CN112165436A

CN112165436A - Flow control method, device and system

Info

Publication number: CN112165436A
Application number: CN202011101957.7A
Authority: CN
Inventors: 曾祝青; 胡健鹰; 刘传奇; 王银虎; 丁业凯
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-10-15
Filing date: 2020-10-15
Publication date: 2021-01-01
Anticipated expiration: 2040-10-15
Also published as: CN112165436B

Abstract

The application relates to a flow control method, a flow control device and a flow control system, wherein the method comprises the following steps: in the current period, responding to a service access request aiming at a virtual storage area to be accessed, and determining a target preset storage area corresponding to the virtual storage area to be accessed in a storage cluster; determining a target single connection corresponding to the target preset storage area; sending the service access request to the storage cluster through the target single connection; receiving request processing abnormal state information fed back by the storage cluster through the target single connection; adjusting the number of service access requests sent to the storage cluster through the target single connection based on at least one item of the request handling exception state information fed back through the target single connection in the current period. The flow control method and the flow control device can adjust the flow in real time so as to avoid system function abnormity caused by instantaneous high flow, and the flow control granularity is small, so that the accuracy of flow control is improved.

Description

Flow control method, device and system

Technical Field

The present application relates to the field of internet technologies, and in particular, to a flow control method, apparatus, and system.

Background

The requests to be processed by the system are not controllable randomly, and the processing capacity of the system is limited, so that the flow needs to be controlled according to the processing capacity of the system.

A flow control method commonly used in the related art is to specify a QPS (query per second) threshold for a specific system service, and when a request received by a system exceeds the QPS threshold, deny the service or add the request to a waiting queue for uniform processing; some distributed systems also adopt independent flow control modules to configure specific flow thresholds for each client, monitor the state of the server according to a certain time period, and update the flow thresholds of the clients to realize the pressurization of the client flow when the server flow is too high or abnormal until the server state returns to normal; the existing flow control method has a slow response speed due to long flow, is difficult to solve the problem of high instantaneous flow, and the flow control module generally controls the whole client, so that the flow control is not accurate due to large flow control granularity.

Disclosure of Invention

The technical problem to be solved by the present application is to provide a flow control method, device and system, which can adjust the flow in real time to avoid system function abnormality caused by instantaneous high flow, and the flow control granularity is small, so as to improve the accuracy of flow control.

In order to solve the above technical problem, in one aspect, the present application provides a flow control method, which is applicable to a flow management end, and the method may include:

in the current period, responding to a service access request aiming at a virtual storage area to be accessed, and determining a target preset storage area corresponding to the virtual storage area to be accessed in a storage cluster; the storage cluster comprises a plurality of preset storage areas;

determining a target single connection corresponding to the target preset storage area based on a plurality of preset single connections established between the traffic management terminal and the storage cluster and the corresponding relation between each preset storage area and each preset single connection in the storage cluster;

sending the service access request to the storage cluster through the target single connection;

receiving request processing abnormal state information fed back by the storage cluster through the target single connection;

adjusting the number of service access requests sent to the storage cluster through the target single connection based on at least one item of the request handling exception state information fed back through the target single connection in the current period.

In another aspect, the present application provides another flow control method, which may be applied to a storage cluster, and the method may include:

in the current period, receiving a service access request sent by a flow management end through a target single connection;

processing the service access request, and generating request processing abnormal state information based on a processing result of the service access request;

and feeding back the request processing abnormal state information to the traffic management terminal through the target single connection, so that the traffic management terminal adjusts the number of service access requests sent through the target single connection based on at least one item of request processing abnormal state information fed back through the target single connection in the current period.

In another aspect, the present application provides a flow control device, which may include:

the service access request response module is used for responding to a service access request aiming at a virtual storage area to be accessed in the current period and determining a target preset storage area corresponding to the virtual storage area to be accessed in the storage cluster; the storage cluster comprises a plurality of preset storage areas;

the target single connection determining module is used for determining a target single connection corresponding to the target preset storage area based on a plurality of preset single connections established between a flow management end and the storage cluster and the corresponding relation between each preset storage area and each preset single connection in the storage cluster;

a service access request sending module, configured to send the service access request to the storage cluster through the target single connection;

the state information receiving module is used for receiving the request processing abnormal state information fed back by the storage cluster through the target single connection;

and the flow adjusting module is used for adjusting the quantity of service access requests sent to the storage cluster through the target single connection based on at least one item of request processing abnormal state information fed back through the target single connection in the current period.

In another aspect, the present application provides another flow control device that may include:

the service access request receiving module is used for receiving a service access request sent by the flow management terminal through the target single connection in the current period;

the state information generating module is used for processing the service access request and generating request processing abnormal state information based on the processing result of the service access request;

a status information feedback module, configured to feed back the request processing abnormal status information to the traffic management end through the target single connection, so that the traffic management end adjusts the number of service access requests sent through the target single connection based on at least one item of request processing abnormal status information fed back through the target single connection in the current period.

In another aspect, the present application provides a flow control system that may include a flow control device as described above.

In another aspect, the present application provides an apparatus, which includes a processor and a memory, where at least one instruction or at least one program is stored in the memory, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the flow control method as described above.

