CN112445585A

CN112445585A - Method and device for controlling service quality of server cluster

Info

Publication number: CN112445585A
Application number: CN201910809106.9A
Authority: CN
Inventors: 田彦博
Original assignee: Beijing Jingdong Century Trading Co Ltd; Beijing Jingdong Shangke Information Technology Co Ltd
Current assignee: Beijing Jingdong Century Trading Co Ltd; Beijing Jingdong Shangke Information Technology Co Ltd
Priority date: 2019-08-29
Filing date: 2019-08-29
Publication date: 2021-03-05

Abstract

The invention discloses a method and a device for controlling the service quality of a server cluster, and relates to the technical field of computers. One specific implementation of the method comprises the following steps: monitoring state data of each server in the server cluster; controlling the service quality of the server cluster according to the state data value of each server, and for each server, if the state data value is greater than a first threshold value and less than or equal to a second threshold value, rejecting the requests classified by the part according to a step degradation strategy aiming at the requests; and if the state data value of the server is greater than the second threshold value, or the state data value is still greater than the first threshold value and less than or equal to the second threshold value after all the types of requests are rejected, closing the services of the service classification according to the step degradation strategy for the services. The implementation method can timely and quickly solve the problem of server cluster service quality reduction caused by various reasons such as burst flow and the like, reduce the unavailable range of service, ensure the cluster service quality and improve the user experience.

Description

Method and device for controlling service quality of server cluster

Technical Field

The present invention relates to the field of computer technologies, and in particular, to a method and an apparatus for controlling a service quality of a server cluster.

Background

At present, almost all internet companies adopt a service cluster as a service carrier, a cluster server is limited by hardware resources, such as a Central Processing Unit (CPU), a memory, a network card, and the like, when traffic exceeding the processing capacity of the cluster occurs in a certain scale, the service provided by the cluster enters a partially unavailable state, and if the traffic is continuous and the cluster has no any specific measures, the service may be completely unavailable, and the cluster is shut down. For example, in a large-scale e-commerce application, because of the huge amount of users, when a website is promoted in a centralized manner, in the face of an abrupt flow exceeding the cluster service capability, if the cluster is unavailable, the user experience is seriously affected, and the effect of promoting the sale of the marketing manner is greatly reduced.

In the prior art, a mode of adding servers in a cluster in an emergency is usually adopted for solving the problems, but a certain time is needed for adding the servers to the cluster, and the drilling needs to be performed in advance, the data preparation and the server preparation are inevitable time-consuming operations, a part of cluster services cannot be inevitably used in the operation process, for example, a part of important services cannot be used at all, and in addition, the mode cannot completely cope with the condition of sudden large flow, the reaction to the sudden flow is seriously delayed, and the user experience is influenced.

In the process of implementing the invention, the inventor finds that at least the following problems exist in the prior art:

the problem of service quality reduction of the server cluster caused by various reasons such as burst flow and the like cannot be solved in time and quickly, and the user experience is poor.

Disclosure of Invention

In view of this, embodiments of the present invention provide a method and an apparatus for controlling a service quality of a server cluster, which can timely and quickly solve a problem of a decrease in the service quality of the server cluster caused by various reasons such as a burst traffic, reduce a scope of unavailable service, ensure the service quality of the server cluster, and improve user experience.

To achieve the above object, according to an aspect of embodiments of the present invention, a method for controlling quality of service of a server cluster is provided.

A method of controlling quality of service for a cluster of servers, comprising: monitoring state data of each server in a server cluster to obtain a state data value of each server; controlling the service quality of the server cluster according to the state data value of each server, wherein: for each server, if the state data value of the server is greater than a first threshold value and less than or equal to a second threshold value, performing a rejection operation on the request of the partial category according to a step degradation strategy aiming at the request; and if the state data value of the server is greater than the second threshold value, or the state data value of the server is still greater than the first threshold value and less than or equal to the second threshold value after all the classes of requests are rejected, performing closing operation on the services classified according to the step degradation strategy for the services.

Optionally, before the step of monitoring the status of each server in the server cluster, the method further includes: configuring a service priority for a service provided by the server cluster and a request priority for a request of each class of service, the service priority and the request priority indicating an importance of the service and the request, respectively.

Optionally, the state data of the server includes one or more of the following data: the CPU utilization rate, the memory utilization rate, the bandwidth utilization rate and the average response time of a specified service processing request of the server are obtained, and the specified service is determined according to the service priority.

Optionally, in a case that the status data of the server includes the plurality of types of data, each type of data has the corresponding first threshold and the corresponding second threshold, where when the plurality of types of data values are all smaller than or equal to the corresponding second thresholds, and at least one of the plurality of types of data values is larger than the corresponding first threshold, it indicates that the status data value of the server is larger than the first threshold and smaller than or equal to the second threshold; when at least one of the plurality of data values is greater than the corresponding second threshold value, indicating that the state data value of the server is greater than the second threshold value; when the plurality of data values are all smaller than or equal to the first threshold value corresponding to each data value, the state data value representing the server is smaller than or equal to the first threshold value.

