CN114827159A

CN114827159A - Network request path optimization method, device, equipment and storage medium

Info

Publication number: CN114827159A
Application number: CN202210344154.7A
Authority: CN
Inventors: 谷崇明; 高岩
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2022-03-31
Filing date: 2022-03-31
Publication date: 2022-07-29
Anticipated expiration: 2042-03-31
Also published as: CN114827159B

Abstract

The disclosure provides a network request path optimization method, a network request path optimization device, network request path optimization equipment and a storage medium, relates to the technical field of computers, particularly relates to the field of artificial intelligence such as big data and cloud computing, and can be applied to a media cloud scene. The network request path optimization method comprises the following steps: receiving a resource request message for acquiring a target resource; if the target resource belongs to a hot spot resource in the current period, determining a target storage server in a plurality of candidate storage servers corresponding to the target resource, wherein the hot spot resource in the current period comprises: a first hot spot resource and a second hot spot resource, the first hot spot resource being determined based on a resource request message within a first period, the second hot spot resource being determined based on a resource request message within a second period, the first period being a previous period of the current period, the second period being earlier than the first period; and forwarding the resource request message to the target storage server. The load balancing method and the load balancing device can balance loads and improve system stability.

Description

Network request path optimization method, device, equipment and storage medium

Technical Field

The disclosure relates to the technical field of computers, in particular to the field of artificial intelligence such as big data and cloud computing, can be applied to a media cloud scene, and particularly relates to a network request path optimization method, device, equipment and storage medium.

Background

A Content Delivery Network (CDN) is an intelligent virtual Network built on the basis of the existing Network, and through functional modules such as load balancing, Content Delivery, scheduling, and the like, a user can obtain required Content nearby, so that Network congestion is reduced, and the access response speed and hit rate of the user are improved. The CDN may be divided into multiple tiers, and each tier of CDN may include at least one load balancing server and at least one storage server.

When a user acquires a resource through the CDN, generally, the load balancing server receives a resource request, and determines a target storage server in at least one storage server based on a resource identifier in the resource request.

In the related art, when the load balancing server determines the target storage server, the load balancing server is based on a hash algorithm, and at this time, the resource request of the same resource is sent to the same target storage server.

Disclosure of Invention

The disclosure provides a network request path optimization method, device, equipment and storage medium.

According to an aspect of the present disclosure, there is provided a network request path optimization method, including: receiving a resource request message for acquiring a target resource; if the target resource belongs to the hot spot resource in the current period, determining a target storage server in a plurality of candidate storage servers corresponding to the target resource, wherein the hot spot resource in the current period comprises: a first hot spot resource and a second hot spot resource, the first hot spot resource being determined based on a resource request message within a first period, the second hot spot resource being determined based on a resource request message within a second period, the first period being a previous period of the current period, the second period being earlier than the first period; and forwarding the resource request message to the target storage server.

According to another aspect of the present disclosure, there is provided a network request path optimizing apparatus including: a receiving module, configured to receive a resource request message for acquiring a target resource; a determining module, configured to determine a target storage server in a plurality of candidate storage servers corresponding to the target resource if the target resource belongs to a hot resource in a current period, where the hot resource in the current period includes: a first hot spot resource and a second hot spot resource, the first hot spot resource being determined based on a resource request message within a first period, the second hot spot resource being determined based on a resource request message within a second period, the first period being a previous period of the current period, the second period being earlier than the first period; a sending module, configured to forward the resource request message to the target storage server.

According to another aspect of the present disclosure, there is provided an electronic device including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of the above aspects.

According to another aspect of the present disclosure, there is provided a non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method according to any one of the above aspects.

According to another aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of the above aspects.

According to the technical scheme disclosed by the invention, the load can be balanced and the system stability can be improved.

It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.

Drawings

The drawings are included to provide a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:

FIG. 1 is a schematic diagram according to a first embodiment of the present disclosure;

FIG. 2 is a schematic illustration of different cycles in an embodiment of the present disclosure;

fig. 3 is a schematic diagram of an application scenario of forwarding a resource request message based on a hash algorithm according to an embodiment of the present disclosure;

FIG. 4 is a schematic diagram according to a second embodiment of the present disclosure;

FIG. 5 is a flow chart of a method for generating current policy information in an embodiment of the present disclosure;

FIG. 6 is a schematic diagram of generating current policy information based on first policy information and second policy information in an embodiment of the disclosure;

FIG. 7 is a schematic diagram according to a third embodiment of the present disclosure;

fig. 8 is a schematic diagram of an electronic device for implementing a network request path optimization method according to an embodiment of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of the embodiments of the disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

Because the resource request of the same resource is sent to the same target storage server, if the resource is a hot resource, the access pressure of the corresponding target storage server is large, and the overall stability and throughput of the CDN are seriously affected.

In order to equalize the access pressure of the storage server, the present disclosure provides the following embodiments.

Fig. 1 is a schematic diagram according to a first embodiment of the present disclosure, where this embodiment provides a method for optimizing a network request path, where the method includes:

101. a resource request message to acquire a target resource is received.

