CN107645519B

CN107645519B - Data processing method and system, client and server

Info

Publication number: CN107645519B
Application number: CN201610579241.5A
Authority: CN
Inventors: 朱家稷
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2016-07-21
Filing date: 2016-07-21
Publication date: 2021-08-06
Anticipated expiration: 2036-07-21
Also published as: CN107645519A

Abstract

The embodiment of the application discloses a data processing method and system, a client and a server. The method comprises the following steps: a client sends a reading request to a processing server; the processing server receives the reading request, and processes the reading request when the processing server determines that the data state of the processing server is consistent with that of the main server; wherein the processing server is subordinate to the main server. The embodiment of the application can reduce the read request processing burden of the main server, utilize the idle time of the slave server, reduce the processing delay and improve the read request processing efficiency of the high-consistency communication system.

Description

Data processing method and system, client and server

Technical Field

The present application relates to the field of information processing technologies, and in particular, to a data processing method and system, a client, and a server.

Background

In a communication system formed by a client and a server, the client or an application installed on the client often needs to send a write request and a read request to the server, the write request often needs to change a data state of the server, for example, operations such as adding, deleting, modifying the data state of the server, and the read request often does not need to change the data state of the server, and the server completes processing and returns a processing result to the client after receiving the write request or the read request.

For communication systems requiring multiple servers to operate in parallel, such as distributed systems, the demand for a high-consistency environment is often high. In order to maintain high consistency of data states among a plurality of servers in a communication system, a Paxos algorithm, a RAFT algorithm, a zookeeper zab protocol algorithm, and the like can be used for implementation. For example, the Paxos algorithm may select a master server from N servers and take the other servers as slaves. The master server can process the write request and the read request, and when the write request is processed, the master server also sends a data state change log to the slave servers, and if N/2 slave servers successfully accept the data state change, the master server can determine that the write request is successfully processed and return a request result to the client. Slave servers that have not successfully received a log of data state changes can asynchronously update the data state in the background without affecting the front-end write request response. Heartbeat detection is arranged between the master server and the slave servers, and if the master server is abnormal, the slave servers initiate Paxos elections again to select a new master server after finding that the heartbeat is overtime.

In the prior art, a client sends a read request, and after receiving the read request, a master server may select a slave server to process the read request. And after the slave server processes the read request, returning a processing result to the client. Thus, the master server can utilize the processing power of the slave servers with less processing overhead. However, in the process of processing the read request, the master server is required to forward the read request to the slave server, and the overall processing capacity is limited by the upper connection limit and the forwarding capacity of the master server. In addition, in the processing process, the conventional processing mode can increase one time of RPC communication from the master server to the slave server, increase the processing delay of the read request and reduce the processing efficiency of the read request.

Disclosure of Invention

An object of the embodiments of the present application is to provide a data processing method and system, a client, and a server, which can reduce processing delay in a read request process and improve processing efficiency of the read request.

The embodiment of the application provides a data processing method and system, a client and a server, which are realized as follows:

a method of data processing, the method comprising:

a client sends a reading request to a processing server;

the processing server receives the reading request, and processes the reading request when the processing server determines that the data state of the processing server is consistent with that of the main server; wherein the processing server is subordinate to the main server.

A method of data processing, the method comprising:

receiving a reading request of a client;

and processing the read request when the data state is determined to be consistent with the data state of the main server.

A method of data processing, the method comprising:

sending a read request to a processing server;

and receiving a processing result of the processing server on the read request.

A server, the server comprising:

the receiving unit is used for receiving a reading request of a client;

and the processing unit is used for processing the read request when the data state is consistent with the data state of the main server.

A client, the client comprising:

a read request sending unit for sending a read request to the processing server;

and the processing result receiving unit is used for receiving the processing result of the reading request by the processing server.

A data processing system, the system comprising:

the processing server is used for receiving a reading request of the client; and processing the read request when the data state of the main server is determined to be consistent.