In another aspect, the present application provides a computer storage medium, in which at least one instruction or at least one program is stored, and the at least one instruction or the at least one program is loaded by a processor and executes the flow control method as described above.

The embodiment of the application has the following beneficial effects:

in the application, a traffic management end sends a service access request to a storage cluster through target single connection with a target preset storage area; and receiving the request processing abnormal state information fed back by the storage cluster through the target single connection, so that the quantity of the service access requests sent to the storage cluster through the target single connection can be adjusted based on the request processing abnormal state information fed back by the target single connection in the current period. After a service request is sent to a storage cluster through a target single connection, request processing abnormal state information sent by the storage cluster through the target single connection can be directly received, so that the request processing abnormal state information corresponding to the target single connection can be known in real time, corresponding flow adjustment can be timely carried out according to the request state information, service state monitoring and sending of flow adjustment instructions are not needed to be carried out by a flow control server, the flow control response speed is high, and the problem of instantaneous high flow can be solved so as to adjust the flow in real time and avoid system function abnormity caused by the instantaneous high flow; in addition, the sending of the access request and the receiving of the request processing state are carried out through the target single connection, independent flow control can be carried out on each single connection, the single connections are not affected mutually, flow control is not carried out on the whole flow management end, the flow control granularity is small, and therefore the accuracy of the flow control is improved.

Drawings

In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative efforts.

FIG. 1 is a schematic diagram of an implementation environment provided by an embodiment of the present application;

fig. 2 is a flowchart of a flow control method according to an embodiment of the present application;

fig. 3 is a flowchart of a method for determining a target preset storage area according to an embodiment of the present application;

fig. 4 is a flowchart of another method for determining a target preset storage area according to an embodiment of the present application;

fig. 5 is a flowchart of a method for establishing a correspondence between a preset storage area and a preset single connection according to an embodiment of the present application;

fig. 6 is a flowchart of a method for adjusting the number of service access requests sent according to an embodiment of the present application;

fig. 7 is a flowchart of another method for adjusting the number of service access request transmissions according to an embodiment of the present application;

fig. 8 is a flowchart of another method for adjusting the number of service access request transmissions according to an embodiment of the present application;

fig. 9 is a flowchart of another flow control method provided in an embodiment of the present application;

fig. 10 is a schematic view of a cloud hard disk application scenario provided in an embodiment of the present application;

FIG. 11 is a schematic view of a flow control device provided in an embodiment of the present application;

FIG. 12 is a schematic view of another flow control device provided by an embodiment of the present application;

fig. 13 is a schematic structural diagram of an apparatus according to an embodiment of the present application.

Detailed Description

To make the objects, technical solutions and advantages of the present application more clear, the present application will be further described in detail with reference to the accompanying drawings. It is to be understood that the described embodiments are merely a few embodiments of the present application and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or server that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

The application relates to a cloud storage technology (cloud storage), wherein cloud storage is a new concept extended and developed on the cloud computing concept, and a distributed cloud storage system (hereinafter referred to as a storage system) refers to a storage system which integrates a large number of storage devices (storage devices are also referred to as storage nodes) of various different types in a network through functions of cluster application, a grid technology, a distributed storage file system and the like to cooperatively work through application software or an application interface and provides a data storage function and a service access function to the outside. At present, a storage method of a storage system is as follows: logical volumes are created, and when created, each logical volume is allocated physical storage space, which may be the disk composition of a certain storage device or of several storage devices. The client stores data on a certain logical volume, that is, the data is stored on a file system, the file system divides the data into a plurality of parts, each part is an object, the object not only contains the data but also contains additional information such as data identification (ID, ID entry), the file system writes each object into a physical storage space of the logical volume, and the file system records storage location information of each object, so that when the client requests to access the data, the file system can allow the client to access the data according to the storage location information of each object.

Referring to fig. 1, a schematic diagram of an implementation environment provided by an embodiment of the present application is shown, where the implementation environment may include: the user terminal 110, the traffic management terminal 120, and the storage cluster terminal 130 may be directly or indirectly connected between the user terminal 110 and the traffic management terminal 120, and between the traffic management terminal 120 and the storage cluster terminal 130 through wired or wireless communication methods, so as to implement data communication.

Specifically, the user terminal 110 may send a service access request to the traffic management terminal 120; the traffic management terminal 120 determines a corresponding target single connection according to the received service access request, and sends the service access request to the storage cluster terminal 130 through the target single connection; the storage cluster end 130 feeds back the request processing abnormal state information to the traffic management end 120 through the target single connection, and the traffic management end 120 adjusts the number of service access requests sent to the storage cluster end 130 through the target single connection according to the received request processing abnormal state information.

The user terminal 110 may include: the system comprises a smart phone, a tablet computer, a notebook computer, a digital assistant, a smart wearable device, a vehicle-mounted terminal, a server and other types of entity devices, and can also comprise a virtual machine and the like. The operating system running on the user terminal 110 in the embodiment of the present application may include, but is not limited to, an android system, an IOS system, linux, windows, and the like.

The traffic management end 120 and the storage cluster end 130 may specifically be servers, which may be independent physical servers, or may be a server cluster or a distributed system formed by a plurality of physical servers, or may be cloud servers that provide basic cloud computing services such as cloud service, cloud database, cloud computing, cloud functions, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN, and big data and artificial intelligence platform.