102. If the target resource belongs to a hot spot resource in the current period, determining a target storage server in a plurality of candidate storage servers corresponding to the target resource, wherein the hot spot resource in the current period comprises: the resource allocation method includes determining a first hot spot resource based on a resource request message in a first period and determining a second hot spot resource based on a resource request message in a second period, where the first period is a previous period of the current period and the second period is earlier than the first period.

103. And forwarding the resource request message to the target storage server.

The network request path optimization method of this embodiment may be executed by a load balancing server in the CDN.

The resource request message may come from a user terminal, that is, when the user needs to obtain the target resource, the resource request message may be sent through the user terminal.

The target resource refers to a resource to be acquired or accessed by a user, such as a webpage, a game, an application, an audio/video, and the like.

The resource request message may include a resource identifier of the target resource, so as to indicate that the resource request message is used to acquire the target resource.

The Resource identifier refers to information for identifying a Resource, and may be, for example, a Uniform Resource Locator (URL) address, or may also be a Resource name.

The hot spot resource refers to a resource with a large access amount in a certain period, and can be determined based on the resource request message in the corresponding period.

The hotspot resources are updatable, and the hotspot resources of different periods can be the same or different.

The current period refers to a period to which a time point of receiving the resource request message belongs.

Referring to fig. 2, assuming that the starting time point of the current cycle is represented as T0 and the ending time point of the current cycle is T1, T1 is T0+ T, where T is the duration of the cycle, and the duration of the cycle is a settable value, for example, T is 10 seconds. Therefore, for the resource request messages received in the period from T0 to T1, the current period is [ T0-T, T0 ].

The first cycle is a previous cycle of the current cycle, and as shown in FIG. 2, the time range of the first cycle is [ T0-T, T0 ].

The second period is earlier than the first period, and the second period in fig. 2 has a time range [ T0-k × T, T0- (k-1) × T ], where k is a positive integer greater than 1, and the specific value may be determined based on a preset delayed aging strategy, for example, k is 30.

For the resource request message received in the current period [ T0, T0+ T ], if the target resource corresponding to the resource request message belongs to the hot resource in the current period, the target storage server may be determined from a plurality of candidate storage servers, for example, one candidate storage server may be randomly selected from the plurality of candidate storage servers as the target storage server.

And aiming at a plurality of candidate storage servers corresponding to the target resource, a plurality of candidate storage servers refers to at least two candidate storage servers. The candidate storage servers may be predetermined, for example, a plurality of storage servers in a CDN hierarchy where the load balancing server is located may be used as the candidate storage servers. Further, the candidate storage servers may be determined based on the number of the resource request messages of the target resource in the corresponding period, for example, if the target resource is a first hot resource, the candidate storage servers are determined based on the number of the resource request messages of the target resource in the first period, or if the target resource is a second hot resource, the candidate storage servers are determined based on the number of the resource request messages of the target resource in the second period. For the determination process of the candidate storage server, reference may be made to the description of the subsequent embodiments.

After the target storage server is determined, the resource request message may be sent to the target storage server, so as to obtain the target resource from the target storage server.

In addition, since the hot spot resources are variable in different periods, if the hot spot resources in the current period only include the first hot spot resource, that is, only relate to the resource request message in the previous period of the current period, the hot spot resources in the current period may generate a shock problem, that is, for a certain resource, the resource may belong to the hot spot resources in the period in a certain period, does not belong to the hot spot resources in the next period of the period, and belongs to the hot spot resources in the next period. The oscillation problem of the hot spot resource can cause the oscillation of the load balancing strategy of the load balancing server, and the overall stability and other performances are affected.

In this embodiment, for a hot resource in a current period, a target storage server may be determined from a plurality of candidate storage servers, and compared with a method of fixing one target storage server, access pressure of the storage servers may be balanced, and a load balancing effect of the CDN is improved. In addition, in this embodiment, the hot spot resources in the current period include a first hot spot resource and a second hot spot resource, the first hot spot resource is determined based on the resource request message in the first period, and the second hot spot resource is determined based on the resource request message in the second period, because the second period is earlier than the first period, the hot spot resources in the current period are not only related to the resource request message in the previous period (i.e., the first period) of the current period, but also related to the resource request message in the earlier period (i.e., the second period), so that delayed aging of the hot spot resources is achieved.

Fig. 3 is a schematic diagram of an application scenario in which a resource request message is forwarded based on a hash algorithm according to an embodiment of the present disclosure.

As shown in fig. 3, taking a certain hierarchy of the CDN as an example, the CDN hierarchy 300 may include at least one load balancing server and at least one storage server therein, and it is assumed that the hierarchy includes two load balancing servers (represented by load balancer-1 and load balancer-2, respectively) and three storage servers (represented by storage server-1, storage server-2, and storage server-3, respectively).

When a user needs to access a certain resource, a resource request message can be sent through a user terminal, and the resource request message can contain a resource identifier of a target resource.