Optionally, the step of performing a rejection operation on the requests classified according to the step downgrading policy for the requests includes: and executing a refusal operation on the requests according to the request priority level one by one from the request of the category with the lowest request priority level until the state data value of the server becomes less than or equal to the first threshold value or the interval range of the first threshold value after the refusal operation is executed on the request of the category, or stopping the refusal operation when the refusal request of the category with higher request priority level does not exist.

Optionally, before the step of performing a denial operation on the requests classified according to the step downgrading policy for the requests, the method further includes: if there is a service that has been closed, the closed service is restored.

Optionally, the step of performing a shutdown operation on the classified services according to the step downgrading policy for the services includes: and starting from the service of the class with the lowest service priority, executing closing operation on the services according to the service priority class by class until the state data value of the server becomes smaller than or equal to a second threshold value or an interval range of the second threshold value after the closing operation is executed on the service of a certain class, or stopping the closing operation when the service which can be closed and has a higher service priority class does not exist.

According to another aspect of the embodiments of the present invention, an apparatus for controlling quality of service of a server cluster is provided.

An apparatus for controlling quality of service for a cluster of servers, comprising: the cluster monitoring module is used for monitoring the state data of each server in the server cluster to obtain the state data value of each server; a service quality control module, configured to control the service quality of the server cluster according to the state data value of each server, where: for each server, if the state data value of the server is greater than a first threshold value and less than or equal to a second threshold value, performing a rejection operation on the request of the partial category according to a step degradation strategy aiming at the request; and if the state data value of the server is greater than the second threshold value, or the state data value of the server is still greater than the first threshold value and less than or equal to the second threshold value after all the classes of requests are rejected, performing closing operation on the services classified according to the step degradation strategy for the services.

Optionally, the system further comprises a service priority configuration module and a request priority configuration module, wherein: the service priority configuration module is used for configuring service priority for the service provided by the server cluster; the request priority configuration module is used for configuring request priority aiming at each class of service request; wherein the service priority and the request priority indicate a degree of importance of the service and the request, respectively.

Optionally, the quality of service control module includes a request degradation submodule configured to: and executing a refusal operation on the requests according to the request priority level one by one from the request of the category with the lowest request priority level until the state data value of the server becomes less than or equal to the first threshold value or the interval range of the first threshold value after the refusal operation is executed on the request of the category, or stopping the refusal operation when the refusal request of the category with higher request priority level does not exist.

Optionally, the service quality control module further includes a service restoration sub-module, configured to restore the closed service before the request downgrading sub-module performs a reject operation on the requests of the different categories according to the stepped downgrading policy for the requests.

Optionally, the quality of service control module includes a service degradation submodule configured to: and starting from the service of the class with the lowest service priority, executing closing operation on the services according to the service priority class by class until the state data value of the server becomes smaller than or equal to a second threshold value or an interval range of the second threshold value after the closing operation is executed on the service of a certain class, or stopping the closing operation when the service which can be closed and has a higher service priority class does not exist.

According to yet another aspect of an embodiment of the present invention, an electronic device is provided.

An electronic device, comprising: one or more processors; a memory for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the method of controlling quality of service of a cluster of servers provided by the present invention.

According to yet another aspect of an embodiment of the present invention, a computer-readable medium is provided.

A computer-readable medium, on which a computer program is stored which, when being executed by a processor, carries out the method of controlling the quality of service of a cluster of servers according to the invention.

One embodiment of the above invention has the following advantages or benefits: monitoring state data of each server in the server cluster to obtain a state data value of each server; controlling the service quality of the server cluster according to the state data value of each server, wherein: for each server, if the state data value of the server is greater than a first threshold value and less than or equal to a second threshold value, executing rejection operation on the requests of the partial category according to a step degradation strategy aiming at the requests; and if the state data value of the server is larger than the second threshold value, or the state data value of the server is still larger than the first threshold value and smaller than or equal to the second threshold value after all the classes of requests are rejected, performing closing operation on the services classified according to the step degradation strategy for the services. The method can timely and quickly solve the problem of service quality reduction of the server cluster caused by various reasons such as burst flow and the like, reduce the range of unavailable service, ensure the service quality of the server cluster and improve the user experience.

Further effects of the above-mentioned non-conventional alternatives will be described below in connection with the embodiments.

Drawings

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

FIG. 1 is a schematic diagram of the main steps of a method of controlling the quality of service of a cluster of servers according to one embodiment of the invention;

FIG. 2 is a schematic flow diagram for controlling quality of service for a cluster of servers, according to one embodiment of the invention;

FIG. 3 is a schematic diagram of the main modules of an apparatus for controlling the quality of service of a server cluster according to one embodiment of the present invention;

FIG. 4 is a schematic diagram of an architecture of an apparatus for controlling quality of service of a server cluster according to an embodiment of the present invention;

FIG. 5 is an exemplary system architecture diagram in which embodiments of the present invention may be employed;

fig. 6 is a schematic block diagram of a computer system suitable for use in implementing a terminal device or server of an embodiment of the invention.

