the above embodiments.
According to another aspect of the present invention, there is provided a streaming service node, including but not limited to:
a node instruction memory for storing instructions;
a node instruction processor, configured to execute the instructions, so that the caching policy server performs operations of implementing the streaming media caching method according to any one of the above embodiments.
According to another aspect of the present invention, there is provided a streaming media caching system, including the caching policy server according to any of the above embodiments, and the streaming service node according to any of the above embodiments.
According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions, and when the instructions are executed by a processor, the method for caching streaming media according to any one of the above embodiments is implemented.
According to the method and the device, the hot videos are determined based on the relevance prediction, so that the CDN streaming media cache can timely store hot resources, non-hot resources are replaced, and the cache hit rate is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic diagram of some embodiments of a streaming media caching method according to the present invention.
Fig. 2 is a schematic diagram of a cache policy server according to some embodiments of the present invention.
FIG. 3 is a diagram illustrating another embodiment of a cache policy server according to the present invention.
Fig. 4 is a schematic diagram of another embodiment of a streaming media caching method according to the present invention.
Fig. 5 is a schematic diagram of some embodiments of a streaming service node.
Fig. 6 is a schematic diagram of another embodiment of a streaming service node of the present invention.
Fig. 7 is a diagram of a streaming media caching system according to some embodiments of the present invention.
Fig. 8 is a schematic diagram of a streaming media caching method according to still another embodiment of the invention.
Fig. 9 is a schematic diagram of a streaming media caching method according to still another embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The relative arrangement of the components and steps, the numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.
Meanwhile, it should be understood that the sizes of the respective portions shown in the drawings are not drawn in an actual proportional relationship for the convenience of description.
Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.
In all examples shown and discussed herein, any particular value should be construed as merely illustrative, and not limiting. Thus, other examples of the exemplary embodiments may have different values.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.
The applicant found that: the related technology determines that the predicted click rate of the video in the future 24 hours is mainly according to historical playing data, but actual correlation factors influencing the click behavior of a user are ignored, so that the predicted value of the click rate of the video is inaccurate; meanwhile, when the hot spot prediction deviation is large, the CDN cache is inaccurate, the hit rate of the CDN stream service cache is low, cache resource waste is caused, the server pressure is too large, and the user experience is poor.
In view of at least one of the above technical problems in the related art, the present invention provides a CDN network streaming media caching method and system based on relevance prediction, which are described in detail below with reference to an embodiment.
Fig. 1 is a schematic diagram of some embodiments of a streaming media caching method according to the present invention. Preferably, this embodiment may be performed by the cache policy server of the present invention. The method comprises the following steps:
and step 11, determining the click probability of the video in a future preset time period according to the historical correlation data of the videos of the streaming service nodes.
In some embodiments of the invention, the historical relevance data includes at least two of an amount of video searches, an amount of video clicks, an amount of video collections, and a video presentation time.
In some embodiments of the present invention, step 11 may comprise: and counting index values related to the video click rate in a plurality of time dimensions, and determining the click probability of the video in a future preset time period according to the index values.
In some embodiments of the present invention, the step of determining the click probability of the video in a predetermined time period in the future according to the index value may include: and inputting the index value into a preset relevance prediction model, and determining the click probability of the video in a future preset time period.
In some embodiments of the invention, the indicator value comprises at least one of a click probability, a collection probability, and a search probability of the video in a plurality of time dimensions.
In some embodiments of the invention, the plurality of time dimensions comprises a plurality of time periods corresponding to predetermined time periods.
In some embodiments of the invention, the predetermined period of time is one day; accordingly, the multiple time dimensions may include yesterday, last week of today, last month of today, last year of today.
In some embodiments of the present invention, step 11 may comprise:
step 111, determining a correlation prediction model.
In some embodiments of the present invention, step 111 may comprise:
first, since the showing time of the video, i.e. the relevance of the video to a certain day of a year (for example, the popularity of the video related to spring and late during the spring festival becomes high), will affect the popularity of the video, the statistical time is first divided into four time dimensions: yesterday, the last week today, the last month today and the last year today, so that the relevance of the video and the video display time is met.