The user terminal may include: personal computers (Personal computers, PCs), mobile devices, smart home devices, wearable devices and the like, wherein the mobile devices include mobile phones, laptops, tablet computers and the like, the smart home devices include smart speakers, smart televisions and the like, and the wearable devices include smart watches, smart glasses and the like.

The target resource is a resource to be acquired or accessed by the user, such as a webpage, a game, an application, an audio/video and the like.

The resource identifier of the target resource is, for example, a URL address of the target resource, or a resource name.

After the load balancing server receives the resource request message, the load balancing server may determine a target storage server based on a preset algorithm, where the target storage server is a storage server serving as a forwarding destination.

Generally speaking, the load balancing server may determine the target storage server based on a hash algorithm, and at this time, the load balancing server may perform a hash operation on the resource identifier in the resource request message to obtain a hash value, and determine the target storage server based on the hash value.

Based on the hash algorithm, the same resource will correspond to the same target storage server, for example, referring to fig. 3, the resource request messages for requesting resource a for 8 times are all sent to storage server-1, the resource request messages for requesting resource C for 10 times are all sent to storage server-2, and the resource request messages for requesting resource B for 220 times are all sent to storage server-3.

Because the resource request message of the same resource is sent to the same storage server, if the resource is a hot resource, that is, the resource with a larger number of resource request messages, for example, the resource B, the storage server corresponding to the hot resource, for example, the storage server-3, has a larger access pressure.

In order to balance the access pressure of the storage servers, the load balancing server may employ some load balancing algorithms, for example, a hotspot balancing table may be configured, and at this time, the same resource may be distributed to multiple storage servers.

For example, based on the hash algorithm, the target storage server of resource B is only a fixed one, i.e., storage server-3. However, after the balancing algorithm is adopted, a plurality of storage servers can be configured for the resource B, the plurality of storage servers can be referred to as candidate storage servers, and the load balancing server can randomly select one of the plurality of candidate storage servers as the target storage server of the resource B.

The determination method of the candidate storage server may refer to related technologies, which is not limited in this embodiment. For example, a storage server determined based on a hash algorithm may be referred to as a hash server; the number of the storage servers to be dispersed can be determined based on the number of the resource request messages corresponding to the same resource in a certain period, and then the storage servers in the number are randomly selected to serve as non-hash servers; and taking the hash server and the non-hash server as candidate storage servers.

As shown in table 1, for resource a and resource B, the candidate storage servers corresponding thereto may be determined.

TABLE 1

For a certain target resource, the candidate storage servers may include a hash server and a non-hash server, where the hash server is a storage server determined based on a hash algorithm, and the non-hash server is another storage server determined to implement load balancing.

Referring to Table 1, taking resource A as an example, c [31] is the hash server corresponding to resource A, and c [30] and c [32] are the non-hash servers corresponding to resource A.

After receiving the resource request message, the load balancing server may determine a target storage server among the candidate storage servers, for example, for resource a, the target storage server may be determined in c [30], c [31], and c [32 ].

The embodiment is not limited as to how to determine the target storage server from the candidate storage servers, for example, one of the candidate storage servers may be randomly selected.

After the target storage server is determined, the load balancing server may forward the resource request message to the target storage server.

As analyzed above, after the equalization algorithm is introduced, each resource may correspond to a plurality of candidate storage servers, so that the problem of high access pressure caused by fixing a resource request message to a certain storage server based on the hash algorithm can be solved.

However, for a certain resource, there may be a shock problem only corresponding to a plurality of candidate storage servers, and in order to reduce the shock problem caused by switching a target resource between a hot resource and a non-hot resource, in the embodiment of the present disclosure, each resource may correspond to a plurality of candidate storage servers, delay aging of the hot resource may also be introduced, policy merging may also be implemented, which may specifically refer to the subsequent embodiments.

Fig. 4 is a flowchart illustrating a second embodiment of the present disclosure, where this embodiment provides a method for optimizing a network request path, where the method of this embodiment may be applied to a load balancing server, and the method provided by this embodiment includes:

401. acquiring current policy information, wherein the current policy information comprises: the server comprises a resource identifier and a plurality of server identifiers corresponding to the resource identifier.

The current policy information is at least one group, and each group of current policy information comprises: the resource identification and a plurality of server identifications corresponding to the resource identification.

Plural means at least two. The server identification may be the IP address of the server,

at least one set of current policy information may constitute a current policy set.

Wherein the resource identification comprises: the resource identifier of the first hot spot resource, and the resource identifier of the second hot spot resource.

The first hot spot resource is determined based on a resource request message in a first period, the second hot spot resource is determined based on a resource request message in a second period, the first period is a previous period of the current period, and the second period is earlier than the first period.

That is, the resource identification in the current policy information includes the resource identification of the delayed aging hotspot resource (i.e., the second hotspot resource).

The current policy information may further include an expiration time, and updating of the current policy information may be implemented based on the expiration time.

The current policy information may be as shown in table 2.