Detailed Description

Exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which various details of embodiments of the invention 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 invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

As will be appreciated by one skilled in the art, embodiments of the present invention may be embodied as a system, apparatus, device, method, or computer program product. Accordingly, the present disclosure may be embodied in the form of: entirely hardware, entirely software (including firmware, resident software, micro-code, etc.), or a combination of hardware and software.

A cluster is a set of servers working in coordination to provide a more stable, efficient, and extensible service platform than a single service. From the external view, a cluster is an independent service entity, but actually, two or more service entities in the cluster coordinate and cooperate to complete a series of complex work. A cluster is generally composed of two or more servers, each of which is called a cluster node, and the cluster nodes can communicate with each other. In a cluster, the same service may be provided by multiple service entities. Therefore, when one node fails, the other node of the cluster can automatically take over the resources of the failed node, thereby ensuring that the service runs durably and uninterruptedly.

The invention controls the service quality of the server cluster by degrading each single server in the server cluster according to the service priority and the request priority, so as to ensure that the important requests of various types of services can be responded to by the cluster as soon as possible, ensure the normal use of the important services as soon as possible, and quickly ensure the service quality of the server cluster in time for the scene of sudden flow such as e.g. e-commerce promotion, and improve the overall user experience of the service.

Fig. 1 is a schematic diagram of the main steps of a method for controlling the quality of service of a server cluster according to one embodiment of the present invention.

As shown in fig. 1, the method for controlling the quality of service of a server cluster of this embodiment mainly includes the following steps S101 to S102.

Step S101: and monitoring the state data of each server in the server cluster to obtain the state data value of each server.

Before monitoring the state of each server in the server cluster, service priority may be configured for the service provided by the server cluster in advance, and request priority may be configured for the request of each class of service, where the service priority and the request priority indicate the importance of the service and the request, respectively, and the more important the service is, the higher the service priority is. Likewise, the more important the request, the higher the request priority.

The state data of the server includes one or more of the following data: the service processing method comprises the following steps of CPU utilization rate, memory utilization rate, bandwidth utilization rate and average response time of processing requests of a specified service of a server, wherein the specified service can be determined according to service priority, and specifically the specified service can comprise one or more services with the highest service priority. In a case where the specified service includes a plurality of services having the highest service priority, the average response time of processing the request by the specified service includes the average response time of processing the request by each service of the plurality of services. The average response time for a service to process a request refers to the average response time for the service to process a request.

The CPU utilization rate, the memory utilization rate and the bandwidth utilization rate can be obtained through open source monitoring software, and the average response time of the appointed service processing request can be obtained in real time through writing codes.

Step S102: controlling the service quality of the server cluster according to the state data value of each server, wherein: for each server, if the state data value of the server is greater than a first threshold value and less than or equal to a second threshold value, executing rejection operation on the requests of the partial category according to a step degradation strategy aiming at the requests; and if the state data value of the server is larger than the second threshold value, or the state data value of the server is still larger than the first threshold value and smaller than or equal to the second threshold value after all the classes of requests are rejected, performing closing operation on the services classified according to the step degradation strategy for the services.

The state data of the server includes one of CPU usage, memory usage, bandwidth usage, and average response time of the designated service processing request of the server.

Taking the CPU utilization as an example, the first threshold and the second threshold are two thresholds corresponding to the CPU utilization, respectively, and for each server, if the value of the CPU utilization is greater than the first threshold and less than or equal to the second threshold, the step degradation policy for the request is followed, and a reject operation is performed on the requests classified in the part; and if the value of the CPU utilization rate is larger than a second threshold value, or the value of the CPU utilization rate is still larger than the first threshold value and smaller than or equal to the second threshold value after all the classes of requests are rejected, closing the services classified according to the step degradation strategy aiming at the services.

Taking the average response time of a specific service processing request as an example, if the specific service includes a plurality of services with the highest service priority, each service in the plurality of services has a respective average response time for processing the request, and when the average response time value of at least one service processing request is greater than a first threshold and the average response time values of the plurality of service processing requests are all less than or equal to a second threshold, performing a rejection operation on the classified requests according to a step-down policy for the requests; and if the average response time value of at least one service processing request is larger than the second threshold value, or after all the classes of requests are rejected, the average response time value of at least one service processing request is still larger than the first threshold value, and the average response time values of the plurality of service processing requests are smaller than or equal to the second threshold value, closing the services of the classes according to the step degradation strategy for the services.

As another embodiment, the state data of the server includes multiple data (two or more) of a CPU usage rate, a memory usage rate, a bandwidth usage rate, and an average response time of a specific service processing request of the server, where each data has a corresponding first threshold and a corresponding second threshold, for example, the CPU usage rate corresponds to the first threshold and the second threshold of the CPU usage rate, and the memory usage rate corresponds to the first threshold and the second threshold of the memory usage rate.

Then, when the plurality of data values are all smaller than or equal to the respective corresponding second threshold values, and at least one of the data values is larger than the corresponding first threshold value, the status data value of the server is larger than the first threshold value and smaller than or equal to the second threshold value; when at least one of the multiple data values is greater than the corresponding second threshold value, the state data value of the server is greater than the second threshold value; when the various data values are all smaller than or equal to the corresponding first threshold value, the state data value representing the server is smaller than or equal to the first threshold value.