In order to solve the problems of slow response speed caused by long flow path of flow control and inaccurate flow control caused by large flow control granularity in the prior art, an embodiment of the present application provides a flow control method, and specifically refer to fig. 2, an execution main body of the method may be the flow management end in fig. 1, and the method may specifically include:

s210, in the current period, responding to a service access request aiming at a virtual storage area to be accessed, and determining a target preset storage area corresponding to the virtual storage area to be accessed in a storage cluster; the storage cluster comprises a plurality of preset storage areas.

The service access request in the embodiment of the present application may be any type of network access request, and when a user needs to perform network access, the user may send a service access request for a to-be-accessed virtual storage area to a traffic management end through a user end, where the to-be-accessed virtual storage area is a virtual storage area specified by the user or a virtual storage area determined according to the service access request of the user, and needs to be corresponding to an actual storage area in a storage cluster, so as to determine a corresponding target preset storage area, specifically refer to fig. 3, which shows a method for determining a target preset storage area, and may include:

s310, determining a virtual start address and a virtual end address of the virtual storage area to be accessed.

S320, converting the virtual starting address and the virtual ending address into an actual starting address and an actual ending address respectively.

S330, determining an actual storage area in the storage cluster based on the actual starting address and the actual ending address.

S340, determining a preset storage area containing the actual storage area in the storage cluster as the target preset storage area.

Taking access to a file system as an example, the corresponding service access request may specifically be a file processing request, and specifically may include a file write request, a file read request, a file delete request, a file copy request, a file cut request, and the like. When a user plans to read a file from a file system, for example, file a in D disk needs to be read, the D disk therein can be accessed through a user side and a read operation is performed on file a, so that a corresponding file read request can be triggered and sent to a traffic management side, where a virtual area to be accessed may be a virtual storage area corresponding to a virtual disk (e.g., D disk), or may be a virtual storage area corresponding to a file to be accessed (e.g., file a), and a virtual start address and a virtual end address of the virtual storage area to be accessed can be correspondingly determined. The D disk can be regarded as a virtual disk, is a disk identifier used for showing a storage position to a user, and is not an actual storage hard disk; the file a is not a real file, but is a file identifier corresponding to the file a, and the real file a is stored in the storage cluster.

According to the corresponding relation between the preset virtual address and the actual address, the virtual starting address and the virtual ending address can be converted into the actual starting address and the actual ending address in the storage cluster, so that the corresponding actual storage area is determined in the storage cluster according to the actual starting address and the actual ending address, and the preset storage area comprising the actual storage area is determined as the target preset storage area, namely the size of the actual storage area is smaller than or equal to that of the target preset storage area.

In the foregoing fig. 3, a corresponding target preset storage area is determined directly based on a corresponding relationship between a virtual address and an actual address, a service access request sent by a user for a virtual storage area to be accessed may further include a service type, and in a specific implementation process, the target preset storage area may also be determined in combination with the service type, please refer to fig. 4, which shows another method for determining the target preset storage area, where the method may include:

s410, determining a target area corresponding to the service type in the storage cluster based on the service type in the service access request aiming at the virtual storage area to be accessed, wherein the target area comprises at least one preset storage area.

S420, determining an actual storage area in the storage cluster based on the preset corresponding relation between the virtual address and the actual address in the target area.

And S430, determining a preset storage area containing an actual storage area as a target preset storage area corresponding to the virtual storage area to be accessed.

The method for determining the virtual address is consistent with the method in fig. 3, and is not described herein again; the method for determining the target preset storage area by combining the service types corresponds to the situation that the storage is performed in a storage cluster in a partitioned mode according to different service types during specific implementation, and the different service types correspond to different preset storage areas.

It should be noted that after the target preset storage area is determined, the service content to be accessed specifically included in the service access request may be determined from the target preset storage area according to the corresponding service identifier; for example, in the above example, the file a needs to be read, and after the target preset storage area corresponding to the virtual storage area is determined, the storage cluster end may read the file a from the target preset storage area according to the identification information of the file a, return the file a to the traffic management end, and then return the file a to the user end by the traffic management end.

S220, determining target single connection corresponding to the target preset storage area based on a plurality of preset single connections established between the flow management end and the storage cluster and the corresponding relation between each preset storage area and each preset single connection in the storage cluster.

A single connection in the embodiments of the present application may refer to a single network connection from end to end; a plurality of preset single connections can be pre-established between the flow management end and the storage cluster, each preset storage area in the storage cluster is provided with a corresponding preset single connection, and one preset single connection can correspond to one or more preset storage areas; specifically referring to fig. 5, a method for establishing a corresponding relationship between a preset storage area and a preset single connection is shown, where the method includes:

and S510, setting the service type corresponding to each preset storage area in the storage cluster.

Generally, each preset storage area is set to correspond to the same service type.

S520, distributing corresponding preset single connection for each service type based on the plurality of preset single connections between the flow management end and the storage cluster.

For different preset single connections, it may have different connection network ports, different routes, different network conditions, etc., so that for each service type, it may adapt to the corresponding preset single connection; for example, for a service type with a high processing requirement, a preset single connection with a better network connection condition can be allocated, and for a service type with a low processing requirement, a preset single connection with a general network connection condition can be allocated; therefore, the appropriate preset single connection can be distributed to different types of services as required, and the efficiency of the service is improved under the condition that the normal service is not influenced.