According to the data processing method and system, the client and the server, the client can directly send the reading request to the slave server, and after the slave server receives the reading request, when the data state of the slave server is consistent with that of the master server, the slave server can directly process the reading request. Compared with the prior art, the client does not need to send a read request to the master server, and the read request is distributed to the slave servers by the master server for processing. The read request processing method provided by the application can reduce the read request processing burden of the master server, utilize the idle time of the slave server, and simultaneously, the master server is not required to work in the key processing process of the read request, so that the processing delay is not increased, and the read request processing efficiency of the high-consistency communication system is improved. In addition, the method of the embodiment can enable a plurality of slave servers in the high-consistency communication system to process the read requests simultaneously, the maximum processing capacity of the communication system can be expanded to the sum of the processing capacities of all the processing servers, and the read request processing capacity of the high-consistency communication system is expanded.

Drawings

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

FIG. 1 is a schematic method flow diagram of one embodiment of a data processing method provided herein;

FIG. 2 is a flowchart of one embodiment of a method for determining whether a processing server is bound to a primary server provided herein;

FIG. 3 is a method flow diagram of one embodiment of a method for handling a read request processing failure as provided herein;

FIG. 4 is a schematic diagram of an application scenario of the data processing method provided in the present application;

FIG. 5 is a method flow diagram of another embodiment of a read request processing method provided herein;

FIG. 6 is a flowchart of one embodiment of a method for determining whether a processing server is bound to a primary server provided herein;

FIG. 7 is a block diagram illustrating an exemplary server module according to the present disclosure;

FIG. 8 is a block diagram illustrating an exemplary processing unit;

FIG. 9 is a method flow diagram of another embodiment of a read request processing method provided herein;

FIG. 10 is a method flow diagram of one embodiment of a method for handling a read request processing failure as provided herein;

FIG. 11 is a block diagram illustrating an exemplary client architecture;

fig. 12 is a schematic block diagram of another embodiment of the client provided in the present application.

Detailed Description

In order to make those skilled in the art better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the 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.

The data processing method described in the present application is described in detail below with reference to the drawings. Fig. 1 is a schematic method flow diagram of an embodiment of a data processing method provided in the present application. Although the present application provides method steps as shown in the following examples or figures, more or fewer steps may be included in the method based on conventional or non-inventive efforts. In the case of steps where no necessary causal relationship exists logically, the order of execution of the steps is not limited to that provided by the embodiments of the present application. In actual data processing, the methods can be executed sequentially or in parallel according to the embodiments or the methods shown in the figures (for example, a parallel processor or a multi-thread processing environment, or even a distributed processing environment).

Specifically, as shown in fig. 1, an embodiment of a data processing method provided in the present application may include:

s1: the client sends a read request to the processing server.

In this embodiment, the client may be a terminal device capable of accessing a communication network based on a network protocol or application software running on the terminal device, such as a mobile smart phone, a computer, a pay-for-all client, and the like. The client may detect an operation from an upper layer of the application, may trigger generation of a read request or a write request based on the operation, and thereafter, may send the read request to a processing server or a host server and send the write request to the host server. The processing server and the main server comprise hardware equipment with a data information processing function and necessary software for driving the hardware equipment to work, and the processing server and the main server can analyze and process the received write request or read request and feed back corresponding request processing results to the client. Wherein, the processing server can process the read request, and the main server can process the write request and the read request. The processing server, the main server, may provide a predetermined port through which requests of clients may be received. For example, the processing server and the main server can perform network data interaction with the client based on a network protocol such as HTTP, TCP/IP or FTP and the network communication module.