TABLE 2

The resource a and the resource D may be first hot-spot resources, and t0 is a statistical time point corresponding to the current period, for example, a start time point t0 of the current period, that is, an end time point of the first period (a previous period of the current period). The candidate storage servers for resource a and resource D may be determined based on the resource a or resource D resource request messages during the first period. ttl is a preset lifetime, for example, ttl is 30 × T, T is a period duration, for example, 10 seconds, and ttl is 300 seconds.

Resource C is a second hotspot resource, and resource C and its corresponding candidate storage server may be determined based on the resource request message of resource C in the second period. t 0' is the corresponding statistical time point of resource C, such as the end time point of the second period. Since the second cycle is earlier than the first cycle, t0 'is also earlier than t0, i.e., t 0' < t 0.

The generation process of the current policy information can be referred to the embodiment shown in fig. 5.

402. Receiving a resource request message, wherein the resource request message comprises a resource identifier of a target resource.

401 and 402 have no timing limitation relationship.

403. And judging whether the resource identifier of the target resource belongs to the current policy information, if so, executing 404, and otherwise, executing 405.

For example, as shown in table 2, the current policy information indicates that the resource of the target resource is identified as resource a, and since resource a belongs to the current policy information, 404 is executed.

For another example, the current policy information is shown in table 2, the resource identifier of the target resource is resource B, and since resource B does not belong to the current policy information, 405 is executed.

404. And randomly selecting one candidate storage server from a plurality of candidate storage servers corresponding to the target resource as a target storage server.

The plurality of servers indicated by the plurality of server identifiers corresponding to the resource identifier of the target resource in the current policy information may be used as the plurality of candidate storage servers.

For example, referring to table 2, the resource identifier of the target resource is resource a, and since the identifiers of the servers corresponding to resource a in the current policy information are c [30], c [31], and c [32], one of the servers c [30], c [31], and c [32] may be randomly selected as the target storage server, for example, the storage server corresponding to c [30] is selected as the target storage server.

By selecting a candidate storage server as a target storage server in a random manner, randomness can be improved, and resource request messages can be better balanced.

405. And determining the target storage server based on a hash algorithm.

For example, the resource identifier of the target resource is resource B, the current policy information is shown in table 2, and since the current policy information does not include resource B, the target storage server may be determined based on a hash algorithm for resource B, that is, a hash value corresponding to resource B is calculated, and the storage server corresponding to the hash value is used as the target storage server.

406. Forwarding the resource request message to a target storage server.

After the load balancing server forwards the resource request message to the target storage server, if the target storage server stores the target resource, the target storage server can feed back the target resource to the load balancing server after acquiring the target resource based on the resource identifier. Or, if the target storage server does not store the target resource, the target storage server may execute a source returning operation to obtain the target resource and feed back the target resource to the load balancing server. The process of the back source operation can be implemented by using the related art, and this embodiment does not limit this.

In this embodiment, by acquiring current policy information, where the current policy information includes a resource identifier, whether a target resource belongs to a hot resource in a current period may be simply determined based on whether the resource identifier of the target resource belongs to the current policy information, and then different determination manners of a target storage server may be adopted, that is, if the resource identifier of the target resource belongs to the current policy information, one of a plurality of candidate storage servers is randomly selected as the target storage server; and if the resource identifier of the target resource does not belong to the current strategy information, determining the target storage server based on a hash algorithm. The target storage server can be determined according to actual conditions, and the load balancing effect is improved.

Fig. 5 is a flowchart of a method for generating current policy information in an embodiment of the present disclosure, and as shown in fig. 5, the method for generating current policy information includes:

501. determining first policy information based on the resource request message within the first period, the first policy information comprising: the resource management system comprises a first resource identifier, a first expiration time and a first server identifier corresponding to the first resource identifier, wherein the first resource identifier is the resource identifier of the first hot spot resource, and the first server identifiers are multiple.

Wherein the first policy information may be determined based on the resource request message in the first period (i.e., the previous period of the current period) at a start time point of the current period.

For example, referring to fig. 2, the first policy information may be determined based on the resource request message in the period of [ T0-T, T0] at a time point of T0.

The first resource identifier is a resource identifier of a hot resource (i.e., a first hot resource) in a first period, and the hot resource in the first period may be determined based on the number of resource request messages of the same resource, for example, a resource in the first period for which the number of resource request messages of the same resource is greater than a preset value may be determined as the hot resource, or a preset number of resources in the first period for which the number of resource request messages of the same resource is greater may be determined as the hot resource.

After the first hotspot resource is determined, the corresponding expiration time and a plurality of candidate storage servers can be determined, the expiration time corresponding to the first hotspot resource can be referred to as a first expiration time, and the server identifier of the candidate storage server corresponding to the first hotspot resource can be referred to as a first server identifier.

Wherein the first expiration time may be determined based on a statistical time point of the current period and a preset lifetime.

For example, the first expiration time is T0+ ttl, where T0 is a statistical time point of the current cycle, generally is a starting time point of the current cycle, ttl is a preset lifetime, and may be selected according to actual requirements, for example, ttl is 30 × T, and T is a duration of the cycle, for example, 10 seconds.