S530, determining preset single connection corresponding to each preset storage area based on the service type corresponding to each preset storage area.

And S540, establishing a corresponding relation between each preset storage area and each preset single connection in the storage cluster based on each preset storage area and each preset single connection corresponding to each preset storage area.

Taking the above file system access as an example, the service type may be a file type, and the file types in this embodiment may include, but are not limited to: video files, audio files, picture files, text files, executable files, and the like; the files of the same file type can be stored in the same preset storage area, the condition that the files of the same type are stored in a plurality of preset storage areas can exist, the preset storage areas in which the files of different file types are stored can correspond to different preset single connections, and therefore the corresponding relation between each preset storage area and each preset single connection is established.

And S230, sending the service access request to the storage cluster through the target single connection.

And S240, receiving the request fed back by the storage cluster through the target single connection to process abnormal state information.

After the traffic management terminal sends each service access request to the storage cluster, request processing state information for the service access request, which is fed back by the storage cluster based on the corresponding target single connection, is received, wherein the request processing state information may be request processing normal state information or request processing abnormal state information; the subsequent related operation is performed based on the request processing abnormal state information.

S250, based on at least one item of request processing abnormal state information fed back through the target single connection in the current period, adjusting the quantity of service access requests sent to the storage cluster through the target single connection.

In the current period, the number of service access requests received by the traffic management end may be one or more, so that the number of corresponding determined target single connections may also be one or more, and the number of determined target single connections may be less than or equal to the number of received service access requests.

Referring to fig. 6, it shows a method for adjusting the sending number of service access requests, where request processing exception status information includes a first type of exception status information related to the processing status of the storage cluster; the method specifically comprises the following steps:

s610, according to the preset abnormal grade division information of the first type abnormal state information, respectively counting the number of items of the first type abnormal state information belonging to each abnormal grade in the current period.

For example, the exception levels may be classified into level 1, level 2, … …, and level N, each having a level index corresponding thereto; each item of first-type abnormal state information in the current period also has a corresponding index, and the abnormal grade to which each item of first-type abnormal state information belongs can be determined through index matching, so that the number of items of first-type abnormal states belonging to each abnormal grade can be counted. Taking the first type abnormal state information as the request response delay height as an example, the corresponding grade index may be a specific response delay length, for example, multiple grades such as 10ms \100ms \1000ms and the like; when the abnormal level is determined specifically, the level corresponding to the level index closest to the index of the current item of the first type state information can be determined as the abnormal level to which the current item of the first type state information belongs.

It should be noted that the first-type abnormal state information may also include a plurality of different types of first-type abnormal state information, and when performing specific processing, the number of items belonging to each abnormal level may be counted for each type of first-type abnormal state information.

S620, weighting and calculating are carried out on the basis of the number of items of the first type abnormal state information belonging to each abnormal level in the current period and the weight corresponding to the abnormal level, and a first state numerical value is obtained.

When the first type abnormal state information comprises one type of first type abnormal state information, carrying out weighting and calculation according to the number of terms of the first type abnormal state information belonging to each abnormal level and the weight of the corresponding abnormal level to obtain a first state numerical value; when the first type abnormal state information comprises a plurality of types of first type abnormal state information, respectively calculating a first state value corresponding to each type of first type abnormal state information based on the method, thereby obtaining a plurality of first state values, and performing preset calculation on the plurality of first state values to obtain a final first state value; the specific preset calculation may be an average calculation, a weighted sum calculation, and the like.

S630, matching the first state value with first threshold values of multiple levels, and determining a target first threshold value matched with the first state value; wherein the first state value is greater than or equal to the target first threshold.

The target first threshold may specifically be a threshold that is smaller than the first state value and closest to the first state value.

And S640, determining a first descending gradient corresponding to the target first threshold, and reducing the number of service access requests sent to the storage cluster through the target single connection in a preset time period based on the first descending gradient.

For each level of the first threshold, there are corresponding descending gradients and ascending gradients, and since the current step is executed after the request processing abnormal state information is received, there is no request processing abnormal state information before, so that the current cycle can be regarded as an initial cycle, the flow rate needs to be reduced on the original basis, and what needs to be determined is the descending gradient corresponding to the target first threshold. Specifically, the sending number of the service access requests in the preset time period may be adjusted based on the preset sending number of the service access requests and the first descending gradient.

After the service access request is sent to the storage cluster through the target single connection, the number of the service access requests sent to the storage cluster through the target single connection in a preset time period can be adjusted in time according to the first type abnormal state information fed back by the storage cluster through the target single connection, and therefore the condition of system function abnormality caused by long flow control process and high instantaneous flow can be avoided; in addition, for different target single connections, the number of service access requests can be adjusted by the same method, the target single connections are not affected mutually, and the flow control granularity is small due to the adjustment of the number of the service access requests of the single connections, so that the flow control effect is improved.

After the sending number of the service access requests in the current period is adjusted, the next period can adjust the number of the service access requests sent to the storage cluster in a preset time period based on the adjustment details of the current period and the request processing abnormal state information received by the next period; referring specifically to fig. 7, another method for adjusting the number of service access requests to be sent is shown, where the method may include:

and S710, in the next period of the current period, according to the preset abnormal grade classification information of the first type abnormal state information, respectively counting the number of items of the first type abnormal state information belonging to each abnormal grade in the next period.