The step of performing a denial operation on the requests classified according to the ladder destaging policy for the requests of this embodiment may include: and executing rejection operation on the requests in the category by category according to the request priority from the request in the category with the lowest request priority until the status data value of the server becomes less than or equal to the first threshold or the interval range of the first threshold after the rejection operation is executed on the request in the category, or stopping the rejection operation when the rejectable request in the category with higher request priority does not exist.

Wherein, stopping the rejection operation means that the request with higher priority is not rejected any more.

The range of the first threshold may be a range in which the first threshold is a lower limit, a predetermined value greater than the first threshold and less than the second threshold is an upper limit, the range may be an open range or a half-open and half-closed range, and the value of the upper limit of the range is set to a value close to the first threshold. Taking the state data of the server as the CPU utilization as an example, assuming that the first threshold of the CPU utilization is 60% and the second threshold is 80%, the interval range of the first threshold may be set to (60%, 65% ]. then, after rejecting a request of a category corresponding to a certain request priority, if the value of the CPU utilization is 62%, the operation of rejecting is stopped, that is, a request of a category of a higher request priority is not further rejected.a current CPU utilization value is greater than 60% of the first threshold, but is within the small interval range of (60%, 65% ]), the influence on the server cluster is not great, so that requests of categories of higher request priorities can be processed as usual, thereby ensuring that requests of a plurality of user groups are processed as much as possible, and improving user experience.

If the state data value of the server becomes less than or equal to the first threshold value, on the basis of stopping the above-mentioned rejection operation, the method also includes recovering the request of the category of the first-level request priority which is rejected last to normal processing, and if the state of the server continuously becomes better (is always less than or equal to the first threshold value), the method also can gradually recover the request of the category corresponding to each request priority according to the sequence from high to low of the request priority in the rejected requests. Therefore, the requests of a plurality of user groups can be processed as much as possible, and the user experience is improved.

If there is no refusable request with higher request priority, on the basis of stopping the above-mentioned refusing operation, the method also includes recovering the refused requests of each category to normal processing, so as to close the services of the categories according to the step degradation strategy for the services, and the service quality of the server cluster is guaranteed through the step degradation strategy for the services.

As one implementation, the request that may be rejected may be a request that has not been currently rejected.

As another embodiment, the requests that may be rejected may be requests other than the pre-specified particular category of requests among the requests that are not currently rejected. The denial operations may be pre-specified to be disallowed for one or more categories of requests, which are typically requests of significant groups of users. For example, if the guest user is an important group user, it may be specified in advance that the request of the guest user is not allowed to perform the rejection operation, and the request of the guest user is not a request that can be rejected. If only the request of the honored user exists in the requests which are not rejected currently, the requests which are higher in request priority and can be rejected are considered to be absent.

It should be noted that, since the second embodiment specifies in advance that the request of a specific category is not allowed to perform the rejection operation, after all the rejectable requests have been rejected, the requests of all categories are considered to have been rejected. Then, if the state data value of the server is still greater than the first threshold and less than or equal to the second threshold at this time, the shutdown operation is performed on the part of the service classes according to the step degradation policy for the service.

According to the step downgrade strategy for the request, before the reject operation is performed on the request classified into the groups, if the closed service exists, the closed service can be restored.

The step of performing the closing operation on the classified services according to the step degradation policy for the services in this embodiment may include: and starting from the service of the class with the lowest service priority, executing closing operation on the services according to the service priority class by class until the state data value of the server becomes less than or equal to the second threshold value or the interval range of the second threshold value after the closing operation is executed on the service of a certain class, or stopping the closing operation when the service which can be closed and has a higher service priority class does not exist.

Wherein, stopping the closing operation means not closing the service of the higher service priority class.

The range of the second threshold may be a range in which the second threshold is a lower limit and a predetermined value larger than the second threshold is an upper limit, the range may be an open range or a half-open and half-closed range, and the value of the upper limit of the range is set to a value close to the second threshold. Taking the state data of the server as the CPU utilization as an example, assuming that the second threshold of the CPU utilization is 80%, the interval range of the second threshold may be set to (80%, 85% ]. then, after the service of a certain service priority class is closed, if the value of the CPU utilization is 82%, the closing operation is stopped, i.e., the service of a higher service priority class is not further closed. thus, although the current CPU utilization value is greater than the second threshold 80%, it is within the small interval range of (80%, 85% ]. in order to ensure that as many services as possible are provided to the user, the step downgrading for the service is not continued at this time, but the step downgrading for the request is started, i.e., the service quality of the server cluster is improved by performing the reject operation of classifying the requests according to the step downgrading policy for the request, thereby improving the service quality for the part, but also ensures that the service is available as much as possible and improves the user experience.

If the state data value of the server becomes less than or equal to the second threshold value, on the basis of stopping the closing operation, the method also comprises the steps of completely recovering the closed service and re-executing the steps of the method.

If there is no service with higher service priority that can be closed, on the basis of stopping the above-mentioned closing operation, it also includes adding new server, and recovering the service of each type that has been rejected, then re-executing the steps of said method.

As an embodiment, a service that may be closed may be a service that is not currently closed.