Secondly, counting the click probability, the collection probability and the search probability of the video in the dimension from the four time dimensions respectively to obtain 12 index values of the video. Establishing a probability mathematical model shown as formula (1), and predicting a click probability value P of a video 24 hours in the future:
in the formula (1), Pk(k is 1,2, …,12) is the value of the k-th index of the video, WkThe weight occupied by each index.
In the formula (1), the weight value W of each indexkDepending on the degree of influence of the index on the result, the considerations adopted in the modelThe more stable the value of a certain index of a plurality of videos is, the smaller the utility value of the index is, that is, the smaller the weight value is, and conversely, the larger the weight value is. Thereby obtaining the weight W occupied by each index as shown in the formula (2)k。
In the formula (2), dkIs a measure of the degree of disorder of the data information, as shown in equation (3):
in the formula (3), dkThe calculation model of (2) involves n video samples, m indexes, pikIs the probability value of the ith video under the kth index.
Step 112, counting index values related to the video click rate in a plurality of time dimensions, inputting the index values into a predetermined correlation prediction model shown in formula (1), and determining the click probability of the video in a predetermined time period (for example, 24 hours in the future).
And step 12, determining the hotspot attribute of the video according to the click probability of the video, wherein the hotspot attribute of the video is whether the video is a hotspot video or not.
In some embodiments of the present invention, step 12 may comprise:
in step 121, for each video, it is determined whether the click probability P of the video is greater than or equal to a predetermined threshold α.
And step 122, if the click probability P of the video is greater than or equal to a preset threshold value α, determining that the video is a hot video.
And step 123, if the click probability P of the video is smaller than a preset threshold α, determining that the video is a non-hotspot video.
And step 13, issuing a replacement instruction to each stream service node according to the hotspot attributes of the video so as to facilitate each stream service node to perform cache replacement.
Based on the streaming media caching method provided by the embodiment of the invention, most historical relevant factors influencing the video click rate are described by a unified mathematical model, and a relevance prediction model is designed, so that the predicted video click rate in the future 24 hours can approach the real situation infinitely, thereby enabling the CDN streaming media cache to store hot resources in time, replacing non-hot resources, improving the cache hit rate, and finally solving the problems of low hit rate of CDN cache hot points and poor user experience.
Fig. 2 is a schematic diagram of a cache policy server according to some embodiments of the present invention. As shown in fig. 2, the cache policy server of the present invention may include a click probability determining module 21, a hotspot video determining module 22, and a replacement instruction issuing module 23, where:
and the click probability determining module 21 is configured to determine the click probability of the video in a future predetermined time period according to the historical correlation data of the videos of the streaming service nodes.
The hotspot video determining module 22 is configured to determine a hotspot attribute of the video according to the click probability of the video, where the hotspot attribute of the video is whether the video is a hotspot video.
And the replacement instruction issuing module 23 is configured to issue a replacement instruction to each stream service node according to the hotspot attribute of the video, so that each stream service node performs cache replacement.
In some embodiments of the present invention, the caching policy server is configured to perform operations for implementing the streaming media caching method according to any one of the embodiments (for example, the embodiment of fig. 1) described above.
FIG. 3 is a diagram illustrating another embodiment of a cache policy server according to the present invention. As shown in fig. 3, the cache policy server of the present invention may comprise a server memory 31 and a server processor 32, wherein:
a server memory 31 for storing instructions.
A server processor 32, configured to execute the instructions, so that the caching policy server performs operations of implementing the streaming media caching method according to any one of the embodiments (for example, the embodiment in fig. 1) described above.
Based on the cache strategy server provided by the embodiment of the invention, most of historical relevant factors influencing the video click rate are described by a unified mathematical model, and a relevance prediction model is designed, so that the predicted video click rate in the future 24 hours can approach the real situation infinitely, thereby enabling the CDN streaming media cache to store hot resources in time, replacing non-hot resources, improving the cache hit rate, and finally solving the problems of low cache hit rate and poor user experience of the CDN.
Fig. 4 is a schematic diagram of another embodiment of a streaming media caching method according to the present invention. Preferably, this embodiment can be performed by the streaming service node of the present invention. The method comprises the following steps:
And 42, replacing the current stream service cache according to the replacement instruction.
In some embodiments of the present invention, step 42 may comprise:
step 421, in case that the video is a hot video, determining whether the video exists in the current streaming service cache.