S720, weighting and calculating are carried out on the basis of the number of items of the first type abnormal state information belonging to each abnormal level in the next period and the weight corresponding to the abnormal level, and a second state value is obtained.

The calculation method of the second state value is the same as the calculation method of the first state value, and is not described herein again.

S730, when the second state value is smaller than the target first threshold value, matching the second state value with first threshold values of multiple levels, and determining a target second threshold value matched with the second state value; wherein the second state value is less than the target second threshold.

First, the second state value is compared with the target first threshold, when the second state value is smaller than the target first threshold, the situation that the abnormal state occurs in the period is relieved, at the moment, threshold matching can be carried out again, and the target second threshold is determined.

The target second threshold may specifically be a threshold that is greater than and closest to the second state value.

S740, determining a first lifting gradient corresponding to the target second threshold, and increasing the quantity of service access requests sent to the storage cluster through the target single connection in a preset time period based on the first lifting gradient.

Since the abnormal state is relieved, the number of the service requests can be properly increased, and the increase gradient corresponding to the target second threshold needs to be determined; specifically, the sending number of the service access requests in the preset time period may be adjusted based on the adjusted sending number of the service access requests in the previous period and the first gradient.

Therefore, after the sending number of the service access requests in the preset time period is adjusted, the abnormal state can be monitored in the next period, whether the abnormal state is relieved or not is determined, and when the abnormal state is relieved, the sending number of the service access requests in the preset time period can be properly increased based on a corresponding flow control adjustment strategy in order to ensure the processing efficiency of the service requests; when there is no release, the number of service access request transmissions in a preset time period may be further reduced based on a corresponding flow control adjustment policy, so as to ensure normal operation of the service system and avoid system crash, which may be specifically implemented by the method shown in fig. 6.

Referring to fig. 8, it shows another method for adjusting the sending number of service access requests, which is directed to a case that the request processing exception state information includes a second type exception state information related to the service access request, and specifically, the method may include:

and S810, when the second type abnormal state information is received, adjusting the quantity of the service access requests sent to the storage cluster through the target single connection in a preset time period to a first preset quantity.

S820, after the number of the service access requests sent to the storage cluster through the target single connection in the preset time period is adjusted to be the preset number, when the number of the items continuously receiving the normal state information reaches a second preset number, the number of the service access requests sent to the storage cluster through the target single connection is increased based on a second lifting gradient.

As the second type abnormal state is a state which may cause that the request cannot be processed, in the specific implementation process, as long as the traffic management end receives the second type abnormal state information through the target single connection, the sending number of the service access requests is directly adjusted to the first preset number; and subsequently, when a second preset number of requests for processing the normal state information items are continuously received, the sending number of the service access requests in a preset time period is properly increased according to a corresponding flow control adjustment strategy.

Through the method for adjusting the number of service access request transmissions shown in fig. 8, a real-time response can be made according to the currently received second-type abnormal state, so that the number of service access request transmissions in the subsequent preset time period of the current time node can be adjusted in time.

An embodiment of the present application further provides another flow control method, an execution subject of which may be a storage cluster end in fig. 1, where the method specifically includes:

s910, in the current period, receiving a service access request sent by a traffic management end through a target single connection.

S920, the service access request is processed, and request processing abnormal state information is generated based on the processing result of the service access request.

According to the processing result of the service access request, the corresponding request processing abnormal state information, such as the first type abnormal state information related to the storage cluster processing state or the second type abnormal state information related to the service access request, can be determined.

S930, feeding back the request processing abnormal state information to the traffic management end through the target single connection, so that the traffic management end adjusts the number of service access requests sent through the target single connection based on at least one item of request processing abnormal state information fed back through the target single connection in the current period.

Through the interaction between the traffic management end and the storage cluster end, the traffic from the traffic management end to the storage cluster end can be controlled. The storage cluster terminal processes the service access request sent by the traffic management terminal through the target single connection and feeds back corresponding request processing abnormal state information through the target single connection, so that the traffic management terminal determines a corresponding flow control strategy according to the received request processing abnormal state information, and the quantity of the service access requests sent to the storage cluster terminal through the target single connection is adjusted.

The storage cluster processes the abnormal state information through the target single connection feedback request, so that the flow management end can timely adjust the number of service access requests sent to the storage cluster through the target single connection in a preset time period, and the condition of system function abnormity caused by long flow control process and high instantaneous flow can be avoided; in addition, for different target single connections, the number of service access requests can be adjusted by the same method, the target single connections are not affected mutually, and the flow control granularity is small due to the adjustment of the number of the service access requests of the single connections, so that the flow control effect is improved.