As another embodiment, the services that can be closed may be services other than a pre-specified specific class of services among services that have not been currently closed. It may be pre-specified that the shutdown operation is not allowed for one or more classes of services, the pre-specified services that are not allowed to perform the shutdown operation typically being important services. For example, for an e-commerce system where a payment service is an important service, it may be specified in advance that a closing operation is not allowed for the payment service, and the payment service is not a service that can be closed. If only the payment service is available among the services that have not been currently switched off, it is considered that there is no service that can be switched off of the service priority higher category. Since in this embodiment it is pre-specified that the closing operation is not allowed for a particular class of services, then after all services that can be closed have been closed, it is considered that all classes of services have been closed. If the state data value of the server is still larger than the second threshold value at this time, the service quality of the server cluster is improved by the manner of adding the server, and the normal provision of the important service and the normal processing of the request of the important group user are ensured.

In addition, the invention can further subdivide the degradation strategy (including the step degradation strategy for the service and the request) into multiple steps under the overall inventive concept of step degradation of the request and the step degradation of the service. For example, when the reject operation is performed on the requests classified into two categories according to the step downgrading policy for the requests, the two categories may be specifically subdivided into step downgrading for important requests and step downgrading for non-important requests, and specifically, a third threshold may be further set between the first threshold and the second threshold, that is: the first threshold value < the third threshold value < the second threshold value. If the state data value of the server is greater than the first threshold and less than or equal to the second threshold, according to the step downgrading strategy aiming at the request, in the process of executing rejection operation on the requests of the partial classes: and when the state data value of the server is greater than the third threshold value, carrying out step degradation on the non-important request according to the request priority, and when the state data value of the server is greater than the third threshold value, carrying out step degradation on the non-important request according to the request priority. It can be understood by those skilled in the art that, according to the foregoing concept, the present invention can also subdivide the step degradation for the service into step degradation for important services and step degradation for non-important services, and accordingly, another fourth threshold greater than the second threshold needs to be set, if the status data value of the server is greater than the second threshold, the non-important services are step degraded according to the service priority if the status data value of the server is less than or equal to the fourth threshold, and then the non-important services are step degraded on the basis of the step degradation for the non-important services when the status data value of the server is greater than the fourth threshold. Which services are important services, which services are non-important services, which requests are important requests, and which requests are non-important requests can be preset according to business needs.

In summary, the downgrading strategy of the present invention can be flexibly adapted under the inventive concept that entirely surrounds the step downgrading of requests in combination with the step downgrading of services.

In the embodiment, each individual server in the server cluster is degraded in a stepped manner according to the service priority and the request priority, and targeted and differential degradation can be performed, the request is rejected step by step according to the sequence from low to high of the importance of the request, so that complete unavailability of part of services is avoided, and the normal processing of the request of the important group of users is ensured, thereby improving the experience of the important group of users, the services are closed step by step according to the sequence from low to high of the importance of the services when the state of the server is deteriorated, the problem of service quality degradation of the server cluster caused by various reasons such as burst traffic can be solved timely and quickly, the unavailable service is recovered timely after the state of the server is improved, the service range of the server cluster is reduced, the service quality of the server cluster can be ensured, the total traffic of the server cluster is reduced, and the pressure of, and improves the user experience.

Fig. 2 is a schematic flow chart of controlling the quality of service of a server cluster according to an embodiment of the present invention.

As shown in fig. 2, the flow of controlling the quality of service of the server cluster of the present embodiment includes steps S201 to S205 as follows.

Step S201: and monitoring the state data of the servers in the server cluster to obtain the state data value of the servers.

And monitoring the state data of all the servers in the server cluster to obtain the state data value of each server, and controlling the service quality of each server according to the following process so as to control the service quality of the whole server cluster.

Step S202: and judging whether the state data value of the server is larger than a first threshold value and smaller than or equal to a second threshold value. If yes, go to step S203; if not, and the status data value is greater than the second threshold, step S205 is executed.

If the state data value of the server is less than or equal to the first threshold, the process is skipped, which means that step downgrading is not currently performed for the request and the service.

Step S203: a ladder destage for requests is initiated including performing a denial operation on requests of the category for which the current request has the lowest priority.

E.g., the request of the paying user has a higher priority than the request of the ordinary user, the request of the ordinary user is rejected.

If the step degradation aiming at the service is started currently, the step degradation aiming at the service is stopped first, and then the rejection operation is executed on the request of the category with the lowest priority of the current request.

Step S204: and judging whether the state data value of the server is greater than a second threshold value, if so, executing step S205, otherwise, executing step S201.

In order to prevent the service quality of the server cluster from being rapidly deteriorated due to the condition of burst traffic after the rejection operation is performed on the request of the category with low request priority, whether the state data value of the server is greater than a second threshold value or not can be judged, and if the state data value is greater than the second threshold value, the step degradation for the service is started. If not, the process returns to step S201 to re-execute the present flow.

