A parent bandwidth reduction method and device based on global cold and hot

Abstract

The application provides a global cold and hot-based parent bandwidth reduction method and device, which comprises the following steps: receiving a user request, and collecting the user request according to a preset unit time to obtain a request set; performing cache query on the request set through a timing task to obtain a query result comprising at least a cold and hot attribute corresponding to the user request; establishing a plurality of load-balanced content scheduling services; updating cache data according to the cache result by the plurality of content scheduling services to obtain updated cache data; performing cache according to the updated cache data and cold and hot scheduling according to the cold and hot attribute; and responding to the jump based on the cold and hot scheduling to serve the user request. The application solves the problem of uneven request distribution, global heat and single machine cold that may exist in single machine statistics, and can effectively cope with the request url hotspot burst scene.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of Internet, and particularly relates to a parent bandwidth reduction method and device based on global cold and hot. BACKGROUND

[0002] Through analysis of CDN acceleration requests of a video website, it is known that most of the requests are cold requests, and most of the parent bandwidth of the website is caused by these cold requests. More parent bandwidth will cause waste of parent node bandwidth cost. An effective method of saving parent cost is to perform 302 scheduling on the cold requests, so that the cold requests are jumped to the same machine through the 302 jump of HTTP, so that there is only one copy of the cache, thereby reducing the parent bandwidth.

[0003] The traditional statistical global cold and hot technical means is that the edge collects request logs, the support center collects request logs, the number of requests of each request URL unit time of the domain name is counted, the cold and hot information of each request URL is calculated, and the cold and hot information is sent to each edge machine in the form of a data file. However, for a video website with a large number of request URLs, such an implementation mode not only has long link and large delay, but also has huge data volume, cannot flexibly feedback the real-time cold and hot properties of resources, and cannot effectively cope with sudden scenes. SUMMARY

[0004] In view of the above deficiencies of the prior art, the purpose of the application is to provide a parent bandwidth reduction method and device based on global cold and hot.

[0005] The first aspect of the application provides a parent bandwidth reduction method based on global cold and hot, comprising:

[0006] S1: receiving a user request, and according to a preset unit time, the user request is summarized to obtain a request set;

[0007] S2: performing cache query on the request set through a timing task to obtain a query result including at least a cold and hot attribute corresponding to the user request;

[0008] S3: establishing a plurality of load-balanced content scheduling services, and updating cache data according to the cache result by the plurality of content scheduling services to obtain updated cache data;

[0009] S4: caching according to the updated cache data, and cold and hot scheduling according to the cold and hot attribute;

[0010] S5: gateway response based on the cold and hot scheduling jump, and serving the user request.

[0011] According to the global cold and hot based parent bandwidth reduction method, the request set in step S1 comprises a request URL and a request number.

[0012] According to the global cold and hot based parent bandwidth reduction method, step S2 further comprises the following steps.

[0013] S21: The edge scheduling component sends the request set to the content scheduling center through a timing task.

[0014] S22: The content scheduling center calculates the MD5 value of the request set.

[0015] S23: The content scheduling center performs cache query according to the MD5 value, judges the cold and hot attribute according to the statistical value in the request set when there is no cache item corresponding to the MD5 value, judges the cold and hot attribute according to the statistical value in the cache when there is the cache item corresponding to the MD5 value, and obtains a query result.

[0016] According to the global cold and hot based parent bandwidth reduction method, in step S21, the edge scheduling component sends the request set to the content scheduling center through an HTTP mode.

[0017] According to the global cold and hot based parent bandwidth reduction method, in step S3, a plurality of load-balanced content scheduling services are realized through the LVS technology.

[0018] According to the global cold and hot based parent bandwidth reduction method, in step S4, the edge scheduling component further comprises the following steps when performing cache according to the updated cache data.

[0019] Cleaning the expired cache data.

[0020] According to the global cold and hot based parent bandwidth reduction method, the cold and hot scheduling in step S4 specifically comprises the following steps.

[0021] When the cold and hot attribute of the user request is a cold request, a serviceable virtual IP address is obtained in the planning area coverage hash to hit the cache; when the cold and hot attribute of the user request is a hot request, the local service is used.

[0022] The second aspect of the application further provides a global cold and hot based parent bandwidth reduction device, comprising:

[0023] The aggregation module is used for receiving a user request, aggregating the user request according to a preset unit time, and obtaining a request set.

