CN113596109A

CN113596109A - Service request operation method, system, device, equipment and storage medium

Info

Publication number: CN113596109A
Application number: CN202110770113.XA
Authority: CN
Inventors: 靳瑞涛; 尉惠敏; 李雪梅; 王路; 冯雅靖
Original assignee: Shanghai Pudong Development Bank Co Ltd
Current assignee: Shanghai Pudong Development Bank Co Ltd
Priority date: 2021-07-07
Filing date: 2021-07-07
Publication date: 2021-11-02
Anticipated expiration: 2041-07-07
Also published as: CN113596109B

Abstract

The application relates to a service request operation method, a system, a device, equipment and a storage medium, which are characterized in that a target service queue corresponding to each service request is determined according to the actual success rate of each service request in the current pressure time period of the current service queue and the success rate threshold corresponding to the current pressure time period by acquiring the actual success rate of each service request in the current pressure time period and the success rate threshold corresponding to the current pressure time period; the current service queue is any one of a main service queue, a semi-fusing queue and a fusing queue, and the pressure time period represents the time length of the maximum pressure which can be borne by the current service queue when the service request is operated. The method ensures the normal operation of the service request, and prevents the service loss caused by accidental conditions because the service request is directly fused instead of being directly fused.

Description

Service request operation method, system, device, equipment and storage medium

Technical Field

The present application relates to the field of network technologies, and in particular, to a method, a system, an apparatus, a device, and a storage medium for service request operation.

Background

In the internet system, when the response of the downstream service is slow or fails due to excessive access pressure, the upstream service may temporarily cut off the call to the downstream service in order to protect the availability of the entire system. The fusing mechanism plays a role in protecting driving and navigating in a system with a large number of users.

However, the fusing mechanism in the related art is based on directly fusing after the service request fails to operate, and there is a possibility that the fusing is directly fused due to accidental extreme conditions, thereby causing loss.

Disclosure of Invention

The embodiment of the application provides a service request operation method, a system, a device, equipment and a storage medium, which can directly fuse a service request and prevent service loss caused by accidental conditions.

In a first aspect, an embodiment of the present application provides a method for service request operation, where the method includes:

acquiring the actual success rate of each service request in the current pressure time period of the current service queue; the current service queue is any one of a main service queue, a semi-fusing queue and a fusing queue; the pressure time period represents the time length of the maximum pressure which can be borne by the current service queue when the service request is operated;

acquiring a success rate threshold corresponding to the current pressure time period;

determining a target service queue corresponding to each service request according to the actual success rate of each service request in the current pressure time period and the success rate threshold of the current pressure time period

In one embodiment, obtaining the success rate threshold corresponding to the current pressure time period includes:

acquiring a first average success rate of the first pressure time period and a second average success rate of the second pressure time period by executing the average success rate calculation step; the first pressure time period is the last pressure time period of the current pressure time period, and the second pressure time period is the last pressure time period of the last pressure time period;

acquiring a minimum error between the first pressure time period and the second pressure time period according to the first average success rate and the second average success rate;

and determining the sum of the minimum error and the first average success rate as a success rate threshold value of the current pressure time period.

In one embodiment, obtaining the minimum error between the first pressure time period and the second pressure time period according to the first average success rate and the second average success rate includes:

constructing a square loss function according to the first average success rate and the second average success rate;

and optimizing the square loss function through a preset optimization algorithm, and determining the minimum error between the first pressure time period and the second pressure time period.

In one embodiment, the average success rate calculating step includes:

equally dividing any pressure time period into N sub-time periods, and acquiring the total request number, the request success number and the weight of each sub-time period in each sub-time period; n is a positive even number;

calculating to obtain N/2 reference average success rates according to the total request number in each sub-time period, the request success number in each sub-time period and the weight in each sub-time period;

and determining the average success rate of the pressure time period according to the N/2 reference average success rates.

In one embodiment, if the current service queue is the main service queue; determining a target service queue corresponding to each service request according to the actual success rate of each service request in the current pressure time period and the success rate threshold of the current pressure time period, wherein the determining comprises the following steps:

comparing the actual success rate of each service request in the current pressure time period of the main service queue with the success rate threshold value of the current pressure time period of the main service queue;

determining a target service queue of the service request with the actual success rate greater than the success rate threshold value as a main service queue;

and determining the target service queue of the service request with the actual success rate less than or equal to the success rate threshold value as a semi-fusing queue.

In one embodiment, if the current service queue is a semi-fusible queue; determining a target service queue corresponding to each service request according to the actual success rate of each service request in the current pressure time period and the success rate threshold of the current pressure time period, wherein the determining comprises the following steps:

comparing a first actual success rate of each service request in the current pressure time period of the semi-fusing queue with a first success rate threshold of the current pressure time period of the semi-fusing queue; comparing the second actual success rate of each service request in the last pressure time period of the semi-fusing queue with a second power forming threshold value of the last pressure time period of the semi-fusing queue;

determining a target service queue of the service request with the first actual success rate greater than a first power forming threshold value and the second actual success rate greater than a second power forming threshold value as a main service queue;

determining a target service queue of the service request with the first actual success rate smaller than a first power forming threshold value and the second actual success rate smaller than a second power forming threshold value as a fusing queue;

and determining the target service queue of the service request with the first actual success rate greater than the first power forming threshold value or the second actual success rate greater than the second power forming threshold value as a semi-fusing queue.

In one embodiment, if the current service queue is a fused queue, acquiring an actual success rate of each service request in the current pressure time period of the current service queue includes:

releasing a preset number of service requests from the fusing queue to run according to a preset interval time; the preset quantity is determined according to the quantity of the service requests of the main service queue in the current time period;

and acquiring the actual success rate of the released service request.