In this embodiment, the data processing method may be applied to a high-consistency communication system formed by a main server and a processing server, where the processing server is subordinate to the main server, and the main server may be obtained by election through an algorithm such as Paxos. As can be seen from the above, the main server may process a read request and a write request sent by a client, where the write request may change a data state of the main server, the main server may synchronize a data state change log to the processing server after processing the write request, and return a processing result to the client after the main server determines that the system completes synchronization. For example, in a system including N processing servers, when at least N/2 processing servers successfully receive the data state change log, it may be determined that the synchronization of the communication system is successful, and at this time, the main server may return the processing result to the client. It should be noted that the high-consistency communication system may include one or more main servers, and when the high-consistency communication system includes a plurality of main servers, consistency of data states may be maintained between the main servers through internal communication.

The processing server, which is a slave to the primary server, generally does not process write requests, but the processing server can obtain a data state change log of the primary server and thus can maintain data state consistency with the primary server. The processing server may process the read request when the processing server maintains data state consistency with the primary server.

In one embodiment of the present application, the processing server may include a processing server selected by the client from a provided server set. The server set may be used to store names of processing servers in a high-consistency communication system and corresponding address information, which may include IP addresses or access ports of the slave servers. The client may obtain the server set, and may select a processing server from the server set and send a read request to the selected processing server through address information of the processing server in the server set. For example, in a communication system based on the TCP/IP protocol, the client may obtain an IP address and a port of a processing server selected from the server set, and initialize a TCP connection to the processing server by creating a socket object. Based on the created TCP connection, the client may send a read request to the processing server.

In this embodiment, the client may send the read request to the processing server, and compared with the prior art, the client may not need to send the read request to the main server, and then the main server allocates the read request to the processing server for processing. The processing server can be used for processing the read request, and the processing pressure of the read request of the main server is relieved.

S2: the processing server receives the reading request, and processes the reading request when the processing server determines that the data state of the processing server is consistent with that of the main server; wherein the processing server is subordinate to the main server.

Generally, in a high-consistency communication system, when the processing server is subordinate to the main server, it is necessary that the data states of the main server and the processing server are kept consistent. However, when the processing server is abnormal or the system is delayed, the processing server often cannot keep the data state consistent with the main server. Therefore, the processing server can confirm whether the data state of the processing server is consistent with that of the main server after receiving the read request of the client. When the processing server determines that the data state of the processing server is consistent with the data state of the main server, it may be determined that the processing server and the main server successfully maintain a real-time synchronization state, and at this time, the processing server may process the read request. It should be noted that, the data state in this embodiment may include a data logic state of a data storage unit such as a database and a data file system in a server, and the data state of the processing server and the data state of the main server being consistent may include: when the same data is input to the processing server and the main server, the same data processing result can be obtained. Of course, the data storage form in the processing server and the main server may not be identical, for example, the main server may be a database storage form, and the processing server may be a file system storage form, and if the data logic states of the two are consistent, it may be determined that the data states of the processing server and the main server are consistent.

In one embodiment of the present application, the determining by the processing server that the data state of the main server is consistent may include:

SS 1: when the processing server determines that the processing server is in a binding state with a main server, the data states of the processing server and the main server are consistent.

When the processing server and the main server are in a binding state, it may be determined that the data states of the processing server and the main server are consistent. During the binding of the processing server and the main server, the processing server may process a read request sent by a client and return a processing result. Specifically, this embodiment provides a method how to determine whether the processing server is in a binding state with the host server, fig. 2 is a flowchart of a method of an embodiment of determining whether the processing server is bound with the host server, and as shown in fig. 2, the determining whether the processing server is in a binding state with the host server may include:

s21: and acquiring the time when the processing server receives a reply message last time, wherein the reply message comprises a message that the processing server is successfully bound with the main server.

In the high-consistency communication system of this embodiment, the processing server may periodically send a synchronous binding request to the master server, and the master server may determine whether the processing server meets a synchronous binding requirement after receiving the synchronous binding request sent by the processing server, and when the processing server meets the synchronous binding requirement, the master server may send a reply message, where the reply message may include a message that the processing server and the master server are successfully bound. In this embodiment, the main server sends a reply message to the processing server when at least the following conditions are satisfied:

the data state of the processing server is consistent with the data state of the main server;

and the response speed of the processing server meets the preset binding requirement.