Therefore, for a certain hot spot resource, the aging time is 30 × 10 — 300 seconds.

The candidate storage servers corresponding to the first hot resource may be determined by using a correlation technique, for example, based on a resource identifier of the first hot resource, a hash algorithm may be used to determine a hash server, and the number of servers to be distributed (non-hash servers) may be determined according to the number of resource request messages of the first hot resource, in general, the larger the number of resource request messages is, the larger the number of servers to be distributed is, and the storage servers whose number is the number of the servers to be distributed are randomly selected as the non-hash servers from among the storage servers in the CDN hierarchy where the load balancing server is located. And the candidate storage server is composed of a hash server and a non-hash server. The corresponding server identification may then be taken as the first server identification.

For example, as shown in table 2, the first expiration time t0+ ttl corresponds to resource a. The first server identifier corresponding to the resource a includes: c 30, c 31, c 32; the first server identifier corresponding to the resource D includes: c 43 and c 44.

502. Acquiring existing second policy information, wherein the second policy information includes: and a second resource identifier, a second expiration time corresponding to the second resource identifier, and a plurality of second server identifiers, where the second resource identifier is the resource identifier of the second hotspot resource.

The current policy information is updatable, that is, the current policy information may be determined based on the existing policy information of the previous cycle, and the current policy information may also be used as the existing policy information of the next cycle. For the purpose of distinction, the existing policy information may be referred to as existing second policy information. The initial value of the existing second policy information may be null.

The determination process of the first policy information and the second policy information is similar, but resource request messages of different periods are adopted.

The first policy information and the second policy information may be both one or more sets, and accordingly, a first policy set and a second policy set may be formed.

As shown in fig. 6, it is assumed that the second policy information forms a second policy set, the second policy set includes three sets of second policy information, each set of second resource policy information includes a second resource identifier, and a second expiration time and a second server identifier corresponding to each second resource identifier, where the second resource identifiers are respectively represented by resource a, resource B, and resource C.

The first policy information is assumed to form a first policy set, the first policy set includes two sets of first policy information, each set of first policy information includes a first resource identifier, and a first expiration fact and a first server identifier corresponding to each first resource identifier, where the first resource identifiers are respectively represented by a resource a and a resource D.

After the first policy information and the second policy information are obtained, the second policy information may be updated by using the first policy information to obtain updated second policy information, and the updated second policy information may form an updated second policy set.

The update process of the second policy information may be as follows:

503. and traversing each first resource identifier in the first strategy information.

504. And judging whether the currently traversed first resource identifier belongs to the second strategy information, if so, executing 505, and otherwise, executing 509.

Based on the example of fig. 6, if the first resource identifier currently traversed is resource a, because the second policy information also includes resource a, 505 is executed; or, if the currently traversed first resource is identified as resource D, since the second policy information does not include resource D, execute 509.

505. And judging whether the number of the first server identifications corresponding to the currently traversed first resource identification is larger than the number of the second server identifications corresponding to the resource identification in the second strategy information, if so, executing 506, and otherwise, executing 507.

For the resource a, the number of first server identifiers corresponding to the resource a may be obtained from the first policy information (assuming that the number is 3), and for the resource a, the number of second server identifiers corresponding to the resource a may also be obtained from the second policy information (assuming that the number is 2), and since the number of first server identifiers is greater than the number of second server identifiers, 506 is executed.

506. And updating a second server identifier corresponding to the first resource identifier into a first server identifier corresponding to the first resource identifier in the second policy information.

For example, referring to fig. 6, in the updated second policy set, the server identifier corresponding to the resource a is updated to be the first server identifier.

507. And in the second policy information, keeping a second server identifier corresponding to the first resource identifier unchanged.

For example, for resource a, if the number of the first server identifiers corresponding to resource a is less than or equal to the number of the second server identifiers corresponding to resource B, the server identifier corresponding to resource a is kept as the second server identifier in the updated second policy set.

508. And updating a second expiration time corresponding to the first resource identifier to a first expiration time corresponding to the first resource identifier in the second policy information.

For example, referring to fig. 6, in the updated second policy set, the expiration time corresponding to the resource a is updated to the first expiration time.

In addition, for resource a, if the server identifier corresponding to resource a still keeps the second server identifier unchanged, the expiration time corresponding to resource a also needs to be updated to the first expiration time.

Based on 505 and 508, the update of the second policy information in the second policy set can be realized, and the server identifier is updated to a larger number of server identifiers, so that the dispersibility of the candidate servers can be improved, and the load balancing effect can be further improved.

509. And adding the first strategy information corresponding to the first resource identifier as new second strategy information.

For example, referring to fig. 6, since the resource D does not belong to the second policy information, the resource D and the corresponding first expiration time and the first server identifier are added in the updated second policy set.

Therefore, by adding the first policy information corresponding to the first resource identifier as the new second policy information, the new addition of the policy information can be realized, and the accuracy of the updated second policy information is ensured.

510. And after each first resource identifier in the first strategy information is traversed, obtaining updated second strategy information.