[0024] The query module is configured to cache query the request set through a timing task, and obtain a query result including at least a cold and hot attribute corresponding to the user request;

[0025] The update module is configured to establish a plurality of load-balanced content scheduling services, and update cache data according to the cache result by the plurality of content scheduling services, to obtain updated cache data;

[0026] The scheduling module is configured to cache according to the updated cache data, and perform cold and hot scheduling according to the cold and hot attribute;

[0027] The execution module is configured to perform gateway jump in response to the cold and hot scheduling, and serve the user request.

[0028] The third aspect of the application provides a parent bandwidth reduction device based on global cold and hot, comprising:

[0029] a memory and at least one processor, wherein the memory stores instructions;

[0030] The at least one processor invokes the instructions in the memory, so that the parent bandwidth reduction device based on global cold and hot executes the parent bandwidth reduction method based on global cold and hot as any one of the above.

[0031] The fourth aspect of the application provides a computer readable storage medium, wherein the computer readable storage medium stores instructions, and the instructions are executed by a processor to implement the parent bandwidth reduction method based on global cold and hot as any one of the above.

[0032] The application has the following advantages:

[0033] The application provides a global cold and hot based parent bandwidth reduction method, device, equipment and storage medium, through the edge scheduling component, request URL data in the aggregation unit time is collected, and a unified request is initiated to a content scheduling center for query, on one hand, edge scheduling execution time consumption can be reduced, and influence on the first packet time of a client request can be reduced, on the other hand, the interaction times between the edge scheduling and the content scheduling center can be reduced, the pressure of the content scheduling center can be reduced, and the high availability of the server can be ensured; secondly, the cold and hot of the request URL is calculated through the statistical data of the global edge machine, so that the resource cold and hot judgment is more accurate; the problem of uneven request distribution, global hot and single machine cold that may exist in single machine statistics is solved; in addition, the edge scheduling component and the content scheduling component interact in real time, and the request number of the request URL is updated constantly, so that the latest data statistics are ensured, the statistics are more timely, and the scene of request URL hotspot burst can be effectively solved; and for the traditional coverage planning, basically, the planning is to the regional granularity, and the cold resource may exist in multiple cache situations across provinces, therefore, the scheme proposes a regional planning mode that can realize cross-regional planning, automatically plans the coverage of each region to the same region, sets the region across several regions, and realizes cold resource hashing on the region, so that the parent bandwidth is saved, and the parent cost is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0034] The accompanying drawings are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings:

[0035] Fig. 1 A global cold and hot based parent bandwidth reduction method flowchart provided by the embodiment of the application;

[0036] Fig. 2 A global cold and hot based parent bandwidth reduction device structure diagram provided by the embodiment of the application.

[0037] Reference signs:

[0038] 100, aggregation module; 200, query module; 300, update module; 400, scheduling module; 500, execution module. DETAILED DESCRIPTION

[0039] In the following, the technical solutions of the present application will be described clearly and completely in connection with the drawings, which are obviously a part of the embodiments of the present application, but not all the embodiments. It should be understood that the description is only exemplary, and is not intended to limit the scope of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of the present application.

[0040] In addition, in the following description, the description of well-known structures and techniques is omitted to avoid unnecessary confusion of the concepts disclosed in the present application.

[0041] In the description of the present application, it should be noted that, unless otherwise explicitly specified and limited, the orientation or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second", "third" are only for description purposes and cannot be understood as indicating or implying relative importance. The terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixed connection, or detachable connection, or integral connection; it can be mechanical connection, or electrical connection; it can be direct connection, or indirect connection through intermediate medium, or internal communication of two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0042] The exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments are not intended to represent all implementations consistent with the present application. Rather, they are merely examples of methods and systems consistent with some aspects of the present application as detailed in the appended claims.

[0043] The following is based on Figs. 1-2 The embodiments of the present application are described.

[0044] As Fig. 1 shown, the present application provides a global hot and cold-based parent bandwidth reduction method, comprising:

[0045] S1: receiving a user request, and aggregating the user request according to a preset unit time to obtain a request set;

[0046] The request set in step S1 includes a request url and a request number.

[0047] In this phase, the edge scheduling component first aggregates requests in a unit of time. Specifically, after receiving a customer request, the edge scheduling component aggregates requests in a unit of time, and counts the request number of each request url in the planned area in a unit of time. For example, the request number of the request url http: / / www.abc.com / 1.jpg in 5 seconds is counted to be 10 times.