Step 422, under the condition that the hotspot video exists in the current streaming service cache, maintaining the current streaming service cache unchanged.
Step 423, injecting the hot video into the current streaming service cache when the hot video does not exist in the current streaming service cache.
Step 424, in case that the video is a non-hotspot video, determining whether the video exists in the current streaming service cache.
Step 425, deleting the non-hotspot video from the current streaming service cache in case the non-hotspot video exists in the current streaming service cache.
And 426, maintaining the current streaming service cache unchanged under the condition that the non-hotspot video does not exist in the current streaming service cache.
Based on the streaming media caching method provided by the embodiment of the invention, the video click rate predicted by the caching strategy server in the future 24 hours is adopted to approach the real situation infinitely, and the situation that the predicted value obtained by only adopting single historical data of a period of continuous time and the actual situation have larger errors is solved. According to the embodiment of the invention, the accuracy of cache video hotspot replacement is improved, the hit rate and the resource utilization rate of CDN stream service are effectively improved, and the use experience of a user is effectively improved.
Fig. 5 is a schematic diagram of some embodiments of a streaming service node. As shown in fig. 5, the streaming service node of the present invention may include a replacement instruction receiving module 51 and a cache replacement module 52, wherein:
the replacement instruction receiving module 51 is configured to receive a replacement instruction issued by a cache policy server, where the cache policy server determines, according to historical correlation data of videos of each streaming service node, a click probability of a video in a predetermined time period in the future, determines a hotspot attribute of the video according to the click probability of the video, where the hotspot attribute of the video is whether the video is a hotspot video, and issues the replacement instruction to each streaming service node according to the hotspot attribute of the video.
And a cache replacement module 52, configured to replace the current streaming service cache according to the replacement instruction.
In some embodiments of the present invention, the streaming service node is configured to perform operations for implementing the streaming media caching method according to any one of the embodiments (for example, the embodiment of fig. 4) described above.
In some embodiments of the present invention, the streaming service node may be implemented as an edge node of the CDN network.
Fig. 6 is a schematic diagram of another embodiment of a streaming service node of the present invention. As shown in fig. 6, the streaming service node of the present invention may include a node memory 61 and a node processor 62, wherein:
a node instruction memory 61 for storing instructions.
A node instruction processor 62, configured to execute the instructions, so that the caching policy server performs operations of implementing the streaming media caching method (for example, the embodiment of fig. 4) according to any one of the above embodiments.
Based on the streaming service node provided by the embodiment of the invention, the video click rate predicted by the cache policy server in the future 24 hours is adopted to approach the real situation infinitely, and the situation that the predicted value obtained by only adopting single historical data of a period of continuous time and the actual situation have larger errors frequently is solved. According to the embodiment of the invention, the accuracy of cache video hotspot replacement is improved, the hit rate and the resource utilization rate of CDN stream service are effectively improved, and the use experience of a user is effectively improved.
Fig. 7 is a diagram of a streaming media caching system according to some embodiments of the present invention. As shown in fig. 7, the streaming service node of the present invention may include a cache policy server 71 and at least one streaming service node 72, wherein:
the caching policy server 71 may be implemented as a caching policy server as described in any of the embodiments above (e.g., the embodiments of fig. 2 or fig. 3).
The streaming service node 72 may be implemented as the streaming service node according to any of the embodiments described above (e.g., the embodiment of fig. 5 or fig. 6).
Based on the streaming media cache system provided by the embodiment of the invention, the relevance prediction model is established to enable the predicted video click rate in the future 24 hours to approach the real situation infinitely, so that the situation that a predicted value obtained by only adopting single historical data of a period of continuous time and the actual situation have large errors frequently is solved. According to the embodiment of the invention, the accuracy of cache video hotspot replacement is improved, the hit rate and the resource utilization rate of CDN stream service are effectively improved, and the use experience of a user is effectively improved.
Fig. 8 is a schematic diagram of a streaming media caching method according to still another embodiment of the invention. Preferably, this embodiment can be performed by the streaming media caching system of the present invention. The method comprises the following steps:
In some embodiments of the present invention, step 81 may comprise: the cache policy server, according to historical relevance factors with the video in each streaming service node in the CDN network, for example: and predicting the click probability of the video in the future 24 hours according to historical data such as the search volume, the click volume, the collection volume, the video display time and the like, making a hotspot generation plan, and issuing a replacement instruction to the cache of the edge node according to whether the video is the hotspot video.