In one embodiment, determining a target service queue corresponding to each service request according to an actual success rate of each service request in a current pressure time period and a success rate threshold of the current pressure time period includes:

and if the actual success rates of the preset number of service requests in the pressure time periods of two continuous fusing queues are greater than the success rate threshold of the latest pressure time period of the semi-fusing queue, determining the target service queue of the preset number of service requests as the semi-fusing queue.

In a second aspect, an embodiment of the present application provides a service request operating system, where the system includes: the system comprises a management server and at least one service processing link, wherein each service processing link comprises an application server, a database and a network server;

a management server for executing the service request operation method of any one of claims 1 to 8;

the application server is used for forwarding the plurality of service requests sent by the user side to the corresponding network server and storing the service information carried in each service request into the database; receiving a service request result returned by the corresponding network server, and storing the service request result into a database;

and the network server is used for forwarding each service request sent by the corresponding application server to servers in other local area networks and sending service request results returned by the servers in other local area networks to the application server.

In one embodiment, the service processing link comprises a main service queue, a semi-fusing queue and a fusing queue;

the main service queue comprises a first application server, a database and a first network server; the main service queue is used for running service requests in a normal state;

the semi-fusing queue comprises a second application server, a database and a second network server; the semi-fusing queue is used for running service requests in a semi-fusing state;

the fusing queue comprises a third application server, a database and a third network server; and the fuse queue is used for service requests in a fuse state.

In a third aspect, an embodiment of the present application provides a service request operation apparatus, where the apparatus includes:

the first acquisition module is used for acquiring the actual success rate of each service request in the current pressure time period of the current service queue; the current service queue is any one of a main service queue, a semi-fusing queue and a fusing queue; the pressure time period represents the time length of the maximum pressure which can be borne by the current service queue when the service request is operated;

the second acquisition module is used for acquiring a success rate threshold corresponding to the current pressure time period;

and the processing module is used for determining a target service queue corresponding to each service request according to the actual success rate of each service request in the current pressure time period and the success rate threshold value of the current pressure time period.

In a third aspect, an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the method of the first aspect when executing the computer program.

In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method of the first aspect.

According to the service request operation method, the system, the device, the equipment and the storage medium provided by the embodiment of the application, the actual success rate of each service request in the current pressure time period of the current service queue and the success rate threshold corresponding to the current pressure time period are obtained, and then the target service queue corresponding to each service request is determined according to the actual success rate of each service request in the current pressure time period and the success rate threshold of the current pressure time period; the current service queue is any one of a main service queue, a semi-fusing queue and a fusing queue, and the pressure time period represents the time length of the maximum pressure which can be borne by the current service queue when the service request is operated. In the method, after the actual success rate of the service request of any service queue is obtained, the actual success rate of the service request of the service queue is determined by combining the success rate threshold value of the current pressure time period, and the target service queue is selected to run each service request.

Drawings

FIG. 1a is a schematic diagram of a service request execution system in one embodiment;

FIG. 1b is a schematic diagram of a system for service request operation according to an embodiment;

FIG. 2 is a flow diagram illustrating a method for service request execution according to an embodiment;

FIG. 3 is a flow diagram illustrating a method for service request execution according to one embodiment;

FIG. 4 is a flow diagram illustrating a method for service request execution according to one embodiment;

FIG. 5 is a flowchart illustrating a method for service request execution according to an embodiment;

FIG. 6 is a flow diagram illustrating a method for service request execution according to one embodiment;

FIG. 7 is a block diagram showing the structure of a service request execution apparatus according to an embodiment;

FIG. 8 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application more clearly understood, the embodiments of the present application are described in further detail below with reference to the accompanying drawings and the embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the embodiments of the application and are not intended to limit the embodiments of the application.

The following describes in detail the technical solutions of the present application and how the technical solutions of the present application solve the above technical problems by embodiments and with reference to the drawings. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application.

Before explaining the technical scheme of the service request operation method provided by the embodiment of the present application, a service request operation system provided by the embodiment of the present application is explained first. The service request operation method provided by the embodiment of the application is applied to operation in a service request operation system.

As shown in fig. 1a, the service request running system includes a management server 10 and at least one service processing link 20, each service processing link 20 includes an application server 201, a database 202 and a network server 203;

the management server 10 is configured to execute the service request operation method provided in the embodiment of the present application;

the application server 201 is configured to forward a plurality of service requests sent by a user side to a corresponding network server, and store service information carried in each service request in a database; receiving a service request result returned by the corresponding network server, and storing the service request result into a database;

the network server 203 is configured to forward each service request sent by the corresponding application server 201 to a server in another local area network, and send a service request result returned by the server in the other local area network to the application server 201.

The service request operation system is constructed by adopting a Software Defined Network (SDN) architecture, wherein the SDN is used as a technology for realizing Network reconfiguration by improving Network programmability, converting the Network into Software, and carrying out centralized control on a bottom layer communication Network.

Based on the SDN architecture idea, the embodiment of the application divides the original bottom communication network into a control layer (management server) and a forwarding layer (at least one service link), thereby realizing the decoupling between communication networks and the centralized control of the forwarding layer.

When a service request runs in a service request running system, please refer to fig. 1a, a user side in an external network in fig. 1a sends a service request to an internal network through a firewall, and after receiving the service request, an application server 201 in the internal network forwards the service request to a network server 203, and stores service information carried in each service request in a database 202. The service information carried in the service request includes, but is not limited to, a service name, a URL (Uniform Resource Locator) of the request, and message time of the request. Where the URL is a uniform resource locator, i.e., a network address, of the WWW.