[0048] S2: cache query of the request set by a timing task to obtain a query result including at least a cold and hot attribute corresponding to the user request;

[0049] In step S2, further comprising:

[0050] S21: the edge scheduling component sends the request set to the content scheduling center by a timing task;

[0051] Further, the edge scheduling component initiates cache query at a timing. Specifically, the edge scheduling component reports the request url and the request number aggregated in a unit of time to the content scheduling center in the form of an HTTP request in a timing task.

[0052] Further, the edge scheduling component performs certain operations at a timing by executing a timing task, that is, performs a preset task according to a preset time interval. This mechanism helps to reduce waste of system resources and ensures that data is processed and reported at an appropriate time.

[0053] In step S21, the edge scheduling component sends the request set to the content scheduling center in the form of HTTP.

[0054] Further, the HTTP request packet body can be but is not limited to in the form of json.

[0055] S22: the content scheduling center calculates an MD5 value of the request set;

[0056] S23: the content scheduling center performs cache query according to the MD5 value, judges the cold and hot attribute according to the statistical value in the request set when there is no cache item corresponding to the MD5 value, judges the cold and hot attribute according to the statistical value in the cache when there is a cache item corresponding to the MD5 value, and obtains a query result.

[0057] In steps S21-S23, after receiving the edge scheduling query request, the content scheduling system calculates the MD5 value of the URL+region, and then queries the cache according to the MD5 value. If the cache does not exist, the hot / cold attribute of the URL is determined according to the reported statistical value combined with the hot / cold threshold. If the cache exists, the statistical value in the cache is accumulated and combined with the hot / cold threshold to determine the hot / cold attribute of the URL.

[0058] Further, in order to efficiently manage and query the cache, the system combines the requested URL with the region of the request source to calculate an MD5 value, which is used as a unique identifier for subsequent cache query and update operations. The system will evaluate the hot / cold attribute of the URL based on the reported statistical value. Specifically, if the cache does not exist, the statistical value is compared with the preset hot / cold threshold. If the access frequency is higher than the hot threshold or lower than the cold threshold, it is marked as hot content or cold content, respectively. If the cache exists, the system reads the existing statistical value of the URL from the cache and accumulates (or updates) it to reflect the latest access situation. Then, it compares the accumulated statistical value with the hot / cold threshold to re-evaluate or confirm the hot / cold attribute of the URL.

[0059] The hot / cold statistical method of content scheduling can be implemented by maintaining time slots. In a specific embodiment, the implementation is as follows: assuming the statistical period is 300s and the number of time slots is 5. The specific steps are as follows: 1) Time slot statistical process: 0-60s, statistics to the 1st time slot; 60s-120s, statistics to the 2nd time slot; 120s-180s, statistics to the 3rd time slot; 180s-240s, statistics to the 4th time slot; 240s-300s, statistics to the 0th time slot; 300s-360s, statistics to the 1st time slot;...; 1320s-1380s, statistics to the 2nd time slot;...; 2) Accumulate the statistical value of the previous 5 time slots to get the total statistical value; 3) When time slots are reused, the total statistical value needs to be pre-subtracted from the original statistical value of the reused time slots;

[0060] 4) For request intervals spanning multiple time slots, expired time slot statistical data needs to be cleaned up.

[0061] S3: Establish multiple load-balanced content scheduling services, and update the cache data according to the cache result to obtain updated cache data;

[0062] In step S3, the content scheduling center needs to face the requests of a large number of edge machines, in order to ensure the high availability of the system, a multi-pod mode can be realized by hanging an lvs, each content scheduling service can process all requests, and load balancing is realized by lvs, wherein a plurality of content scheduling pods update cache data by means of grabbing a distributed lock.

[0063] Specifically, when the content scheduling grabs the distributed lock, first determine whether the lock is successfully grabbed, if not, exit the logic, if the lock is successfully grabbed, batch get data from the redis update queue, then gather the statistical value of the url to be updated, batch read the current statistical value of the url to be updated from redis, and then update the statistical value of the url to be updated to redis in batches, and the execution is ended.

[0064] In step S3, a plurality of load-balanced content scheduling services are realized by hanging an LVS technology.

[0065] In the Linux environment, LVS is a high-performance and high-availability server cluster technology based on Linux, which realizes load balancing by distributing network traffic to multiple servers, and the hanging mode of LVS, i.e. its deployment and configuration mode, mainly depends on its working mode (such as NAT mode, DR mode, TUN mode, etc.) and specific system environment and requirements.