After receiving the synchronous binding request sent by the processing server, the main server may determine whether the data state of the processing server is consistent with the data state of the main server. As described above, the main server is configured to process a write request from a client, and after the write request is processed, the main server changes the data state, and a processing server can ensure the data state to be consistent with the main server only by timely synchronizing with the main server, so that the main server can allow the processing server to bind with the main server. In addition, the main server further needs to determine that the response speed of the processing server meets a preset binding requirement, and specifically, for example, the main server may test whether the current response speed of the processing server is greater than or equal to a preset response speed. Generally, the binding time of the main server and the processing server is short, and if the response speed of the processing server is slow, the comprehensive processing efficiency of the main server may be affected. For example, after the main server makes a trade-off between response delay of read request processing capability and response delay of write request processing capability, binding can be simultaneously implemented with N processing servers, and if the response speed of one or more processing servers is slow, the number of the main server binding processing servers is affected, and the processing efficiency of the main server is reduced.

The following describes the binding process between the main server and the processing server through a specific application scenario: the processing server A sends a synchronous binding request to the main server, the main server judges whether the processing server A meets the binding requirement after receiving the synchronous binding request, and if the synchronous binding requirement is met, a reply message of successful synchronous binding is returned. The main server returns a reply message of successful synchronous binding, which indicates that the main server is allowed to keep a synchronous binding state with the processing server A within a time period T, and the main server cannot send a processing result of the write request during synchronous binding with the processing server A, and the main server can return the processing result of the write request only when the main server and the processing server A finish binding or the binding is overtime. After receiving the reply message which is successfully synchronously bound with the main server, the processing server A can determine to keep binding with the main server within a time period T-T, wherein T represents the time consumed by the processing server A from sending the synchronous binding request to receiving the reply message, and within the time period T-T, the processing server A can receive the read request and return the processing result. Of course, the processing server a may also send a synchronization binding request again within the time period T-T, and if the synchronization binding is successful, the processing server a may also continue to maintain the binding with the main server. If the processing server A does not receive the reply message which is successfully bound with the main server synchronously within the time period T-T, after the time period T-T is finished, the processing server A loses the binding relationship with the main server, and can refuse to process a new read request.

S22: judging whether the time interval between the moment of receiving the reply message and a reference moment is smaller than a preset time interval, wherein the reference moment is the moment of executing the judging step; and if the time interval is smaller than the preset time interval, determining that the processing server and the main server are in a binding state.

In this embodiment, the binding time between the main server and the processing server is set to T, and the preset time interval may be T-T, where T represents the time consumed by the processing server to send the synchronous binding request until the reply message is received. As described above, in the time period T-T, the processing server is bound with the main server, and may process the read request and return a processing result. In this embodiment, when it is determined that the time interval between the time when the reply message is received and the reference time is smaller than the preset time interval, the reference time is the time when the determining step is executed, and it may be determined that the processing server and the main server are in the binding state, and at this time, the processing server may process the read request. Of course, subsequently, the processing server may also return the processing result to the client.

In this embodiment, when the processing server determines that the data state of the processing server is consistent with the data state of the main server, the processing server may process the read request. However, in the prior art, the client may only send a read request to the master server, and the master server may authorize one of the slave servers to process the read request sent by the client when determining that the slave server meets the requirement for processing the read request. Compared with the prior art, in the embodiment, the client can directly send the read request to the slave server for processing, and does not need to send the read request to the master server and then the read request is distributed to the slave server for processing by the master server. Therefore, the embodiment can not only reduce the processing load of the read request of the master server, but also utilize the idle time of the slave server, and meanwhile, the master server is not required to work in the key processing process of the read request, so that the processing delay is not increased, and the read request processing efficiency of the high-consistency communication system is improved. Further, in an embodiment of the present application, after the processing server determines that the data state of the processing server is consistent with the data state of the main server, the method may further include:

S2S: and marking the processing server as a priority processing server for processing the client read request so that the client preferentially selects the priority processing server in the server set.