For the network server 203, after receiving the service request forwarded by the application server 201, the network server 203 will continue to forward each service request sent by the application server 201 to a server in another local area network, and will receive a service request result returned by a server in another local area network. The service request result may be message information returned by servers of other local area networks or a transmission failure prompt message, where the received message information indicates that the service request transaction is successful, and if the network server 203 receives the transmission failure prompt message, the service request transaction is failed.

Then, the web server 203 returns the received service request result to the application server 201, and the application server 201 receives the service request result and stores the service request result in the database 202. The application server 201 stores the service request result in the database 202, that is, the message information or the transmission failure prompt message is recorded in the database 202, and the response time of the service request is recorded.

Thus, the service request is sent from the user side to the application server 201, and the information related to the result of the service request is stored, that is, the information sending process is completed.

Each service link 20 is responsible for the above information sending process, and the management server 10 monitors the information stored in the database of each service request to implement the service request operation method provided in the embodiment of the present application according to the monitoring result, and the specific process of the service request operation method will be described in detail in the following embodiments.

It should be noted that, the database in fig. 1a is exemplified by a database shared by multiple links, and in practical application, one database may be provided for each link, and each database is controlled by the management server 10.

In a network in a conventional IT architecture, a physical infrastructure (such as a switch and a router) is used to establish a connection and operate normally, and after the network is deployed and brought online according to a service demand, if the service demand changes, the configuration on corresponding equipment needs to be modified again, which is very tedious. Compared with a network in a traditional IT framework, the SDN network framework is used for constructing the service request operation system in the embodiment of the application, the control right on the network equipment is separated, so that the management server can manage and configure the network service from a centralized position, the dependence on the underlying network equipment is not needed, any network routing and transmission rules which are to be realized can be defined, and the flexible and intelligent network can be built.

As shown in fig. 1b, the functions of the service processing links are further explained based on fig. 1a by taking three service processing links as an example.

In fig. 1b, traffic handling link 20 includes a main traffic queue 20a, a semi-fused queue 20b, and a fused queue 20 c.

The main service queue 20a includes a first application server 201a, a database 202, and a first network server 203 a; and the main service queue 20a is used for running service requests in a normal state;

the semi-blown queue 20b comprises a second application server 201b, a database 202 and a second network server 203 b; and the semi-blown queue 20b is used to run service requests in a semi-blown state;

the fusing queue 20c comprises a third application server 201c, a database 202 and a third network server 203 c; and the blown queue 20c is for a blown state of the service request.

In this embodiment, the management server 10 at the management end schedules 3 application servers: a first application server 201a, a second application server 201b, a third application server 201 c; a database 202; and 3 network servers: a first web server 203a, a second web server 203b, and a third web server 203 c.

A first service processing link, which is marked as a "main service queue 20 a", is formed by the first application server 201a, the database 202 and the first network server 203 a; the main service queue 20a is mainly responsible for daily service request operation.

A second service processing link, which is marked as a semi-fusible queue 20b, is formed by a second application server 201b, a database 202 and a second network server 203 b; semi-blown queue 20b is primarily responsible for service request execution in the "semi-blown" state.

A third service processing link, which is marked as a "fuse queue 20 c", is formed by a third application server 201c, a database 202, and a third network server 203 c; the fuse queue 20c is primarily responsible for service request execution in the "fuse" state.

In this embodiment, taking three service processing links as an example, the three service processing links are divided into detailed service running queues, and different service running queues process service requests in different states, so as to undertake different functions, and select different services to run in a targeted manner, thereby ensuring normal running of the services.

Based on the service request operation system provided in the foregoing embodiment, a technical solution of the service request operation method provided in the embodiment of the present application is described.

In the following description, when a service request operation method provided in the embodiment of the present application is described, an execution subject is the management server 10 in the service request operation system as an example.

In an embodiment, a method for operating service requests is provided, where the embodiment relates to a specific process in which the management server 10 compares an actual success rate of each service request in a current pressure time period of any one of the main service queue, the semi-fusing queue, and the fusing queue with a success rate threshold corresponding to the current pressure time period, and then determines a target service queue to be operated by each service request; as shown in fig. 2, this embodiment includes the steps of:

s101, acquiring the actual success rate of each service request in the current pressure time period of the current service queue; the current service queue is any one of a main service queue, a semi-fusing queue and a fusing queue; the pressure time period represents the length of time that the current service queue can withstand the maximum pressure while running the service request.

The current pressure time period represents the time length of the maximum pressure which can be borne by the current service queue when the service request is operated, and is equivalent to the pressure time period which can be borne by the system under the extreme condition. It should be noted that the current pressure time period only refers to the current pressure time period, and for the three queues, namely the main service queue, the semi-fusing queue, and the fusing queue, there are their own current pressure time periods, and the own current pressure time periods of the three queues may be the same time length or different time lengths.

Therefore, when referring to the current pressure time period in the embodiment of the present application, it may be the current pressure time period of any queue, and it is subject to practical definition and description. For example, if the current service queue is the main service queue, the current pressure time period of the current service queue refers to the current pressure time period of the main service queue; the same applies to the other queues, which must be defined for the current pressure period to be considered as the current pressure period for that queue.

The pressure time period may be determined by calculating the data transmission amount of the server in an extreme case, and optionally, the pressure time period is calculated by the number of temporary ports sent to the network server by the application server of the current traffic queue and the maximum throughput processed by the server.

Specifically, assuming that the number of temporary ports from the application server to the web server is 50000, and the maximum number of Transactions Per Second (TPS) processed by the server is 5000, the time period is 10 seconds. Wherein the network server request timeout default is 60 seconds.

The actual success rate of each service request refers to the probability that the actual operation of each service request by the current service queue is successful in the current pressure time period.