Before the state data value of the server does not exceed the second threshold, if the state data value of the server is in the interval range of the first threshold, the rejecting operation of the request of the category with higher request priority is not continued, and in combination with the above example, the rejecting operation of the request of the paying user is not further executed, and the server provides service under the condition of rejecting the request of the ordinary user; if the server state data value is restored to be less than or equal to the first threshold value, the step degradation aiming at the request is cancelled, namely, not only the rejection operation of the request of the paying user is not executed, but also the normal processing of the request of the ordinary user is restored (the request of the ordinary user is not rejected any more); if the state of the server continues to deteriorate (i.e. is greater than the first threshold and outside the range of the first threshold, but is less than or equal to the second threshold), step S203 needs to further reject the higher-level request, i.e. reject the request of the paying user until all the categories of requests are rejected, and then step S205 is executed.

Step S205: and starting the ladder degradation for the service, wherein the ladder degradation comprises the step of executing closing operation on the service of the category with the lowest priority of the current service.

For example, in an e-commerce application, the service priority of the payment service is higher than that of the recommended commodity service, and then the recommended commodity service is closed, that is, degradation results are directly returned to all requests of the recommended commodity service, and processing is not performed any more.

If the server state is kept stable, namely within the interval range of the second threshold value, the current ladder degradation for the service is kept, namely the recommended commodity service is kept closed, and the payment service is not continuously closed; if the server status continues to deteriorate, i.e. continues to be greater than the second threshold and is outside the interval range of the second threshold, step S205 needs to further close the higher-level service, i.e. close the payment service; and if the server state data value is smaller than or equal to the second threshold value, recovering all closed services and re-executing the process.

The embodiment of the invention aims at the degradation of the single server in the server cluster (comprising the step degradation aiming at the service and the request), but not the degradation according to the service of the server cluster, mainly considers that the configuration of each server in the current server cluster is different, and the fine-grained degradation can ensure that the server cluster can provide the service and respond to the request as much as possible, reduce the influence range of the degradation and further reduce the influence on the user experience.

In addition, a two-stage step degradation strategy for the service and the request is adopted, step degradation is firstly carried out on the request of the service, then step degradation is carried out on the service, the degradation range is gradually expanded, fine-grained degradation is further ensured, and the range of unavailable service and the influence on user experience are reduced as much as possible.

Fig. 3 is a schematic diagram of main modules of an apparatus for controlling quality of service of a server cluster according to an embodiment of the present invention.

As shown in fig. 3, the apparatus 300 for controlling the quality of service of a server cluster according to this embodiment mainly includes: cluster monitoring module 301, service quality control module 302.

The cluster monitoring module 301 is configured to monitor status data of each server in the server cluster to obtain a status data value of each server.

The implementation of cluster monitoring module 301 may be an open-source framework or a self-developed tool, or modified on an open-source framework.

A service quality control module 302, configured to control the service quality of the server cluster according to the state data value of each server, where: for each server, if the state data value of the server is greater than a first threshold value and less than or equal to a second threshold value, performing a rejection operation on the requests of the partial category according to a step degradation strategy aiming at the requests; and if the state data value of the server is larger than the second threshold value, or the state data value of the server is still larger than the first threshold value and smaller than or equal to the second threshold value after all the types of requests are rejected, closing the services classified according to the step degradation strategy aiming at the services.

The apparatus 300 for controlling quality of service of a server cluster may further include a service priority configuration module and a request priority configuration module, wherein: the service priority configuration module is used for configuring service priority aiming at the service provided by the server cluster; the request priority configuration module is used for configuring request priority for each class of service request; wherein the service priority and the request priority indicate the importance of the service and the request, respectively.

The state data of the server includes one or more of the following data: CPU utilization, memory utilization, bandwidth utilization, and average response time of a given service processing request of a server. The designated service is determined according to a service priority, and particularly, the designated service may include one or more services having the highest service priority. In a case where the specified service includes a plurality of services having the highest service priority, the average response time of processing the request by the specified service includes the average response time of processing the request by each service of the plurality of services. The average response time for a service to process a request refers to the average response time for the service to process a request.

When the state data of the server includes a plurality of types of data, each type of data has a corresponding first threshold and a corresponding second threshold.

When the various data values are all smaller than or equal to the respective corresponding second threshold values, and at least one data value is larger than the corresponding first threshold value, the state data value of the server is larger than the first threshold value and smaller than or equal to the second threshold value;

when at least one data value in the multiple data values is larger than the corresponding second threshold value, the state data value of the server is larger than the second threshold value;

when the various data values are all smaller than or equal to the corresponding first threshold value, the state data value representing the server is smaller than or equal to the first threshold value.

The service quality control module 302 includes a request downgrade sub-module for: the method includes the steps that a rejection operation is executed on requests in a category according to request priorities from a request in a category with the lowest request priority, and the rejection operation is stopped when the state data value of a server becomes smaller than or equal to a first threshold value or an interval range within the first threshold value after the rejection operation is executed on the requests in the category, or a rejectable request in a category with a higher request priority does not exist.

The service quality control module 302 further includes a service restoration sub-module for restoring the closed service before the request downgrading sub-module performs a reject operation on the request of the classified category according to the stepped downgrading policy for the request.