After the processing server determines that the data state of the processing server is consistent with the data state of the main server, the selected processing server can be marked as a priority processing server for processing a client read request, so that the client preferentially selects the priority processing server in the server set. When the same client sends a read request next time or different clients send read requests, a priority processing server in the server set can be selected preferentially. Generally, the frequency of sending the read request by the client is high, and the time consumed for processing the read request is short compared with the time for binding the main server by the processing server, so that the same client sends the read request to the processing server marked as the priority processing server next time, the probability of successful processing is high, and the processing efficiency of the read request is improved.

In an embodiment of the present application, a method for processing a read request failure of a client is further provided, and fig. 3 is a flowchart of a method of an embodiment of the method for processing a read request failure provided by the present application, and as shown in fig. 3, the method may further include:

s31: and when the processing server determines that the data state of the processing server is inconsistent with the data state of the main server, the processing server refuses to process the read request of the client and returns an error request report.

When the processing server determines that the data state of the processing server is inconsistent with the data state of the main server, the processing server may refuse to process the read request of the client and return an error request report, such as a NAK error report.

S32: and the client receives the error request report and counts the total number of the error request reports.

S33: and if the total number is less than or equal to a preset threshold value, the client reselects a server from the server set to process the read request.

After receiving the error request report returned by the processing server, the client may count the total number of the error request reports. If the total number of the received error request reports obtained by statistics is smaller than or equal to a preset threshold value, the client can reselect a processing server from the server set to process the read request.

In another embodiment of the present application, when the client sends a read request to a plurality of processing servers, but none of the read requests is successfully processed, the client may further select to process the read request by the main server, as shown in fig. 3, and the method may further include:

s34: and if the total number is larger than the preset threshold value, the client sends the reading request to a main server.

In conjunction with S33 and S34, that is, when the second processing server cannot process the read request, the read request may be continuously sent to the third processing server until N processing servers cannot process the read request, where N is a preset threshold of the total number of processing servers to which the client sends the read request, and the client may send the read request to the main server.

According to the method for processing the read request failure of the client, the client can turn to the main server to send the read request under the condition that a plurality of processing servers cannot process the read request in time, and the read request is guaranteed to be completed smoothly.

In another embodiment of the present application, the consistency of the data states of the processing server and the main server may also be determined by another method. Specifically, in this embodiment, the read request sent by the client may include a data state update time of the main server that is obtained by the client last time.

In an embodiment of the present application, the data update time in the read request may be obtained by any one of the following manners:

a client sends a write request to a main server, and the main server responds to the write request and updates the data state sent to the client;

and the client synchronously requests the latest data state updating time of the main server from the main server.

As described above, the client may obtain the data state update time of the host server by any one of the above manners: a client may send a write request to a primary server that updates a data state of the primary server in response to the write request. When the main server determines that the write request processing is successful, the data state update time may be returned to the client side while the write request result is returned. Or before sending the read request to the processing server, the client may send a synchronization request to the main server to request to acquire the latest data state update time of the main server, and the main server sends the latest data state update time to the client in response to the synchronization request. Among the two methods, the first method is more suitable for an application scenario in which the time when the client sends the write request last time is shorter than the time when the client sends the read request, and the second method is suitable for an application scenario in which the time when the client receives the data state update last time is longer than the time when the client sends the read request.

Accordingly, the determining that the data state of the primary server is consistent may include:

SS 2: the processing server acquires the latest data state updating time of the main server and judges whether the latest data state updating time is equal to or later than the data state updating time in the reading request; (ii) a And if so, determining that the data state is consistent with the data state of the main server.