The present application may be applied to various scenes related to a distributed cluster, and a specific implementation flow of the present application is specifically described below by taking a cloud hard disk application scene as an example, please refer to fig. 10, which shows a schematic view of a cloud hard disk application scene, where a cloud hard disk client is the traffic management end in this embodiment, the cloud hard disk client provides a cloud hard disk for a virtual machine to use, the virtual machine initiates an IO (Input/Output) read-write request to a cloud hard disk vdisk1, and after receiving the IO read-write request, the cloud hard disk client finds a specific connection with a storage cluster, such as connection 1, and issues the IO read-write request; if the storage machine 1 corresponding to the connection 1 has an abnormality such as insufficient resources due to excessive requests and the like in the process of processing the IO read-write request, a relevant error code is returned; if the cloud hard disk client side detects that the corresponding connection returns an error code or the request delay is high, the flow threshold value of the corresponding connection is reduced, and the purpose of controlling the flow in real time is achieved; after a period of time, for example, after 100ms, if there is no error code and the delay is normal, the flow threshold is slowly recovered to prevent IO from frequently jittering.

For each connection, performing state monitoring when a request is initiated, detecting whether state abnormality exists or not when the request responds, and if the state abnormality exists, triggering a corresponding flow control strategy for the connection; some negative feedback states and corresponding flow control strategies may be defined specifically, and of course, the negative feedback states and the corresponding flow control strategies may be adjusted by adding or deleting according to actual conditions of the service.

The negative feedback states may include:

1. request response latency is high: if the time difference between the initiation and the response of a request exceeds 10ms, defining the time delay as high time delay; the high delay is graded, and the grades can be 10ms \100ms \1000ms and the like.

2. The available resources of the server are lower than the threshold value: if the storage cluster end with too many requests uses many caches, the residual cache resources are lower than 5% of the original cache resources; the CPU of the specific key thread at the storage cluster end occupies more than 90 percent continuously; the hard disk usage rate used by the server is continuously 100%, the service response delay is continuously at a higher level, and the like.

3. And the storage cluster end processes the request timeout: the timeout mechanism is used inside the storage cluster to monitor each incoming request to determine whether the request is processed for timeout, which may be caused by an excessive number of requests in the storage cluster or a large number of retry operations inside the storage cluster that result in a new incoming request not being processed and resulting in timeout.

4. Request error: and the storage cluster end returns a request error message when recognizing that the request format is wrong, has no authority and exceeds the limit.

5. The requested resource is busy: if the disk is in a busy state, and the request cannot be processed if the disk is abnormal, the state is returned.

Wherein the feedback states 1, 2, and 3 can be regarded as a first type of abnormal state information related to the processing state of the storage cluster, and the feedback states 4 and 5 can be regarded as a second type of abnormal state information related to the service access request.

The flow control policy may include:

1. for any state of 1, 2 and 3 of the negative feedback state, counting the number cnt and the severity of the abnormal state within a plurality of seconds (such as higher request response delay, larger severity weight, and the like), setting a threshold as threshold, and reducing the flow threshold of the connection, such as half of the original flow threshold, if cnt weight > is equal to threshold; when the next period cnt weight is less than threshold, the flow threshold is slowly restored, for example, the flow threshold is raised by 50% once; there may be multiple levels of threshold, with different thresholds corresponding to different threshold lowering and raising steps.

2. Aiming at 4 and 5 serious error states in the negative feedback state, the flow threshold value is directly reduced to be as low as 1, then the flow threshold value is slowly recovered according to the feedback state of the subsequent request, and if the subsequent request continues to be normal for a plurality of times, the flow threshold value is increased by a certain gradient, such as 20%.

In a cloud hard disk application scenario, a cloud hard disk client actively adjusts a flow threshold corresponding to a single connection by using feedback state information such as a message code or delay responded by a storage cluster end through the corresponding single connection, for example, a flow contraction response of a 100ms level can be achieved by the flow control method provided in the embodiment of the application, compared with a flow control response of an independent flow control system with a second level or even a higher delay in the prior art, the flow control method in the application has a faster flow control response speed, so that the problem of instantaneous high flow can be solved, and the problem of abnormal function of a server end, frequent QPS jitter and the like caused by the transient high flow and other scenarios can be avoided by adjusting the flow in real time.

In the application, a traffic management end sends a service access request to a storage cluster through target single connection with a target preset storage area; and receiving the request processing abnormal state information fed back by the storage cluster through the target single connection, so that the quantity of the service access requests sent to the storage cluster through the target single connection can be adjusted based on the request processing abnormal state information fed back by the target single connection in the current period. After a service request is sent to a storage cluster through a target single connection, request processing abnormal state information sent by the storage cluster through the target single connection can be directly received, so that the request processing abnormal state information corresponding to the target single connection can be known in real time, corresponding flow adjustment can be timely carried out according to the request state information, service state monitoring and sending of flow adjustment instructions are not needed to be carried out by a flow control server, the flow control response speed is high, and the flow adjustment can be carried out in real time to avoid system function abnormity caused by instantaneous high flow; in addition, the sending of the access request and the receiving of the request processing state are carried out through the target single connection, independent flow control can be carried out on each single connection, the single connections are not affected mutually, flow control is not carried out on the whole flow management end, the flow control granularity is small, and therefore the accuracy of the flow control is improved.