In some embodiments of the present invention, step 81 may comprise the steps of:
in step 811, the cache policy server counts the relevant records of each index of the factors relevant to the video click rate in the four time dimensions, inputs a relevance prediction model such as formula (1), and obtains the click probability prediction value P of the video 24 hours in the future.
Step 812, the cache policy server obtains a click probability prediction value P of each video in the future 24 hours, and judges whether P is greater than or equal to α, wherein α is a preset threshold value, if yes, the video is a hot video and is issued to an edge node (streaming service node) as a hot video passing instruction, and if not, the video is a non-hot video and is notified to the edge node as the non-hot video passing instruction.
And 82, the stream service node replaces the current stream service cache according to the replacement instruction sent by the cache policy server.
In some embodiments of the present invention, step 82 may comprise: when the CDN is in a non-access peak period in the early morning, the streaming service node starts cache replacement, adjusts hot content in each node cache, deletes non-hot videos for replacement, and injects hot videos which do not exist originally into the streaming service node, so that the streaming service node is guaranteed to have a high cache hit rate in the next 24 hours.
Fig. 9 is a schematic diagram of a streaming media caching method according to still another embodiment of the invention. Preferably, this embodiment can be performed by the streaming media caching system of the present invention. The method comprises the following steps:
And step 92, the cache policy server acquires a click probability predicted value P of each video in the future 24 hours, judges whether P is more than or equal to α, and α is a preset threshold value, if so, the video is a hot video, the video is issued to a streaming service node as a hot video through an instruction, then the streaming service node executes step 93, if not, the video is a non-hot video, the streaming service node is informed of the fact that the video is the non-hot video through the instruction, and then the streaming service node executes step 94.
And step 93, judging whether the video exists in the current streaming service cache by the streaming service node under the condition that the video is the hotspot video. In case the hotspot video exists in the current streaming service cache, performing step 95; otherwise, in case the hot video is not present in the current streaming service cache, step 96 is executed.
And step 93, judging whether the video exists in the current streaming service cache by the streaming service node under the condition that the video is a non-hotspot video. If the non-hotspot video exists in the current streaming service cache, executing step 97; otherwise, if the non-hotspot video does not exist in the current streaming service cache, step 95 is performed.
And step 96, the streaming service node injects the hotspot video into the current streaming service cache.
And step 97, the streaming service node deletes the non-hotspot video from the current streaming service cache.
Based on the streaming media caching method provided by the embodiment of the invention, most historical relevant factors influencing the video click rate are described by a unified mathematical model, and a relevance prediction model is designed, so that the predicted video click rate in the future 24 hours can approach the real situation infinitely, thereby enabling the CDN streaming media cache to store hot resources in time, replacing non-hot resources, improving the cache hit rate, and finally solving the problems of low hit rate of CDN cache hot points and poor user experience.
According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions, and when the instructions are executed by a processor, the streaming media caching method according to any one of the embodiments (for example, any one of fig. 1, fig. 4, and fig. 8 to fig. 9) is implemented.
Based on the computer readable storage medium provided by the above embodiment of the present invention, the video click rate predicted by the cache policy server in the future 24 hours is adopted to approach the real situation infinitely, and the situation that a large error often exists between the predicted value obtained by only using a single historical data of a continuous time and the actual situation is solved. According to the embodiment of the invention, the accuracy of cache video hotspot replacement is improved, the hit rate and the resource utilization rate of CDN stream service are effectively improved, and the use experience of a user is effectively improved.
The cache policy server and the streaming service node described above may be implemented as a general purpose processor, a Programmable Logic Controller (PLC), a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any suitable combination thereof, for performing the functions described herein.
Thus far, the present invention has been described in detail. Some details well known in the art have not been described in order to avoid obscuring the concepts of the present invention. It will be fully apparent to those skilled in the art from the foregoing description how to practice the presently disclosed embodiments.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The description of the present invention has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the invention in the form disclosed. Many modifications and variations will be apparent to practitioners skilled in this art. The embodiment was chosen and described in order to best explain the principles of the invention and the practical application, and to enable others of ordinary skill in the art to understand the invention for various embodiments with various modifications as are suited to the particular use contemplated.