The service quality control module 302 further comprises a service downgrading submodule for: and starting from the service of the class with the lowest service priority, executing closing operation on the services according to the service priority class by class until the state data value of the server becomes less than or equal to a second threshold value or an interval range of the second threshold value after the closing operation is executed on the service of a certain class, or stopping the closing operation when the service which can be closed and has a higher service priority class does not exist.

In the embodiment, through a configuration mode, firstly, the service priority is configured, secondly, the user group feature priority of the service, that is, the request priority of the service is configured, and the degradation is performed on a single server according to the two-level priority, wherein the degradation includes closing part of the request and closing part of the service, so that the influence on the overall user experience of the service is reduced as much as possible.

Fig. 4 is a schematic architecture diagram of an apparatus for controlling quality of service of a server cluster according to an embodiment of the present invention.

As shown in fig. 4, the architecture of the apparatus for controlling quality of service of a server cluster according to this embodiment includes a cluster monitoring module 401, a quality of service control module 402, a degradation policy notification module 403, a service priority configuration module 404, and a request priority configuration module 405. Wherein:

the cluster monitoring module 401 is configured to monitor the operation states of all servers in the server cluster and the service operation state to obtain state data values of each server, specifically including a CPU utilization rate, a memory utilization rate, a bandwidth utilization rate, an average response time of a specified service processing request, and the like, and send the state data values to the degradation policy notification module according to a fixed period.

A service priority configuration module 404, configured to configure service priorities of a plurality of services provided by the server cluster, and notify the degradation policy notification module 403 of the service priorities.

A request priority configuration module 405, configured to configure the request priority of each service, which may be prioritized according to the user level in the request, and notify the degradation policy notification module 403 of the request priority.

A degradation policy notification module 403, configured to notify the service quality control module 402 of a degradation policy according to a configured threshold in combination with specific monitoring content; wherein the configured threshold comprises a first threshold and a second threshold, and the downgrading policy comprises a step downgrading policy for the request and a step downgrading policy for the service.

A service quality control module 402, configured to perform step degradation on a request received by a server according to a state data value of the server and a step degradation policy for the request, so as to perform a denial operation on requests classified into different categories; and performing step degradation according to the step degradation strategy aiming at the service so as to execute closing operation on the service of the partial classification.

In addition, the detailed implementation content of the apparatus for controlling the quality of service of the server cluster in the embodiment of the present invention has been described in detail in the above method for controlling the quality of service of the server cluster, so that the repeated content is not described herein.

Fig. 5 illustrates an exemplary system architecture 500 of a method for controlling a quality of service of a server cluster or an apparatus for controlling a quality of service of a server cluster to which an embodiment of the present invention may be applied.

As shown in fig. 5, the system architecture 500 may include

terminal devices

501, 502, 503, a network 504, and a server 505. The network 504 serves to provide a medium for communication links between the

terminal devices

501, 502, 503 and the server 505. Network 504 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

The user may use the

terminal devices

501, 502, 503 to interact with a server 505 over a network 504 to receive or send messages or the like. The

terminal devices

501, 502, 503 may have installed thereon various communication client applications, such as shopping-like applications, web browser applications, search-like applications, instant messaging tools, mailbox clients, social platform software, etc. (by way of example only).

The

terminal devices

501, 502, 503 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.

The server 505 may be a server providing various services, such as a background management server (for example only) providing support for shopping websites browsed by users using the

terminal devices

501, 502, 503. The backend management server may analyze and perform other processing on the received data such as the product information query request, and feed back a processing result (for example, target push information, product information — just an example) to the terminal device.

It should be noted that the method for controlling the quality of service of the server cluster provided in the embodiment of the present invention is generally executed by the server 505, and accordingly, the apparatus for controlling the quality of service of the server cluster is generally disposed in the server 505.

It should be understood that the number of terminal devices, networks, and servers in fig. 5 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

Referring now to FIG. 6, shown is a block diagram of a computer system 600 suitable for use in implementing a terminal device or server of an embodiment of the present application. The terminal device or the server shown in fig. 6 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present application.

As shown in fig. 6, the computer system 600 includes a Central Processing Unit (CPU)601 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM)602 or a program loaded from a storage section 608 into a Random Access Memory (RAM) 603. In the RAM 603, various programs and data necessary for the operation of the system 600 are also stored. The CPU 601, ROM 602, and RAM 603 are connected to each other via a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.

The following components are connected to the I/O interface 605: an input portion 606 including a keyboard, a mouse, and the like; an output portion 607 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 608 including a hard disk and the like; and a communication section 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the internet. The driver 610 is also connected to the I/O interface 605 as needed. A removable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 610 as necessary, so that a computer program read out therefrom is mounted in the storage section 608 as necessary.

In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 609, and/or installed from the removable medium 611. The above-described functions defined in the system of the present application are executed when the computer program is executed by the Central Processing Unit (CPU) 601.

It should be noted that the computer readable medium shown in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having 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. In the present application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In this application, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules described in the embodiments of the present invention may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as: a processor comprises a cluster monitoring module and a service quality control module. The names of these modules do not form a limitation to the modules themselves in some cases, for example, the cluster monitoring module may also be described as a "module for monitoring status data of each server in the server cluster to obtain a status data value of each server".