[0066] S4: caching according to the updated cache data, and cold and hot scheduling according to the cold and hot attributes;

[0067] In step S4, the edge scheduling component first receives the response of the content scheduling center and performs periodic caching. Specifically, after the edge scheduling component receives the query response, it caches the cold and hot attributes of the response url, and periodically cleans up expired cache data. Through the continuous interaction between the edge scheduling component and the content scheduling component, the cold and hot attributes of the url can be kept up-to-date, and the url that has not been updated for a long time will also be cleaned up due to time expiration.

[0068] In step S4, the edge scheduling component caches according to the updated cache data, and the method further comprises:

[0069] Cleaning up expired cache data.

[0070] The cold and hot scheduling in step S4 specifically comprises:

[0071] When the cold and hot attribute of the user request is cold request, a serviceable virtual IP address is obtained in the planning area coverage hash to hit the cache; when the cold and hot attribute of the user request is hot request, the local service is used.

[0072] The cold request mentioned above refers to a content request with low access frequency or newly appeared, which usually does not occur frequently, but if the content is obtained from the original server every time, a large amount of bandwidth and resources will be consumed. For the cold request, the system selects one or more virtual IP addresses (VIP) within the planned regional coverage according to the source of the user request and a certain hash algorithm, wherein the planned regional coverage refers to a geographical area or network area predefined by the system for optimizing content distribution, and the hash algorithm is used to generate a unique identifier according to certain characteristics of the request (such as URL, user ID, etc.) for subsequent cache lookup or content distribution decision. After selecting the VIP, the system tries to use the VIP to query the cache, if the corresponding content exists in the cache, the content can be directly read from the cache and returned to the user without obtaining from the original server, which can greatly reduce the access times to the original server, reduce bandwidth consumption and improve response speed.

[0073] The hot request mentioned above refers to a content request with very high access frequency, which usually corresponds to popular content in the system and needs to be accessed quickly and frequently. For the hot request, the system usually selects to provide service directly on the server handling the request, i.e. local service, which means that when the user initiates a hot request, the system immediately reads the content from the local storage or memory and returns it to the user without additional steps such as network transmission or cache lookup, which can ensure that hot content can be accessed by users with the lowest delay and the highest throughput.

[0074] S5: The gateway responds to the jump based on the cold-hot scheduling and serves the user request.

[0075] Further, when 302 jump, the server sets a Location field in the HTTP response header at the same time of returning 302 status code, which contains the address of the redirection target URL (URL B), then the browser automatically sends a new request to the URL (URL B) after receiving the 302 status code and the Location field returned by the server, and displays the content of the URL.

[0076] Finally, the gateway will serve the user request according to the result of cold-hot scheduling, including reading content from cache, obtaining content from backend server, or directly returning some form of response, no matter which way is adopted, the goal of the gateway is to satisfy the user request as soon as possible and accurately, while optimizing system performance and resource usage.

[0077] As shown in Fig. 2 The application also provides a parent back bandwidth reduction device based on global cold and hot, which comprises:

[0078] The aggregation module 100 is configured to receive a user request and aggregate the user request according to a preset unit time to obtain a request set.

[0079] The query module 200 is configured to cache query the request set through a timing task to obtain a query result including at least a cold and hot attribute corresponding to the user request.

[0080] The update module 300 is configured to establish a plurality of load-balanced content scheduling services, and update cache data according to the cache result by the plurality of content scheduling services to obtain updated cache data.

[0081] The scheduling module 400 is configured to cache according to the updated cache data and perform cold and hot scheduling according to the cold and hot attribute.

[0082] The execution module 500 is configured to perform gateway jump in response to the cold and hot scheduling and serve the user request.

[0083] The application further provides a global cold and hot based parent bandwidth reduction device, comprising:

[0084] a memory and at least one processor, wherein the memory stores instructions;

[0085] The at least one processor calls the instructions in the memory to enable the global cold and hot based parent bandwidth reduction device to perform the global cold and hot based parent bandwidth reduction method as described in any one of the above.

[0086] The application further provides a computer readable storage medium, wherein the computer readable storage medium stores instructions, and the instructions are executed by a processor to implement the global cold and hot based parent bandwidth reduction method as described in any one of the above.