In the high-consistency communication system according to this embodiment, since the processing server is subordinate to the main server, when the main server updates the data state, the processing server tends to actively synchronize the data state of the main server. In this embodiment, the processing server may synchronize the data state of the main server and may also synchronize the time when the main server updates the data state. When it is determined that the data state update time obtained by the processing server last time is equal to or later than the data state update time in the read request, it may be determined that the data states of the processing server and the main server are consistent.

In this embodiment, the data state update time of the main server is used as the data update flag of the main server, whether the data states of the processing server and the main server are consistent or not can be judged based on the data state update time, and the judgment can be performed by using the sequence of the time, so that the judgment speed can be increased, and the processing efficiency of the read request can be improved.

Based on processing the read request processing procedure of the processing server and the main server, the application provides an embodiment of a data processing system, which may include:

In one embodiment of the present application, the system may further include:

a main server, configured to receive a binding request of the processing server, and upon determining that the processing server meets a binding requirement, send a reply message to the processing server, where the reply message includes a message that the binding of the processing server with the main server is successful;

correspondingly, the processing server is further configured to receive the reply message that is successfully bound, and after receiving the read request of the client, determine whether a time interval between the time when the reply message is received and a reference time is less than a preset time interval, where the reference time is the time when the determining step is executed; and if the time interval is smaller than the preset time interval, determining that the processing server and the main server are in a binding state.

In another embodiment of the present application, the read request includes a time when the client has last acquired the data state of the primary server; accordingly, the processing server is further configured to:

acquiring the latest data state updating time of a main server, and judging whether the latest data state updating time is equal to or later than the data state updating time in the reading request;

and if so, determining that the data state is consistent with the data state of the main server.

The foregoing method embodiment is described below by a specific application scenario, and fig. 4 is a schematic view of an application scenario of the data processing method provided in this application, as shown in fig. 4, in a high-consistency communication system a, a master server is elected from N +1 processing servers by a Paxos algorithm, and the remaining N processing servers are used as slave servers of the master server. As shown in fig. 4, the slave server 1 sends a synchronous binding request to the master server, and after receiving the synchronous binding request, the master server determines whether the slave server 1 meets a preset binding requirement, and if the synchronous binding requirement is met, returns a reply message of successful binding. The master server returns a reply message of successful binding indicating that the synchronous binding state with the slave server 1 is allowed to be maintained for a period of time T. The slave server 1, after receiving the reply message that the binding is successful, may determine to keep the binding with the master server for a time period T-T, where T represents the time consumed by the slave server 1 to send the synchronous binding request until receiving the reply message, and during the time period T-T, the slave server 1 may receive the read request and return the processing result. As shown in fig. 4, the client sends a read request to the slave server 1, and after receiving the read request, when the slave server 1 determines that the data state of the slave server is consistent with that of the master server, the slave server 1 processes the read request and sends a read request processing result to the client. When it is determined that the data status of the slave server is inconsistent with the data status of the master server, the slave server 1 may refuse to process the read request of the client and return an error request report, such as a NAK error report, to the client. After receiving the error request report returned by the slave server 1, the client may reselect the slave server from the server set to process the read request if the total number of the error request reports obtained by statistics is less than or equal to the preset threshold. When the second slave server cannot process the read request, the read request may be continuously sent to the third slave server until M slave servers cannot process the read request, where M is the preset threshold, and the client may send the read request to the master server.

According to the data processing method and system, the client side can directly send the read request to the slave server, and after the slave server receives the read request, when the data state of the slave server is consistent with that of the master server, the slave server can directly process the read request. Compared with the prior art, the client does not need to send a read request to the master server, and the read request is distributed to the slave servers by the master server for processing. The read request processing method provided by the application can reduce the read request processing burden of the master server, utilize the idle time of the slave server, and simultaneously, the master server is not required to work in the key processing process of the read request, so that the processing delay is not increased, and the read request processing efficiency of the high-consistency communication system is improved. In addition, the method of the embodiment can enable a plurality of slave servers in the high-consistency communication system to process the read requests simultaneously, the maximum processing capacity of the communication system can be expanded to the sum of the processing capacities of all the processing servers, and the read request processing capacity of the high-consistency communication system is expanded.