511. And traversing each piece of updated second strategy information.

512. And judging whether the updated second policy information of the current traversal is expired, if so, executing 513, and otherwise, repeatedly executing 514.

Whether the corresponding policy information is expired may be determined according to the expiration time in the updated second policy information that is currently traversed, for example, if the expiration time is earlier than or equal to the current time, specifically, if the current time t0 is 9 points, if the expiration time is 8 points and 50 points, it indicates that the policy information is expired; or, for another example, if the expiration time is 9: 5, then the expiration time is not expired.

513. Deleting the expired updated second policy information.

By deleting the overdue strategy information, the accuracy of the strategy information can be ensured, the storage space can be released, and the resource overhead is reduced.

514. And after each piece of updated second strategy information is traversed, obtaining the updated second strategy information to generate the current strategy information.

The second policy information after deleting the expired updated second policy information may be used as current policy information, and the current policy information may form a current policy set.

For example, referring to fig. 6, if the second expiration time corresponding to the resource B is earlier than the current time, the resource B expires, and the policy information corresponding to the resource B is deleted, so that the current policy set includes the policy information corresponding to the resource a, the resource C, and the resource D.

In this embodiment, because the first policy information is determined based on the resource request message in the first period, the resource identifier of the first hot resource may be determined, and the second policy information is the existing information, which includes the resource identifier of the second hot resource, so that the second policy information is updated based on the first policy information, and the expired policy information is deleted, thereby updating and deleting the policy information, ensuring the accuracy of the current policy information, and further improving the accuracy of load balancing.

In addition, the expiration time of the hotspot resource is determined based on the corresponding statistical time point and the preset survival time, and since the preset survival time is added to the expiration time, the delayed aging of the hotspot resource can be realized, for example, the corresponding strategy information is deleted after 5 minutes of delay (when the preset survival time is 10 × T). Under the condition that the hot spot resources are delayed to age, the problem of oscillation caused by switching of the resources between the hot spot resources and the non-hot spot resources can be reduced, and the stability of the system is improved. By adopting the first strategy information to update the second strategy information, strategy combination can be realized, the accuracy of the current strategy information is ensured, and the accuracy of the whole processing effect is further ensured.

Fig. 7 is a schematic diagram according to a ninth embodiment of the present disclosure, where the embodiment provides a network request path optimization apparatus 700, including: a receiving module 701, a determining module 702 and a sending module 703.

The receiving module 701 is configured to receive a resource request message for acquiring a target resource; the determining module 702 is configured to determine, if the target resource belongs to a hot spot resource in a current period, a target storage server in a plurality of candidate storage servers corresponding to the target resource, where the hot spot resource in the current period includes: a first hotspot resource determined based on a resource request message within a first period, and a second hotspot resource determined based on a resource request message within a second period, the first period being a previous period of the current period, the second period being earlier than the first period; the sending module 703 is configured to forward the resource request message to the target storage server.

In this embodiment, for a hot resource in a current period, a target storage server may be determined from multiple candidate storage servers, and compared with a method of fixing one target storage server, access pressure of the storage servers may be balanced, and a load balancing effect of the CDN is improved. In addition, in this embodiment, the hot spot resources in the current period include a first hot spot resource and a second hot spot resource, the first hot spot resource is determined based on the resource request message in the first period, and the second hot spot resource is determined based on the resource request message in the second period, because the second period is earlier than the first period, the hot spot resources in the current period are not only related to the resource request message in the previous period (i.e., the first period) of the current period, but also related to the resource request message in the earlier period (i.e., the second period), so that delayed aging of the hot spot resources is achieved.

In some embodiments, the resource request message includes a resource identifier of the target resource, and the apparatus 700 further includes: a policy obtaining module, configured to obtain current policy information, where the current policy information includes: the resource identification and the multiple server identifications corresponding to the resource identification comprise: the resource identifier of the first hot spot resource and the resource identifier of the second hot spot resource; the hot spot judging module is used for determining that the target resource belongs to the hot spot resource in the current period if the resource identifier of the target resource belongs to the current strategy information; and the candidate module is used for taking a plurality of servers indicated by a plurality of server identifications corresponding to the resource identification of the target resource in the current policy information as the plurality of candidate storage servers.

In some embodiments, the apparatus 700 further comprises: a first obtaining module, configured to determine first policy information based on the resource request message in the first period, where the first policy information includes: a first resource identifier, and a first expiration time and a first server identifier corresponding to the first resource identifier, where the first resource identifier is a resource identifier of the first hotspot resource, and the first server identifiers are multiple; a second obtaining module, configured to obtain existing second policy information, where the second policy information includes: a second resource identifier, and a second expiration time and a second server identifier corresponding to the second resource identifier, where the second resource identifier is a resource identifier of the second hotspot resource, and the number of the second server identifiers is multiple; the updating module is used for updating the second strategy information based on the first strategy information so as to obtain updated second strategy information; and the generating module is used for deleting the expired updated second policy information so as to generate the current policy information.