Claims (18)
1. A streaming media caching method is characterized by comprising the following steps:
determining the click probability of the video in a future preset time period according to historical correlation data of each streaming service node video;
determining a hotspot attribute of the video according to the click probability of the video, wherein the hotspot attribute of the video is whether the video is a hotspot video or not;
and issuing a replacement instruction to each stream service node according to the hotspot attributes of the video so as to facilitate cache replacement of each stream service node.
2. The streaming media buffering method according to claim 1,
the historical relevance data comprises at least two items of video searching amount, video clicking amount, video collection amount and video showing time.
3. The streaming media caching method according to claim 1 or 2, wherein the determining, according to the historical relevance data of the videos of the streaming service nodes, the click probability of the videos in a predetermined period of time in the future comprises:
and counting index values related to the video click rate in a plurality of time dimensions, and determining the click probability of the video in a future preset time period according to the index values.
4. The streaming media buffering method according to claim 3,
the index value comprises at least one of a click probability, a collection probability and a search probability of the video in a plurality of time dimensions;
and/or the presence of a gas in the gas,
the plurality of time dimensions includes a plurality of time periods corresponding to a predetermined time period.
5. The streaming media buffering method according to claim 4,
the predetermined time period is one day;
the multiple time dimensions include yesterday, last week of today, last month of today, last year of today.
6. The method for caching streaming media according to claim 3, wherein the determining the probability of the video clicking within a predetermined period of time in the future according to the index value comprises:
and inputting the index value into a preset relevance prediction model, and determining the click probability of the video in a future preset time period.
7. The streaming media caching method according to claim 1 or 2, wherein the determining the hotspot attribute of the video according to the click probability of the video comprises:
for each video, judging whether the click probability of the video is greater than or equal to a preset threshold value;
if the click probability of the video is larger than or equal to a preset threshold value, judging that the video is a hot video;
and if the click probability of the video is smaller than a preset threshold value, judging that the video is a non-hotspot video.
8. A streaming media caching method is characterized by comprising the following steps:
receiving a replacement instruction issued by a cache policy server, wherein the cache policy server determines the click probability of a video in a future preset time period according to historical correlation data of the video of each streaming service node, determines the hotspot attribute of the video according to the click probability of the video, wherein the hotspot attribute of the video is whether the video is a hotspot video, and issues the replacement instruction to each streaming service node according to the hotspot attribute of the video;
and replacing the current stream service cache according to the replacement instruction.
9. The streaming media caching method according to claim 8, wherein the replacing a current streaming service cache according to the replacement instruction comprises:
judging whether the video exists in a current streaming service cache or not under the condition that the video is a hot video;
maintaining the current streaming service cache unchanged under the condition that the hotspot video exists in the current streaming service cache;
and under the condition that the hotspot video does not exist in the current streaming service cache, injecting the hotspot video into the current streaming service cache.
10. The streaming media caching method according to claim 9, wherein the replacing a current streaming service cache according to the replacement instruction comprises:
judging whether the video exists in a current streaming service cache or not under the condition that the video is a non-hotspot video;
deleting the non-hotspot video from the current streaming service cache under the condition that the non-hotspot video exists in the current streaming service cache;
and maintaining the current streaming service cache unchanged under the condition that the non-hotspot video does not exist in the current streaming service cache.
11. A cache policy server, comprising:
the click probability determining module is used for determining the click probability of the video in a future preset time period according to the historical correlation data of the videos of all the streaming service nodes;
the hotspot video determining module is used for determining hotspot attributes of the video according to the click probability of the video, wherein the hotspot attributes of the video are whether the video is a hotspot video or not;
and the replacement instruction issuing module is used for issuing a replacement instruction to each stream service node according to the hotspot attributes of the video so as to facilitate cache replacement of each stream service node.
12. The caching policy server of claim 11, wherein the caching policy server is configured to perform operations for implementing the streaming media caching method according to any one of claims 1 to 7.
13. A cache policy server, comprising but not limited to:
a server memory to store instructions;
a server processor for executing the instructions to cause the caching policy server to perform operations for implementing the streaming media caching method according to any one of claims 1 to 7.