As can be seen from the introduction of the above embodiment of the service request operation system, in the information sending process of each service request, the application server stores the relevant information of each service request in the database, and the management server monitors the database, so that the management server can directly obtain the total number of service requests and the number of service requests that have successfully operated in the current pressure time period of the current service queue and the number of service requests that have failed to operate according to the records in the database, thereby calculating the actual success rate of each service request in the current pressure time period of the current service queue.

And S102, acquiring a success rate threshold corresponding to the current pressure time period.

In this step, a success rate threshold corresponding to the current pressure time period needs to be obtained, where the current pressure time period may still be the current pressure time period of the current service queue, for example, the current service queue is a main service queue, and the current pressure time period in this step may continue to be the current pressure time period of the main service queue, or the current service queue is a semi-fusing queue, and the current pressure time period in this step may continue to be the current pressure time period of the semi-fusing queue;

however, the current pressure time period here may still not be the current pressure time period of the current service queue, for example, the current service queue is a blown queue, and the current pressure time period of this step may continue to be the current pressure time period of the semi-blown queue. The embodiments of the present application do not limit this.

Because the service request operates in a public network environment, the problems of data packet loss, data delay and the like may be caused by network jitter, and for each service request which should be in normal operation, semi-fusing or fusing, the judgment elements are mainly as follows: and whether the success rate of the service meets the requirement or not within a certain time period. Therefore, the embodiment of the present application needs to obtain the success rate threshold corresponding to the current pressure time period, so as to conveniently and accurately determine which queue of the three service queues, namely the main service queue, the semi-fusing queue and the fusing queue, each service request operates in the following process.

In an embodiment, the step may be to obtain the success rate thresholds of the three service queues, namely the main service queue, the semi-fuse queue and the fuse queue, in real time, and to obtain the success rate threshold corresponding to each pressure time period of each service queue by recalculation, so as to dynamically determine the success rate thresholds of different pressure time periods. It should be noted that, the success rate thresholds corresponding to the current pressure time periods of different service queues may be the same or different, specifically, the actual calculation is the criterion, and the success rate thresholds of different pressure time periods of the same service queue may also be the same or different.

The success rate threshold corresponding to the current pressure time period may be determined according to a machine learning model constructed in advance. For example, it is determined which service queue the current pressure time period refers to, then the duration of the current pressure time period is calculated according to the number of ports of the application server of the service queue, and then the duration of the current pressure time period is input into the pre-constructed machine learning model, so that the success rate threshold corresponding to the current pressure time period can be obtained.

S103, determining a target service queue corresponding to each service request according to the actual success rate of each service request in the current pressure time period and the success rate threshold of the current pressure time period.

And comparing the actual success rate with the success rate threshold value of the current pressure time period according to the actual success rate of the obtained service requests in the current pressure time period so as to determine the target service queue corresponding to each service request.

For example, the actual success rate of the service request in the current pressure time period may be compared with the success rate threshold of the current pressure time period of the current service queue, or the actual success rate of each service request in the current pressure time period and the actual success rate of each service request in other pressure time periods may be compared with the success rate threshold of the corresponding time period together, and after the comparison result is obtained, the service queue in which each service request should operate is determined based on the comparison result.

For example, the target service queues corresponding to different comparison results may be queried according to a pre-constructed relationship mapping table; or, a deep learning network model may be adopted to take the actual success rate of each service request in the current pressure time period and the success rate threshold of the current pressure time period as inputs, so as to obtain a target service queue corresponding to each service request.

According to the service request operation method provided by the embodiment of the application, the actual success rate of each service request in the current pressure time period of the current service queue and the success rate threshold corresponding to the current pressure time period are obtained, and then the target service queue corresponding to each service request is determined according to the actual success rate of each service request in the current pressure time period and the success rate threshold of the current pressure time period; the current service queue is any one of a main service queue, a semi-fusing queue and a fusing queue, and the pressure time period represents the time length of the maximum pressure which can be borne by the current service queue when the service request is operated. In the method, after the actual success rate of the service request of any service queue is obtained, the actual success rate of the service request of the service queue is determined by combining the success rate threshold value of the current pressure time period, and the target service queue is selected to run each service request. In addition, the success rate threshold value of the current time period in the embodiment of the application is required to be acquired every time and belongs to a dynamic threshold value, so that the operation target service queue of each service request is determined according to the success rate threshold value corresponding to the current time period, and the operation mode of each service request is more accurate.

On the basis of the above embodiment, an embodiment is provided to explain an acquisition process of a success rate threshold corresponding to a current pressure time period. As shown in fig. 3, this embodiment includes the steps of:

s201, acquiring a first average success rate of a first pressure time period and a second average success rate of a second pressure time period by executing an average success rate calculation step; the first pressure time period is a previous pressure time period of the current pressure time period, and the second pressure time period is a previous pressure time period of the previous pressure time period.

The average success rate refers to the average success rate of the service request of one pressure time period, and can be obtained according to the average success rate calculation step for each pressure time period.

For example, the first average success rate of the first pressure period, which is the last pressure period of the current pressure period, may be obtained by performing the average success rate calculation step.

The second average success rate of the second pressure period, which is the last pressure period of the last pressure period, that is, the last pressure period of the current pressure period, may be obtained by performing the average success rate calculation step.

The process of the average success rate calculation step will be explained first. Optionally, the average success rate calculating step includes: equally dividing any pressure time period into N sub-time periods, and acquiring the total request number, the request success number and the weight of each sub-time period in each sub-time period; n is a positive even number; calculating to obtain N/2 reference average success rates according to the total request number in each sub-time period, the request success number in each sub-time period and the weight in each sub-time period; and determining the average success rate of the pressure time period according to the N/2 reference average success rates.