In some embodiments, the update module is further to: if the first resource identifier belongs to the second policy information, executing: updating a second expiration time corresponding to the first resource identifier in the second policy information to a first expiration time corresponding to the first resource identifier; and if the number of the first server identifications corresponding to the first resource identifications is larger than the number of the second server identifications corresponding to the first resource identifications in the second policy information, updating the second server identifications corresponding to the first resource identifications in the second policy information to be the first server identifications.

In this embodiment, the server identifiers are updated to a larger number of server identifiers, so that the dispersibility of the candidate servers can be improved, and the load balancing effect can be further improved.

In some embodiments, the update module is further to: and if the first resource identifier does not belong to the second strategy information, adding the first strategy information corresponding to the first resource identifier as new second strategy information.

In this embodiment, by adding the first policy information corresponding to the first resource identifier as the new second policy information, the new addition of policy information can be realized, and the accuracy of the updated second policy information is ensured.

In some embodiments, the time point of generating the current policy information is a current time, the updated second policy information includes an updated second expiration time, and the deleting module is further configured to: and if the updated second expiration time is earlier than or equal to the current time, deleting the corresponding updated second policy information.

In the embodiment, the accuracy of the policy information can be ensured by deleting the outdated policy information, the storage space can be released, and the resource overhead is reduced.

In some embodiments, the time point of generating the current policy information is a current time, and the first expiration time is a sum of the current time and a preset lifetime.

In this embodiment, the expiration time of the hotspot resource is determined based on the corresponding statistical time point and the preset survival time, and since the preset survival time is added to the expiration time, delayed aging of the hotspot resource can be realized, for example, the corresponding policy information is deleted after 5 minutes of delay (when the preset survival time is 10 × T). Under the condition that the hot spot resources are delayed to age, the problem of oscillation caused by switching of the resources between the hot spot resources and the non-hot spot resources can be reduced, and the stability of the system is improved.

In some embodiments, the determining module is further configured to: and randomly selecting one candidate storage server from the plurality of candidate storage servers as the target storage server.

In the embodiment, a candidate storage server is selected as the target storage server in a random mode, so that the randomness can be improved, and the resource request message can be balanced better.

It is to be understood that in the disclosed embodiments, the same or similar elements in different embodiments may be referenced.

It is to be understood that "first", "second", and the like in the embodiments of the present disclosure are used for distinction only, and do not indicate the degree of importance, the order of timing, and the like.

The present disclosure also provides an electronic device, a readable storage medium, and a computer program product according to embodiments of the present disclosure.

FIG. 8 illustrates a schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 8, the electronic device 800 includes a computing unit 801 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM)802 or a computer program loaded from a storage unit 808 into a Random Access Memory (RAM) 803. In the RAM803, various programs and data required for the operation of the electronic apparatus 800 can also be stored. The calculation unit 801, the ROM802, and the RAM803 are connected to each other by a bus 804. An input/output (I/O) interface 805 is also connected to bus 804.

A number of components in the electronic device 800 are connected to the I/O interface 805, including: an input unit 806, such as a keyboard, a mouse, or the like; an output unit 807 such as various types of displays, speakers, and the like; a storage unit 808, such as a magnetic disk, optical disk, or the like; and a communication unit 809 such as a network card, modem, wireless communication transceiver, etc. The communication unit 809 allows the electronic device 800 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunication networks.

Computing unit 801 may be a variety of general and/or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various dedicated Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and the like. The calculation unit 801 executes the respective methods and processes described above, such as the network request path optimization method. For example, in some embodiments, the network request path optimization method may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program can be loaded and/or installed onto the electronic device 800 via the ROM802 and/or the communication unit 809. When loaded into RAM803 and executed by the computing unit 801, a computer program may perform one or more steps of the network request path optimization method described above. Alternatively, in other embodiments, the computing unit 801 may be configured to perform the network request path optimization method in any other suitable manner (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), Complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable load balancing apparatus such that the program codes, when executed by the processor or controller, cause the functions/acts specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The Server can be a cloud Server, also called a cloud computing Server or a cloud host, and is a host product in a cloud computing service system, so as to solve the defects of high management difficulty and weak service expansibility in the traditional physical host and VPS service ("Virtual Private Server", or simply "VPS"). The server may also be a server of a distributed system, or a server incorporating a blockchain.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, and the present disclosure is not limited herein.

The above detailed description should not be construed as limiting the scope of the disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.

Claims

1. A network request path optimization method comprises the following steps:

receiving a resource request message for acquiring a target resource;

if the target resource belongs to a hot spot resource in the current period, determining a target storage server in a plurality of candidate storage servers corresponding to the target resource, wherein the hot spot resource in the current period comprises: a first hot spot resource and a second hot spot resource, the first hot spot resource being determined based on a resource request message within a first period, the second hot spot resource being determined based on a resource request message within a second period, the first period being a previous period of the current period, the second period being earlier than the first period;

and forwarding the resource request message to the target storage server.