14. A streaming service node, comprising:
the replacement instruction receiving module is used for receiving a replacement instruction issued by the cache policy server, wherein the cache policy server determines the click probability of a video in a future preset time period according to historical correlation data of videos of each streaming service node, determines the hotspot attribute of the video according to the click probability of the video, wherein the hotspot attribute of the video is whether the video is a hotspot video, and issues the replacement instruction to each streaming service node according to the hotspot attribute of the video;
and the cache replacement module is used for replacing the current stream service cache according to the replacement instruction.
15. The streaming service node according to claim 14, wherein the streaming service node is configured to perform an operation of implementing the streaming media caching method according to any one of claims 8 to 10.
16. A streaming service node, comprising but not limited to:
a node instruction memory for storing instructions;
a node instruction processor for executing the instructions to cause the caching policy server to perform operations for implementing the streaming media caching method according to any one of claims 8 to 10.
17. A streaming media caching system comprising a caching policy server according to any one of claims 11 to 13 and a streaming service node according to any one of claims 14 to 16.
18. A computer-readable storage medium storing computer instructions which, when executed by a processor, implement a streaming media caching method as claimed in any one of claims 1 to 10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810957204.2A CN110858912A (en) | 2018-08-22 | 2018-08-22 | Streaming media caching method and system, caching policy server and streaming service node |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810957204.2A CN110858912A (en) | 2018-08-22 | 2018-08-22 | Streaming media caching method and system, caching policy server and streaming service node |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110858912A true CN110858912A (en) | 2020-03-03 |
Family
ID=69634847
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810957204.2A Pending CN110858912A (en) | 2018-08-22 | 2018-08-22 | Streaming media caching method and system, caching policy server and streaming service node |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110858912A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112866724A (en) * | 2020-12-31 | 2021-05-28 | 山东远桥信息科技有限公司 | Video service processing method and system based on software defined network and edge computing technology |
| CN113297486A (en) * | 2021-05-24 | 2021-08-24 | 广州虎牙科技有限公司 | Click rate prediction method and related device |
| CN113453036A (en) * | 2020-03-24 | 2021-09-28 | 中国电信股份有限公司 | Video resource caching method and edge streaming media server of content distribution network |
| CN114727131A (en) * | 2022-03-28 | 2022-07-08 | 慧之安信息技术股份有限公司 | Streaming media stream pushing performance improving method and device based on machine learning |
| CN113297486B (en) * | 2021-05-24 | 2024-04-19 | 广州虎牙科技有限公司 | Click rate prediction method and related device |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020120818A1 (en) * | 2001-02-27 | 2002-08-29 | Markus Hofmann | Method of cache replacement for streaming media |
| KR20080078172A (en) * | 2007-02-22 | 2008-08-27 | 한국전자통신연구원 | Apparatus and method for the replacement of cache for streaming service in the proxy server |
| CN102646160A (en) * | 2012-02-21 | 2012-08-22 | 北京工业大学 | Regional water pollution comprehensive evaluation and optimization method based on entropy weight fuzzy matter element method |
| CN103091480A (en) * | 2013-01-07 | 2013-05-08 | 河北工业大学 | Entropy weight-based underground road bituminous pavement service performance evaluation method |
| CN103312776A (en) * | 2013-05-08 | 2013-09-18 | 青岛海信传媒网络技术有限公司 | Method and device for caching contents of videos by edge node server |
| CN104822068A (en) * | 2015-04-29 | 2015-08-05 | 四达时代通讯网络技术有限公司 | Streaming media proxy cache replacing method and device |
| CN104967861A (en) * | 2015-05-27 | 2015-10-07 | 上海美琦浦悦通讯科技有限公司 | CDN video buffer system and method |
| CN105095646A (en) * | 2015-06-29 | 2015-11-25 | 北京京东尚科信息技术有限公司 | Data prediction method and electronic device |
| CN107844835A (en) * | 2017-11-03 | 2018-03-27 | 南京理工大学 | Multiple-objection optimization improved adaptive GA-IAGA based on changeable weight M TOPSIS multiple attribute decision making (MADM)s |