Based on the network architecture of the service request operation system, the pressure time period T is divided into N (N is an even number) sub-time periods with the time length T according to the previously calculated pressure time period T (any pressure time period applicable to any service queue).

Let the total request of each sub-period tThe quantity is C, the total request quantity of N sub-time periods t is C₁、C₂、C₃、……C_x－1、C_x、……C_N-1、C_N(ii) a If the request success number of each sub-period t is S, the request success numbers of N sub-periods t are S₁、S₂、S₃、……S_x－1、S_x、……S_N-1、S_N(ii) a If the weight of each sub-period t is I, the weights of N sub-periods t are I₁、I₂、I₃、……I_x－1、I_x、……I_N-1、I_N。

And correspondingly calculating to obtain N/2 reference average success rates according to N/2 preset formulas respectively based on the total request number in each sub-time period, the request success number in each sub-time period and the weight in each sub-time period.

Specifically, from 1 to N/2, the first reference average success rate is X₁Second reference average success rate is X₂,.., the reference average success rate of the Nth/2 th is X_N/2；

Then, let the first reference average success rate be X₁The formula (1) is:

in the formula (1), the adjacent numerical value with the difference value of 1, taking C as an example, is C when the difference value of the adjacent numerical values is 1₁To C₂Then C₂To C₃、C₃To C₄Up to C_N-1To C_NThen, multiple calculation is performed, and finally, the average is calculated to obtain the first reference average success rate X₁。

And the first reference average success rate is X2, formula (2) is:

in the formula (2), the adjacent numerical value with the difference of 2, taking C as an example, is C when the difference of the adjacent numerical values is 2₁To C₃Then C₂To C₄、C₃To C₅Up to C_NTo C_N+2Then, multiple calculation is performed, and finally, the average is calculated to obtain a second reference average success rate X₂。

By analogy, the third reference average success rate is X₃It is the adjacent value with the difference of 3, taking C as an example, the difference of the adjacent values is 3, that is C₁To C₄Then C₂To C₅、C₃To C₆Up to C_N-3To C_NThen, multiple calculation is carried out, and finally, the average is calculated to obtain a third reference average success rate X₃。

Then the N/2 th reference average success rate is X_N/2The formula (3) is:

in the formula (3), the adjacent numerical value with the difference of N/2-1, taking C as an example, the difference of the adjacent numerical values is N/2-1, namely C₁To C_N/2Then C₂To C_N/2+1、C₃To C_N/2+2Up to C_N/2To C_NThen, multiple calculation is performed, and finally, the average is calculated to obtain a second reference average success rate X_N/2。

N/2 reference average success rates can be calculated according to the formula, and then the average success rate of the pressure time period T can be determined according to the N/2 reference average success rates.

For example, the average success rate over the pressure period T

Comprises the following steps:

s202, acquiring a minimum error between the first pressure time period and the second pressure time period according to the first average success rate and the second average success rate.

After the first average success rate and the second average success rate are determined, the minimum error between the first pressure time period and the second pressure time period is further acquired.

In one embodiment, the minimum error between the first pressure time period and the second pressure time period may be obtained in a deep learning network in which the input values of the first average success rate and the second average success rate are preset.

In yet another embodiment, a square loss function may be constructed according to the first average success rate and the second average success rate; and optimizing the square loss function through a preset optimization algorithm, and determining the minimum error between the first pressure time period and the second pressure time period.

Wherein, the normal error between the first pressure time period and the second pressure time period can be obtained according to the first pressure time period and the second pressure time period, for example, the normal error is represented by E, and the first average success rate of the first pressure time period is

The second average success rate of the second pressure period is

Then

Further, in order to reduce the distortion of the data, the formula is changed as follows:

based on this, a square loss function is constructed:

the square loss function Q, i.e. the errorAnd (3) optimizing the difference sum of squares function by using a least square method to obtain a final Q value which is the sum of squares of the errors, and obtaining the minimum error epsilon after the sum of squares is followed:

and S203, determining the sum of the minimum error and the first average success rate as a success rate threshold value of the current pressure time period.

After the minimum error epsilon is obtained, the minimum error epsilon and the first average success rate are calculated

Is determined as a success rate threshold for the current pressure period, i.e.

It should be noted that the value of X is not more than 100 at the highest and is more than 0 at the lowest.

In the embodiment of the application, the minimum error is determined according to the average success rate of the last pressure time period and the last pressure time period of the current pressure time period, and the sum of the minimum error and the last pressure time period is the success rate threshold of the current pressure time period.

By determining the success rate threshold of the current pressure time period in the above embodiment, the success rate threshold of the current pressure time period of any one of the main service queue, the semi-fused queue, and the fused queue may be determined.

And judging the actual success rate of each service request in the current pressure time period based on the success rate threshold value of the current time period of each service queue so as to determine a target service queue corresponding to each service request.

As shown in fig. 4, an embodiment is provided to explain a process of determining a target service queue corresponding to each service request when a current service queue is a main service queue, where the embodiment includes the following steps:

s301, comparing the actual success rate of each service request in the current pressure time period of the main service queue with the success rate threshold value of the current pressure time period of the main service queue.

S302, the target service queue of the service request with the actual success rate larger than the success rate threshold is determined as the main service queue.

And S303, determining the target service queue of the service request with the actual success rate less than or equal to the success rate threshold value as a semi-fusing queue.

The main service queue is a queue in charge of running daily service requests, so when the actual success rate of the service request in the current pressure time period of the main service queue is greater than the success rate threshold of the current pressure time period of the main service queue, the service request belongs to a normally running service request, and the service request continues to run in the main service queue, so that the target service queue of the service request with the actual success rate greater than the success rate threshold is still the main service queue.