2. The method of claim 1, wherein the resource request message includes a resource identifier of the target resource, the method further comprising:

acquiring current policy information, wherein the current policy information comprises: the resource identification and the multiple server identifications corresponding to the resource identification comprise: the resource identifier of the first hot spot resource and the resource identifier of the second hot spot resource;

if the resource identifier of the target resource belongs to the current strategy information, determining that the target resource belongs to a hot resource in the current period;

and taking a plurality of servers indicated by a plurality of server identifications corresponding to the resource identification of the target resource in the current policy information as the plurality of candidate storage servers.

3. The method of claim 2, further comprising:

determining first policy information based on the resource request message within the first period, the first policy information comprising: a first resource identifier, and a first expiration time and a first server identifier corresponding to the first resource identifier, where the first resource identifier is a resource identifier of the first hotspot resource, and the first server identifiers are multiple;

acquiring existing second policy information, wherein the second policy information includes: a second resource identifier, and a second expiration time and a second server identifier corresponding to the second resource identifier, where the second resource identifier is a resource identifier of the second hotspot resource, and the number of the second server identifiers is multiple;

updating the second policy information based on the first policy information to obtain updated second policy information;

deleting the expired updated second policy information to generate the current policy information.

4. The method of claim 3, wherein the updating the second policy information based on the first policy information comprises:

if the first resource identifier belongs to the second policy information, executing:

updating a second expiration time corresponding to the first resource identifier in the second policy information to a first expiration time corresponding to the first resource identifier;

and if the number of the first server identifications corresponding to the first resource identifications is larger than the number of the second server identifications corresponding to the first resource identifications in the second policy information, updating the second server identifications corresponding to the first resource identifications in the second policy information to be the first server identifications.

5. The method of claim 3, wherein the updating the second policy information based on the first policy information comprises:

and if the first resource identifier does not belong to the second strategy information, adding the first strategy information corresponding to the first resource identifier as new second strategy information.

6. The method of claim 3, wherein the point in time at which the current policy information is generated is a current time, the updated second policy information includes an updated second expiration time, and the deleting the expired updated second policy information includes:

and if the updated second expiration time is earlier than or equal to the current time, deleting the corresponding updated second policy information.

7. The method of claim 3, wherein the time point of generating the current policy information is a current time, and the first expiration time is a sum of the current time and a preset time-to-live.

8. The method of any of claims 1-7, wherein the determining a target storage server among a plurality of candidate storage servers corresponding to the target resource comprises:

and randomly selecting one candidate storage server from the plurality of candidate storage servers as the target storage server.

9. A network request path optimization apparatus, comprising:

a receiving module, configured to receive a resource request message for acquiring a target resource;

a determining module, configured to determine a target storage server in a plurality of candidate storage servers corresponding to the target resource if the target resource belongs to a hot resource in a current period, where the hot resource in the current period includes: a first hot spot resource and a second hot spot resource, the first hot spot resource being determined based on a resource request message within a first period, the second hot spot resource being determined based on a resource request message within a second period, the first period being a previous period of the current period, the second period being earlier than the first period;

a sending module, configured to forward the resource request message to the target storage server.

10. The apparatus of claim 9, wherein the resource request message includes a resource identifier of the target resource, the apparatus further comprising:

a policy obtaining module, configured to obtain current policy information, where the current policy information includes: the resource identification and the multiple server identifications corresponding to the resource identification comprise: the resource identifier of the first hot spot resource and the resource identifier of the second hot spot resource;

the hot spot judging module is used for determining that the target resource belongs to the hot spot resource in the current period if the resource identifier of the target resource belongs to the current strategy information;

and the candidate module is used for taking a plurality of servers indicated by a plurality of server identifications corresponding to the resource identification of the target resource in the current policy information as the plurality of candidate storage servers.

11. The apparatus of claim 10, further comprising:

a first obtaining module, configured to determine first policy information based on the resource request message in the first period, where the first policy information includes: a first resource identifier, and a first expiration time and a first server identifier corresponding to the first resource identifier, where the first resource identifier is a resource identifier of the first hotspot resource, and the first server identifiers are multiple;

a second obtaining module, configured to obtain existing second policy information, where the second policy information includes: a second resource identifier, and a second expiration time and a second server identifier corresponding to the second resource identifier, where the second resource identifier is a resource identifier of the second hotspot resource, and the number of the second server identifiers is multiple;

the updating module is used for updating the second strategy information based on the first strategy information so as to obtain updated second strategy information;

and the generating module is used for deleting the expired updated second policy information so as to generate the current policy information.

12. The apparatus of claim 11, wherein the update module is further to:

13. The apparatus of claim 11, wherein the update module is further to:

14. The apparatus of claim 11, wherein the point in time at which the current policy information is generated is a current time, the updated second policy information comprises an updated second expiration time, and the deletion module is further configured to:

15. The apparatus of claim 11, wherein the time point of generating the current policy information is a current time, and the first expiration time is a sum of the current time and a preset time-to-live.

16. The apparatus of any of claims 9-15, wherein the means for determining is further configured to:

17. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-8.

18. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any one of claims 1-8.

19. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-8.