-
2018
- 2018-08-22 CN CN201810957204.2A patent/CN110858912A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020120818A1 (en) * | 2001-02-27 | 2002-08-29 | Markus Hofmann | Method of cache replacement for streaming media |
| KR20080078172A (en) * | 2007-02-22 | 2008-08-27 | 한국전자통신연구원 | Apparatus and method for the replacement of cache for streaming service in the proxy server |
| CN102646160A (en) * | 2012-02-21 | 2012-08-22 | 北京工业大学 | Regional water pollution comprehensive evaluation and optimization method based on entropy weight fuzzy matter element method |
| CN103091480A (en) * | 2013-01-07 | 2013-05-08 | 河北工业大学 | Entropy weight-based underground road bituminous pavement service performance evaluation method |
| CN103312776A (en) * | 2013-05-08 | 2013-09-18 | 青岛海信传媒网络技术有限公司 | Method and device for caching contents of videos by edge node server |
| CN104822068A (en) * | 2015-04-29 | 2015-08-05 | 四达时代通讯网络技术有限公司 | Streaming media proxy cache replacing method and device |
| CN104967861A (en) * | 2015-05-27 | 2015-10-07 | 上海美琦浦悦通讯科技有限公司 | CDN video buffer system and method |
| CN105095646A (en) * | 2015-06-29 | 2015-11-25 | 北京京东尚科信息技术有限公司 | Data prediction method and electronic device |
| CN107844835A (en) * | 2017-11-03 | 2018-03-27 | 南京理工大学 | Multiple-objection optimization improved adaptive GA-IAGA based on changeable weight M TOPSIS multiple attribute decision making (MADM)s |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113453036A (en) * | 2020-03-24 | 2021-09-28 | 中国电信股份有限公司 | Video resource caching method and edge streaming media server of content distribution network |
| CN112866724A (en) * | 2020-12-31 | 2021-05-28 | 山东远桥信息科技有限公司 | Video service processing method and system based on software defined network and edge computing technology |
| CN113297486A (en) * | 2021-05-24 | 2021-08-24 | 广州虎牙科技有限公司 | Click rate prediction method and related device |
| CN113297486B (en) * | 2021-05-24 | 2024-04-19 | 广州虎牙科技有限公司 | Click rate prediction method and related device |
| CN114727131A (en) * | 2022-03-28 | 2022-07-08 | 慧之安信息技术股份有限公司 | Streaming media stream pushing performance improving method and device based on machine learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10572565B2 (en) | User behavior models based on source domain | |
| CN105205014B (en) | A kind of date storage method and device | |
| US8356154B2 (en) | Storage system, data relocation method thereof, and recording medium that records data relocation program | |
| CN106709068B (en) | Hot spot data identification method and device | |
| US20110087842A1 (en) | Pre-fetching content items based on social distance | |
| US20140279863A1 (en) | Network context-based content positioning for OTT delivery | |
| CN110858912A (en) | Streaming media caching method and system, caching policy server and streaming service node | |
| US20200145310A1 (en) | Search result suggestions based on dynamic network latency classification | |
| WO2017101576A1 (en) | Data resource storage method and apparatus | |
| US11494413B1 (en) | Query alerts generation for virtual warehouse | |
| CN109413694B (en) | Small cell caching method and device based on content popularity prediction | |
| CN107193754B (en) | Method and apparatus for data storage for searching | |
| CN104391947B (en) | Magnanimity GIS data real-time processing method and system | |
| CN114528231A (en) | Data dynamic storage method and device, electronic equipment and storage medium | |
| CN111225267B (en) | Content cache scheduling method, device and system and content distribution network node | |
| US20140032590A1 (en) | Windowed mid-tier data cache | |
| CN110708361A (en) | System, method and device for determining grade of digital content publishing user and server | |
| CN104714976B (en) | Data processing method and equipment | |
| CN108021464B (en) | Bottom-pocketing processing method and device for application response data | |
| CN109189696A (en) | A kind of photo classification device training method, SSD caching system and caching method | |
| CN105530303B (en) | A kind of network-caching linear re-placement method | |
| CN110019783A (en) | Attribute term clustering method and device | |
| CN103678008A (en) | Data reading method and corresponding data reading device | |
| Wang et al. | A trust-based prediction approach for recommendation system | |
| CN111970327A (en) | News spreading method and system based on big data processing |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200303 |
|
| RJ01 | Rejection of invention patent application after publication |