However, when the actual success rate of the service request in the current pressure time period of the main service queue is less than or equal to the success rate threshold of the current pressure time period of the main service queue, the service request is a problematic service request, the service request needs to be moved to a "semi-fusing queue" for operation, and compared with the main service queue, the service request amount in the semi-fusing queue is much smaller than that of the main service queue. Therefore, the target service queue of the service request with the actual success rate less than or equal to the success rate threshold is the semi-fusible queue.

In the embodiment of the application, for the main service queue, in the current pressure time period, when the actual success rate of the service request is greater than the success rate threshold of the current pressure time period of the main service queue, the service request is not changed and continues to operate in the main service queue, otherwise, the service request is moved to the semi-fusing queue, so that when the success rate of the service request is lower, the service request is not directly fused, but is moved to the semi-fusing queue with less service transaction amount, so that the service loss caused by accidental situations can be prevented, excessive fusing is prevented, associated normal fusing is caused, and service operation is influenced.

As shown in fig. 5, an embodiment is provided to explain a process of determining a target service queue corresponding to each service request when a current service queue is a semi-fused queue, where the embodiment includes the following steps:

s401, comparing a first actual success rate of each service request in the current pressure time period of the semi-fusing queue with a first success rate threshold of the current pressure time period of the semi-fusing queue; and comparing the second actual success rate of each service request in the last pressure time period of the semi-fusing queue with a second power threshold value of the last pressure time period of the semi-fusing queue.

Because the traffic transaction amount in the semi-fusing queue is far smaller than that of the main traffic queue, the pressure time period of the semi-fusing queue may be the pressure time period of the main traffic queue, and of course, may also be the pressure time period of the semi-fusing queue obtained through actual calculation, which is not limited in the embodiment of the present application.

The actual success rate of each service request in the current pressure time period of the semi-fusible queue is used as a first actual success rate, and the success rate threshold value of the current pressure time period of the semi-fusible queue is used as a first success rate threshold value, which can be calculated according to data monitored by the management server in the database, and the first success rate threshold value can be calculated in real time according to the method for determining the success rate threshold value of the current pressure time period.

And then comparing the first actual success rate with the first success rate threshold value to obtain a comparison result.

The last pressure time period refers to a last pressure time period of the current pressure time period, wherein the actual success rate of each service request in the last pressure time period of the semi-fusing queue is a second actual success rate, and the success rate threshold of the last pressure time period of the semi-fusing queue is a second power forming threshold.

The current time represented by the current pressure time period is the current pressure time period of the previous time when the previous pressure time period is the previous time, that is, the second actual success rate may also be calculated based on the data monitored by the management server in the database, and the second success rate threshold may also be calculated in real time according to the above-mentioned manner of determining the success rate threshold of the current pressure time period.

And obtaining a second actual success rate and a second power forming threshold value based on the calculation, and comparing the first actual success rate and the first power forming threshold value to obtain a comparison result.

S402, determining the target service queue of the service request with the first actual success rate larger than the first success rate threshold value and the second actual success rate larger than the second success rate threshold value as the main service queue.

If the first actual success rate is greater than the first success rate threshold and the second actual success rate is greater than the second success rate threshold, it indicates that the success rate of the requested service is higher in two consecutive time periods when the service request is in the semi-fused queue and the success rate is greater than the success rate threshold of the corresponding time period of the semi-fused queue, and the requested service can be moved back to the main service queue.

And S403, determining the target service queue of the service request with the first actual success rate smaller than the first success rate threshold and the second actual success rate smaller than the second success rate threshold as a fusing queue.

If the first actual success rate is smaller than the first success rate threshold and the second actual success rate is smaller than the second success rate threshold, it indicates that the success rate of the requested service is very low in two consecutive time periods, and even the requested service is failed, the service request is moved to the fused service queue, and the transaction is temporarily stopped, so that the target service queue of the service request with the first actual success rate smaller than the first success rate threshold and the second actual success rate smaller than the second success rate threshold is determined as the fused queue.

S404, determining the target service queue of the service request with the first actual success rate larger than the first power forming threshold value or the second actual success rate larger than the second power forming threshold value as a semi-fusing queue.

If the first actual success rate is greater than the first success rate threshold, or the second actual success rate is greater than the second success rate threshold, which indicates that in two consecutive time periods in the semi-fusing queue, sometimes the success rate of the service request is high, and the transaction can be successful, sometimes the success rate is very low, and the transaction fails, that is, the transaction of the service request is not stable, the service request is continuously put in the semi-fusing queue to operate. Namely, the target service queue of the service request with the first actual success rate greater than the first power forming threshold value or the second actual success rate greater than the second power forming threshold value is determined as the semi-fusing queue.

In the embodiment of the application, when the success rate of the service requests in two continuous time periods in the semi-fusing queue is higher, the service requests are moved back to the main service queue, so that the service can be normally traded, and when the service requests in two continuous time periods in the semi-fusing queue fail, the fusing queue is placed to fuse the service requests, so that the service loss caused by accidental conditions is prevented.

In the fusing queue, the transaction is suspended, and therefore, an automatic recovery function can be set to ensure that the service request related to the service request in the fusing queue can also run normally.

As shown in fig. 6, in an embodiment, if the current service queue is a fused queue, the acquiring the actual success rate of each service request in the current pressure time period of the current service queue in the step S101 includes:

s501, releasing a preset number of service requests from a fusing queue to run according to a preset interval time; the preset number is determined according to the number of the service requests of the main service queue in the current time period.

S502, the actual success rate of the released service request is obtained.