As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be separate and not incorporated into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to comprise: monitoring state data of each server in a server cluster to obtain a state data value of each server; controlling the service quality of the server cluster according to the state data value of each server, wherein: for each server, if the state data value of the server is greater than a first threshold value and less than or equal to a second threshold value, performing a rejection operation on the request of the partial category according to a step degradation strategy aiming at the request; and if the state data value of the server is greater than the second threshold value, or the state data value of the server is still greater than the first threshold value and less than or equal to the second threshold value after all the classes of requests are rejected, performing closing operation on the services classified according to the step degradation strategy for the services.

According to the technical scheme of the embodiment of the invention, the state data of each server in the server cluster is monitored to obtain the state data value of each server; controlling the service quality of the server cluster according to the state data value of each server, wherein: for each server, if the state data value of the server is greater than a first threshold value and less than or equal to a second threshold value, executing rejection operation on the requests of the partial category according to a step degradation strategy aiming at the requests; and if the state data value of the server is larger than the second threshold value, or the state data value of the server is still larger than the first threshold value and smaller than or equal to the second threshold value after all the classes of requests are rejected, performing closing operation on the services classified according to the step degradation strategy for the services. The method can timely and quickly solve the problem of service quality reduction of the server cluster caused by various reasons such as burst flow and the like, reduce the range of unavailable service, ensure the service quality of the server cluster and improve the user experience.

The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method of controlling quality of service for a cluster of servers, comprising:

monitoring state data of each server in a server cluster to obtain a state data value of each server;

controlling the service quality of the server cluster according to the state data value of each server, wherein: for each server, if the state data value of the server is greater than a first threshold value and less than or equal to a second threshold value, performing a rejection operation on the request of the partial category according to a step degradation strategy aiming at the request; and if the state data value of the server is greater than the second threshold value, or the state data value of the server is still greater than the first threshold value and less than or equal to the second threshold value after all the classes of requests are rejected, performing closing operation on the services classified according to the step degradation strategy for the services.

2. The method of claim 1, wherein the step of monitoring the status of each server in the cluster of servers is preceded by the step of:

configuring a service priority for a service provided by the server cluster and a request priority for a request of each class of service, the service priority and the request priority indicating an importance of the service and the request, respectively.

3. The method of claim 2, wherein the state data of the server comprises one or more of the following: the CPU utilization rate, the memory utilization rate, the bandwidth utilization rate and the average response time of a specified service processing request of the server are obtained, and the specified service is determined according to the service priority.

4. The method according to claim 3, wherein, in the case that the status data of the server includes the plurality of types of data, each type of data has the corresponding first threshold value and the second threshold value, respectively, wherein,

when the plurality of data values are all smaller than or equal to the corresponding second threshold value and at least one data value is larger than the corresponding first threshold value, indicating that the state data value of the server is larger than the first threshold value and smaller than or equal to the second threshold value;

when at least one of the plurality of data values is greater than the corresponding second threshold value, indicating that the state data value of the server is greater than the second threshold value;

when the plurality of data values are all smaller than or equal to the first threshold value corresponding to each data value, the state data value representing the server is smaller than or equal to the first threshold value.

5. The method of claim 2 or 4, wherein the step of performing a rejection operation on the requests of the partial category according to the step downgrading policy for the requests comprises:

and executing a refusal operation on the requests according to the request priority level one by one from the request of the category with the lowest request priority level until the state data value of the server becomes less than or equal to the first threshold value or the interval range of the first threshold value after the refusal operation is executed on the request of the category, or stopping the refusal operation when the refusal request of the category with higher request priority level does not exist.

6. The method of claim 5, wherein the step of performing a denial operation on the requests of the subset according to the ladder destage policy for the requests further comprises:

if there is a service that has been closed, the closed service is restored.

7. The method according to claim 2 or 4, wherein the step of performing shutdown operations on the classified services according to the step downgrade policy for the services comprises:

and starting from the service of the class with the lowest service priority, executing closing operation on the services according to the service priority class by class until the state data value of the server becomes smaller than or equal to a second threshold value or an interval range of the second threshold value after the closing operation is executed on the service of a certain class, or stopping the closing operation when the service which can be closed and has a higher service priority class does not exist.

8. An apparatus for controlling quality of service for a cluster of servers, comprising:

the cluster monitoring module is used for monitoring the state data of each server in the server cluster to obtain the state data value of each server;

a service quality control module, configured to control the service quality of the server cluster according to the state data value of each server, where: for each server, if the state data value of the server is greater than a first threshold value and less than or equal to a second threshold value, performing a rejection operation on the request of the partial category according to a step degradation strategy aiming at the request; and if the state data value of the server is greater than the second threshold value, or the state data value of the server is still greater than the first threshold value and less than or equal to the second threshold value after all the classes of requests are rejected, performing closing operation on the services classified according to the step degradation strategy for the services.

9. An electronic device, comprising:

one or more processors;

a memory for storing one or more programs,

the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method recited in any of claims 1-7.

10. A computer-readable medium, on which a computer program is stored, which, when being executed by a processor, carries out the method according to any one of claims 1-7.