An embodiment of the present application further provides a flow control device, which can be applied to a flow management end, and specifically refer to fig. 11, the flow control device can include:

a service access request response module 1110, configured to respond to a service access request for a to-be-accessed virtual storage area in a current period, and determine a target preset storage area corresponding to the to-be-accessed virtual storage area in a storage cluster; the storage cluster comprises a plurality of preset storage areas;

a target single connection determining module 1120, configured to determine, based on a plurality of preset single connections established between a traffic management end and the storage cluster and a corresponding relationship between each preset storage area and each preset single connection in the storage cluster, a target single connection corresponding to the target preset storage area;

a service access request sending module 1130, configured to send the service access request to the storage cluster through the target single connection;

a status information receiving module 1140, configured to receive request processing exception status information fed back by the storage cluster through the target single connection;

a traffic adjusting module 1150, configured to adjust the number of service access requests sent to the storage cluster through the target single connection based on at least one item of the request handling exception status information fed back through the target single connection in the current period.

The service access request response module 1110 includes:

the virtual address determining module is used for determining a virtual starting address and a virtual ending address of the virtual storage area to be accessed;

the address conversion module is used for converting the virtual starting address and the virtual ending address into an actual starting address and an actual ending address respectively;

an actual storage area determination module to determine an actual storage area in the storage cluster based on the actual start address and the actual end address;

and a target preset storage area determining module, configured to determine that a preset storage area, which includes the actual storage area, in the storage cluster is the target preset storage area.

Further, the apparatus may further include:

the service type setting module is used for setting a service type corresponding to each preset storage area in the storage cluster;

the preset single connection allocation module is used for allocating corresponding preset single connection for each service type based on a plurality of preset single connections between the flow management end and the storage cluster;

the first determining module is used for determining preset single connection corresponding to each preset storage area based on the service type corresponding to the preset storage area;

and the corresponding relation establishing module is used for establishing the corresponding relation between each preset storage area and each preset single connection in the storage cluster based on each preset storage area and each preset single connection corresponding to each preset storage area.

Requesting processing exception state information including a first type of exception state information associated with the storage cluster processing state; the flow adjustment module 1150 may thus include:

the first statistical module is used for respectively counting the number of items of the first type abnormal state information belonging to each abnormal level in the current period according to the preset abnormal level classification information of the first type abnormal state information;

the first calculation module is used for weighting and calculating based on the number of terms of the first type abnormal state information belonging to each abnormal level in the current period and the weight corresponding to the abnormal level to obtain a first state numerical value;

a first threshold determination module, configured to match the first state value with first thresholds of multiple levels, and determine a target first threshold that matches the first state value; wherein the first state value is greater than or equal to the target first threshold;

a first adjusting module, configured to determine a first descending gradient corresponding to the target first threshold, and reduce, based on the first descending gradient, a number of service access requests sent to the storage cluster through the target single connection within a preset time period.

Further, the apparatus may further include:

the second statistical module is used for respectively counting the number of items of the first type abnormal state information belonging to each abnormal level in the next period according to the preset abnormal level classification information of the first type abnormal state information in the next period of the current period;

the second calculation module is used for weighting and calculating in the next period based on the number of items of the first type abnormal state information belonging to each abnormal level and the weight corresponding to the abnormal level to obtain a second state numerical value;

a second threshold determination module, configured to, when the second state value is smaller than the target first threshold, match the second state value with first thresholds of multiple levels, and determine a target second threshold that matches the second state value; wherein the second state value is less than the target second threshold;

a second adjusting module, configured to determine a first boosting gradient corresponding to the target second threshold, and increase, based on the first boosting gradient, a number of service access requests to be sent to the storage cluster through the target single connection within a preset time period.

The request processing abnormal state information comprises second type abnormal state information related to the service access request; the flow adjustment module 1150 may thus further include:

a third adjusting module, configured to adjust, when the second type abnormal state information is received, the number of service access requests sent to the storage cluster through the target single connection within a preset time period to a first preset number;

and a fourth adjusting module, configured to, after the number of service access requests sent to the storage cluster through the target single connection within a preset time period is adjusted to a preset number, increase, based on a second gradient, the number of service access requests sent to the storage cluster through the target single connection when the number of items continuously receiving requests for processing normal state information reaches a second preset number.

Another flow control device, which may be applied to a storage cluster, is provided in an embodiment of the present application, and referring to fig. 12, the device may include:

a service access request receiving module 1210, configured to receive, in a current period, a service access request sent by a traffic management end through a target single connection;

a status information generating module 1220, configured to process the service access request, and generate request processing exception status information based on a processing result of the service access request;

a status information feedback module 1230, configured to feed back the request processing abnormal status information to the traffic management end through the target single connection, so that the traffic management end adjusts the number of service access requests sent through the target single connection based on at least one item of request processing abnormal status information fed back through the target single connection in the current period.

The device provided in the above embodiment can execute the corresponding method in the embodiment of the present application, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in the above embodiments may be referred to a method provided in any of the embodiments of the present application.

Embodiments also provide a flow control system, which may include a flow management end and a storage cluster end, wherein the flow management end may include the flow control device in fig. 11, and the storage cluster end may include the flow control device in fig. 12.

The embodiment of the present application further provides a computer-readable storage medium, where at least one instruction or at least one program is stored in the storage medium, and the at least one instruction or the at least one program is loaded by a processor and executes any one of the methods described in the embodiment.

Embodiments of the present application also provide a computer program product or computer program comprising computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computer device to perform any one of the methods described above in the embodiments.