[0087] Further, the global cold and hot based parent bandwidth reduction device can have great differences due to different configurations or performances, and can include one or more processors (central processing units, CPUs), for example, one or more processors and memories, one or more storage media storing application programs or data, for example, one or more mass storage devices, wherein the memories and the storage media can be temporary storage or persistent storage, the programs stored in the storage media can include one or more modules, each module can include a series of instruction operations of the global cold and hot based parent bandwidth reduction device, and further, the processor can be configured to communicate with the storage medium and execute the series of instruction operations of the storage medium on the global cold and hot based parent bandwidth reduction device.

[0088] One or more power supplies, one or more wired or wireless network interfaces, one or more input / output interfaces, one or more operating systems, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc., can also be included. Those skilled in the art will understand that the global hot and cold bandwidth reduction device structure provided by the present application is not limited to the device structure illustrated, but can include more or fewer components than shown, or combine some components, or arrange the components differently.

[0089] Finally, it should be noted that the above-mentioned embodiments are merely used to illustrate the technical solutions of the present application, rather than limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalent features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application. Any changes or replacements easily thought of by those skilled in the art within the technical scope disclosed by the present application should be covered within the protection scope of the present application.

Claims

1. A global cold-hot based parent bandwidth reduction method, characterized in that, The method comprises the following steps: S1: receiving a user request and aggregating the user request according to a preset unit time to obtain a request set; S2: performing cache query on the request set by a timing task to obtain a query result comprising at least a cold and hot attribute corresponding to the user request; S21: the edge scheduling component sends the request set to a content scheduling center by the timing task; S22: the content scheduling center calculates an MD5 value of the request set; S23: the content scheduling center performs cache query according to the MD5 value, judges the cold and hot attribute according to statistical values in the request set when there is no cache item corresponding to the MD5 value, judges the cold and hot attribute according to statistical values in the cache when there is the cache item corresponding to the MD5 value, and obtains the query result; S3: establishing a plurality of load-balanced content scheduling services, updating cache data according to the cache result by the plurality of content scheduling services, and obtaining updated cache data; S4: caching according to the updated cache data and performing cold and hot scheduling according to the cold and hot attribute; 2. The global cold-hot based bandwidth reduction method of claim 1, wherein, S5: the gateway responds to the jump based on the cold and hot scheduling and serves the user request.

3. The global cold-hot based bandwidth reduction method of claim 1, wherein, The request set in step S1 comprises a request URL and a request number.

4. The global cold-hot based bandwidth reduction method of claim 1, wherein, In step S21, the edge scheduling component sends the request set to the content scheduling center by HTTP.

5. The global cold-hot based bandwidth reduction method of claim 1, wherein, In step S3, the plurality of load-balanced content scheduling services are implemented by the upper hanging LVS technology. In step S4, when the edge scheduling component caches according to the updated cache data, the edge scheduling component further comprises the following steps:

6. A global cold-hot based parent bandwidth reduction device, comprising: cleaning up expired cache data. The method comprises the following steps: an aggregation module for receiving a user request and aggregating the user request according to a preset unit time to obtain a request set; a query module for performing cache query on the request set by a timing task to obtain a query result comprising at least a cold and hot attribute corresponding to the user request; the query module is further used for sending the request set to a content scheduling center by the timing task; calculating an MD5 value of the request set, performing cache query according to the MD5 value, judging the cold and hot attribute according to statistical values in the request set when there is no cache item corresponding to the MD5 value, judging the cold and hot attribute according to statistical values in the cache when there is the cache item corresponding to the MD5 value, and obtaining the query result; an update module for establishing a plurality of load-balanced content scheduling services, updating cache data according to the cache result by the plurality of content scheduling services, and obtaining updated cache data; a scheduling module for caching according to the updated cache data and performing cold and hot scheduling according to the cold and hot attribute; The cold-hot scheduling specifically comprises: when the cold-hot attribute of the user request is a cold request, obtaining a serviceable virtual IP address in a planning area coverage to hit a cache; when the cold-hot attribute of the user request is a hot request, then serving through a local service. The execution module is configured to perform gateway jumping in response to the cold-hot scheduling, and serve the user request.

7. A global cold-hot based parent bandwidth reduction device, comprising: The method comprises the following steps: A memory and at least one processor, wherein the memory stores instructions; At least one processor calls the instructions in the memory, so that a global cold-hot-based parent bandwidth reduction device executes a global cold-hot-based parent bandwidth reduction method as claimed in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores instructions, and the instructions are executed by a processor to implement a global cold-hot-based parent bandwidth reduction method as claimed in any one of claims 1-5.

Images