The preset interval time may be set according to an actual situation, or may be a pressure time period of the main service queue, that is, after the duration of the pressure time period of the main service queue is set, a preset number of service requests are released from the fuse queue to run, where the preset number is determined according to the number of the service requests in the current time period of the main service queue, for example, if the number of the service requests in the current time period of the main service queue is 100, the preset number may be one-fourth 25 of the preset number, that is, 25 service requests are released from the fuse queue.

Specifically, assuming that there is a service request due to network fluctuation, the service request is moved to a blown queue, the blown queue suspends the operation of the service request until the request is released after a pressure time period of a main service queue, and when the number of requests (in seconds, that is, the number of TPS of the current service request) of the main service queue at the present time is one hundred requests (the service request is currently 100TPS), one fourth of the requests is released, that is, the number of 25 requests is released. Of course, if the number of requests of the main service queue at the present moment is lower than 4, only one request is released.

After releasing the preset number of service requests, operating the preset number of service requests in a fusing queue, and then acquiring the actual success rate of the released service requests. And determining the target service queues of the preset number of service requests by taking the actual success rate of the preset number of service requests in the fusing queue as judgment.

Correspondingly, the process of determining the target service queue corresponding to each service request in S103 includes: and if the actual success rates of the preset number of service requests in the pressure time periods of two continuous fusing queues are greater than the success rate threshold of the latest pressure time period of the semi-fusing queue, determining the target service queue of the preset number of service requests as the semi-fusing queue.

And if the actual success rates of two continuous pressure time periods of the preset number of service requests in the fusing queue are both greater than the success rate threshold of the latest pressure time period of the semi-fusing queue, determining that the transaction success rate of the preset number of service requests is very high, and recovering the preset number of service requests into the semi-fusing queue, namely if the 25 service requests meet the success rate threshold of the latest pressure time period of the semi-fusing queue, moving the 25 service requests to the semi-fusing queue, and otherwise, repeating the process.

It should be noted that, in this embodiment, the actual success rate of the service request in the fusing queue for two consecutive pressure time periods is compared with the success rate threshold of the pressure time period closest to the semi-fusing queue, and because it is determined whether the service request should be moved back to the semi-fusing queue, the service request needs to be compared with the success rate threshold of the semi-fusing queue as a reference, and the success rate threshold of the pressure time period closest to the semi-fusing queue needs to be used, so that the determination is more consistent with the current actual situation of the semi-fusing queue, and the determination result is more accurate.

It should be understood that, although the steps in the flowcharts of the above embodiments are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts of the above embodiments may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.

In one embodiment, as shown in fig. 7, there is provided a service request execution apparatus, including: a first obtaining module 10, a second obtaining module 11 and a processing module 12, wherein:

a first obtaining module 10, configured to obtain an actual success rate of each service request in a current pressure time period of a current service queue; the current service queue is any one of a main service queue, a semi-fusing queue and a fusing queue; the pressure time period represents the time length of the maximum pressure which can be borne by the current service queue when the service request is operated;

the second obtaining module 11 is configured to obtain a success rate threshold corresponding to the current pressure time period;

the processing module 12 is configured to determine a target service queue corresponding to each service request according to an actual success rate of each service request in the current pressure time period and a success rate threshold of the current pressure time period.

In one embodiment, the second obtaining module 11 includes:

a first obtaining unit for obtaining a first average success rate of the first pressure period and a second average success rate of the second pressure period by performing the average success rate calculating step; the first pressure time period is the last pressure time period of the current pressure time period, and the second pressure time period is the last pressure time period of the last pressure time period;

a second obtaining unit, configured to obtain a minimum error between the first pressure time period and the second pressure time period according to the first average success rate and the second average success rate;

and the first determination unit is used for determining the sum of the minimum error and the first average success rate as a success rate threshold value of the current pressure time period.

In one embodiment, the first obtaining unit is further configured to divide any pressure time period into N sub-time periods, and obtain a total number of requests in each sub-time period, a number of successful requests in each sub-time period, and a weight of each sub-time period; n is a positive even number; calculating to obtain N/2 reference average success rates according to the total request number in each sub-time period, the request success number in each sub-time period and the weight in each sub-time period; determining the average success rate of the pressure time period according to the N/2 reference average success rates

In an embodiment, the second obtaining unit is further configured to construct a square loss function according to the first average success rate and the second average success rate; and optimizing the square loss function through a preset optimization algorithm, and determining the minimum error between the first pressure time period and the second pressure time period.

In one embodiment, if the current service queue is the main service queue; the processing module 12 includes:

the first comparison unit is used for comparing the actual success rate of each service request in the current pressure time period of the main service queue with the success rate threshold value of the current pressure time period of the main service queue;

a second determining unit, configured to determine, as a main service queue, a target service queue of a service request with an actual success rate greater than a success rate threshold; and determining the target service queue of the service request with the actual success rate less than or equal to the success rate threshold value as a semi-fusing queue.

In one embodiment, if the current service queue is a semi-fusible queue; the processing module 12 includes:

the second comparison unit is used for comparing the first actual success rate of each service request in the current pressure time period of the semi-fusing queue with the first success rate threshold value of the current pressure time period of the semi-fusing queue; comparing the second actual success rate of each service request in the last pressure time period of the semi-fusing queue with a second power forming threshold value of the last pressure time period of the semi-fusing queue;

a third determining unit, configured to determine, as a main service queue, a target service queue of a service request for which the first actual success rate is greater than the first power forming threshold and the second actual success rate is greater than the second power forming threshold; determining a target service queue of the service request with the first actual success rate smaller than a first power forming threshold value and the second actual success rate smaller than a second power forming threshold value as a fusing queue; and determining the target service queue of the service request with the first actual success rate greater than the first power forming threshold value or the second actual success rate greater than the second power forming threshold value as a semi-fusing queue.