Referring to fig. 13, the apparatus 1300 may have a large difference due to different configurations or performances, and may include one or more Central Processing Units (CPUs) 1322 (e.g., one or more processors) and a memory 1332, and one or more memoriesA storage medium 1330 (e.g., one or more mass storage devices) that stores an application 1342 or data 1344. Memory 1332 and storage media 1330 may be, for example, transitory or persistent storage. The program stored on the storage medium 1330 may include one or more modules (not shown), each of which may include a sequence of instructions operating on a device. Still further, central processor 1322 may be provided in communication with storage medium 1330 for performing a series of instruction operations on storage medium 1330 on device 1300. The apparatus 1300 may also include one or more power supplies 1326, one or more wired or wireless network interfaces 1350, one or more input-output interfaces 1358, and/or one or more operating systems 1341, such as a Windows Server^TM，Mac OS X^TM，Unix^TM，Linux^TM，FreeBSD^TMAnd so on. Any of the methods described above in this embodiment can be implemented based on the apparatus shown in fig. 13.

The present specification provides method steps as described in the examples or flowcharts, but may include more or fewer steps based on routine or non-inventive labor. The steps and sequences recited in the embodiments are but one manner of performing the steps in a multitude of sequences and do not represent a unique order of performance. In the actual system or interrupted product execution, it may be performed sequentially or in parallel (e.g., in the context of parallel processors or multi-threaded processing) according to the embodiments or methods shown in the figures.

The configurations shown in the present embodiment are only partial configurations related to the present application, and do not constitute a limitation on the devices to which the present application is applied, and a specific device may include more or less components than those shown, or combine some components, or have an arrangement of different components. It should be understood that the methods, apparatuses, and the like disclosed in the embodiments may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a division of one logic function, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or unit modules.

Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

Those of skill would further appreciate that the various illustrative components and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.

Claims

1. A flow control method is applied to a flow management end, and the method comprises the following steps:

2. The flow control method according to claim 1, wherein the determining a target preset storage area corresponding to the virtual storage area to be accessed in the storage cluster comprises:

determining a virtual start address and a virtual end address of the virtual storage area to be accessed;

converting the virtual start address and the virtual end address into an actual start address and an actual end address respectively;

determining an actual storage area in the storage cluster based on the actual start address and the actual end address;

and determining a preset storage area containing the actual storage area in the storage cluster as the target preset storage area.

3. The flow control method according to claim 1, characterized in that the method further comprises:

setting a service type corresponding to each preset storage area in the storage cluster;

distributing corresponding preset single connection for each service type based on a plurality of preset single connections between the flow management end and the storage cluster;

determining preset single connection corresponding to each preset storage area based on the service type corresponding to the preset storage area;

and establishing a corresponding relation between each preset storage area and each preset single connection in the storage cluster based on each preset storage area and each preset single connection corresponding to each preset storage area.

4. The traffic control method according to claim 1, wherein the request handling exception state information comprises a first type exception state information relating to the storage cluster handling state;

the adjusting the number of service access requests sent to the storage cluster through the target single connection based on at least one item of the request handling exception state information fed back through the target single connection in the current period comprises:

respectively counting the number of items of the first-type abnormal state information belonging to each abnormal level in the current period according to the preset abnormal level classification information of the first-type abnormal state information;

weighting and calculating based on the number of items of the first type abnormal state information belonging to each abnormal level in the current period and the weight corresponding to the abnormal level to obtain a first state numerical value;

matching the first state value with first thresholds of a plurality of levels, and determining a target first threshold matched with the first state value; wherein the first state value is greater than or equal to the target first threshold;

determining a first descending gradient corresponding to the target first threshold, and reducing the number of service access requests sent to the storage cluster through the target single connection within a preset time period based on the first descending gradient.

5. The traffic control method according to claim 4, wherein the adjusting the number of service access requests sent to the storage cluster via the target single connection based on the at least one item of request handling exception status information fed back via the target single connection in the current cycle further comprises:

in the next period of the current period, according to the preset abnormal grade classification information of the first type abnormal state information, respectively counting the number of items of the first type abnormal state information belonging to each abnormal grade in the next period;

weighting and calculating based on the number of items of the first type abnormal state information belonging to each abnormal level in the next period and the weight corresponding to the abnormal level to obtain a second state numerical value;

when the second state value is smaller than the target first threshold, matching the second state value with first thresholds of multiple levels, and determining a target second threshold matched with the second state value; wherein the second state value is less than the target second threshold;

and determining a first lifting gradient corresponding to the target second threshold, and increasing the number of service access requests sent to the storage cluster through the target single connection within a preset time period based on the first lifting gradient.

6. The traffic control method according to claim 1, wherein the request handling exception state information includes a second type exception state information related to the service access request;

when the second type abnormal state information is received, the number of service access requests sent to the storage cluster through the target single connection in a preset time period is adjusted to a first preset number;

after the number of the service access requests sent to the storage cluster through the target single connection in the preset time period is adjusted to be the preset number, when the number of the items continuously receiving the request processing normal state information reaches a second preset number, the number of the service access requests sent to the storage cluster through the target single connection is increased based on a second lifting gradient.

7. A traffic control method, applied to a storage cluster end, the method comprising:

8. A flow control device, comprising:

9. A flow control device, comprising:

10. A flow control system, comprising:

a flow control device as claimed in claim 8, and a flow control device as claimed in claim 9.