In an embodiment, if the current service queue is a blown queue, the first obtaining module 10 includes:

the release unit is used for releasing a preset number of service requests from the fusing queue to run according to a preset interval time; the preset quantity is determined according to the quantity of the service requests of the main service queue in the current time period;

and the third obtaining unit is used for obtaining the actual success rate of the released service request.

In an embodiment, the processing module 12 further includes a fourth determining unit, configured to determine a target service queue of the preset number of service requests as a semi-fused queue if actual success rates of the preset number of service requests in the pressure time periods of two consecutive fused queues are greater than a success rate threshold of a closest pressure time period of the semi-fused queue.

For specific limitations of the service request operation device, reference may be made to the above limitations on the service request operation method, which is not described herein again. The modules in the service request execution device can be wholly or partially implemented by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in a computer device, and can also be stored in a memory in an electronic device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, the internal structure of which may be as shown in fig. 8. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a service request execution method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.

Those skilled in the art will appreciate that the architecture shown in fig. 8 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

and determining a target service queue corresponding to each service request according to the actual success rate of each service request in the current pressure time period and the success rate threshold of the current pressure time period.

The implementation principle and technical effect of the computer device provided by the above embodiment are similar to those of the above method embodiment, and are not described herein again.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

The implementation principle and technical effect of the computer-readable storage medium provided by the above embodiments are similar to those of the above method embodiments, and are not described herein again.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express a few embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for those skilled in the art, variations and modifications can be made without departing from the concept of the embodiments of the present application, and these embodiments are within the scope of the present application. Therefore, the protection scope of the embodiments of the present application shall be subject to the appended claims.

Claims

1. A service request operation method is characterized by comprising the following steps:

2. The method of claim 1, wherein the obtaining the success rate threshold corresponding to the current pressure time period comprises:

acquiring a first average success rate of a first pressure time period and a second average success rate of a second pressure time period by executing a preset average success rate calculation step; the first pressure time period is the last pressure time period of the current pressure time period, and the second pressure time period is the last pressure time period of the last pressure time period;

determining a sum of the minimum error and the first average success rate as a success rate threshold for the current pressure time period.

3. The method of claim 2, wherein obtaining the minimum error between the first pressure time period and the second pressure time period according to the first average success rate and the second average success rate comprises:

4. A method according to claim 2 or 3, wherein the average success rate calculating step comprises:

calculating to obtain N/2 reference average success rates according to the total number of requests in each sub-time period, the number of success requests in each sub-time period and the weight in each sub-time period;

5. A method according to any one of claims 1 to 3, wherein if the current service queue is a main service queue; determining a target service queue corresponding to each service request according to the actual success rate of each service request in the current pressure time period and the success rate threshold of the current pressure time period, including:

determining a target service queue of the service request with the actual success rate greater than the success rate threshold value as the main service queue;

and determining the target service queue of the service request with the actual success rate less than or equal to the success rate threshold value as the semi-fusing queue.

6. A method according to any one of claims 1 to 3, wherein if the current traffic queue is a semi-fusible queue; determining a target service queue corresponding to each service request according to the actual success rate of each service request in the current pressure time period and the success rate threshold of the current pressure time period, including:

comparing a first actual success rate of each service request in the current pressure time period of the semi-fusing queue with a first success rate threshold of the current pressure time period of the semi-fusing queue; comparing a second actual success rate of each service request in a last pressure time period of the semi-fusing queue with a second power forming threshold value of the last pressure time period of the semi-fusing queue;

determining a target service queue of the service request with a first actual success rate greater than a first power forming threshold value and a second actual success rate greater than a second power forming threshold value as the main service queue;

determining a target service queue of the service request with a first actual success rate smaller than a first power forming threshold value and a second actual success rate smaller than a second power forming threshold value as the fusing queue;

and determining the target service queue of the service request with the first actual success rate greater than the first power forming threshold value or the second actual success rate greater than the second power forming threshold value as the semi-fusing queue.

7. The method according to any one of claims 1 to 3, wherein if the current service queue is a fused queue, the obtaining the actual success rate of each service request in the current pressure time period of the current service queue comprises:

releasing a preset number of service requests from the fusing queue to run according to a preset interval time; the preset number is determined according to the number of the service requests of the main service queue in the current time period;

and acquiring the actual success rate of the released service request.

8. The method according to claim 7, wherein the determining a target service queue corresponding to each service request according to the actual success rate of each service request in the current pressure time period and the success rate threshold of the current pressure time period comprises:

9. A service request execution system, the system comprising: the system comprises a management server and at least one service processing link, wherein each service processing link comprises an application server, a database and a network server;

the management server is used for executing the service request operation method of any one of claims 1 to 8;

the application server is used for forwarding a plurality of service requests sent by a user side to a corresponding network server and storing service information carried in each service request into the database; receiving a service request result returned by the corresponding network server, and storing the service request result into the database;

10. The system of claim 9, wherein the traffic handling link comprises a main traffic queue, a semi-blown queue, and a blown queue;

the main service queue comprises a first application server, the database and a first network server; the main service queue is used for running service requests in a normal state;

the semi-fusing queue comprises a second application server, the database and a second network server; the semi-fusing queue is used for running service requests in a semi-fusing state;

the fusing queue comprises a third application server, the database and a third network server; and the fusing queue is used for business requests in a fusing state.

11. A service request execution apparatus, comprising:

and the processing module is used for determining a target service queue corresponding to each service request according to the actual success rate of each service request in the current pressure time period and the success rate threshold of the current pressure time period.

12. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of any of claims 1 to 8 are implemented by the processor when executing the computer program.

13. A storage medium having a computer program stored thereon, the computer program, when being executed by a processor, realizing the steps of the method of any one of claims 1 to 8.