Based on the above data processing method, another embodiment of a read request processing method is proposed from the perspective of a processing server, fig. 5 is a flowchart of a method of another embodiment of the read request processing method provided in the present application, and as shown in fig. 5, the method may include:

s51: receiving a reading request of a client;

s52: and processing the read request when the data state is determined to be consistent with the data state of the main server.

The specific implementation manners of S51-S52 can refer to the implementation manners of S1-S2, and are not described herein again.

Optionally, in an embodiment of the present application, the read request may include a read request sent by a client to a processing server selected from a provided server set.

Optionally, in an embodiment of the present application, the determining that the data state of the primary server is consistent includes:

when the data state is determined to be in the binding state with the main server, the data state is determined to be consistent with the data state of the main server.

In an embodiment of the present application, there is further provided a method for determining, by the processing server, that the processing server is in a binding state with the main server, as shown in fig. 6, where the method may include:

s61: and acquiring the time when a reply message is received last time, wherein the reply message comprises the reply message which is successfully bound with the main server.

S62: and judging whether the time interval between the moment of receiving the reply message and a reference moment is less than a preset time interval, wherein the reference moment is the moment of executing the judging step.

S63: and if the time interval is smaller than the preset time interval, determining that the server is in a binding state with the main server.

In this embodiment, the reply message that the synchronous binding with the master server is successful may be received when the following conditions are satisfied:

the data state is consistent with the data state of the main server;

and the response speed meets the preset binding requirement.

In this embodiment, it is determined that the processing server and the main server are in the binding state, so that the data states of the processing server and the main server are consistent, and the binding protocol can ensure the consistency of the data states of the processing server and the main server.

In another embodiment of the present application, the read request includes a data state update time of the main server that is obtained by the client last time;

accordingly, the determining that the data state of the primary server is consistent comprises:

Based on the read request processing method proposed in the perspective of the processing server, the present application also proposes a server, fig. 7 is a schematic block structure diagram of an embodiment of the server provided in the present application, and as shown in fig. 7, the server 70 may include:

a receiving unit 71, configured to receive a read request of a client;

a processing unit 72 for processing the read request when it is determined that the data state of the main server is consistent.

In one embodiment of the present application, the read request may include a read request sent by a client to a processing server selected from a provided server set.

In an embodiment of the present application, the processing unit 72 may further be configured to:

Fig. 8 is a schematic block diagram of an embodiment of a processing unit provided in the present application, and as shown in fig. 7, the processing unit 72 may include:

a reply message obtaining unit 81, configured to obtain a time when a reply message is received last time, where the reply message includes a reply message that is successfully bound with the master server;

a time interval determination unit 82, configured to determine whether a time interval between the time when the reply message is received and a reference time is smaller than a preset time interval, where the reference time is the time when the determination step is executed;

and a binding relation determining unit 83, configured to determine that the server is in a binding state with the host server if the time interval is smaller than the preset time interval.

Optionally, in an embodiment of the present application, the reply message may be received when at least the following conditions are satisfied:

the data state is consistent with the data state of the main server;

and the response speed meets the preset binding requirement.

In this embodiment, whether the processing server and the main server are in the binding state is determined by judging whether the processing server and the main server are in the binding state, and the binding protocol can ensure the consistency of the data states of the processing server and the main server.

In an embodiment of the present application, the read request includes a data state update time of a main server that is obtained by a client last time;

accordingly, the processing unit 72 may be further configured to:

acquiring the latest data state updating time of a main server, and judging whether the latest data state updating time is equal to or later than the data state updating time in the reading request; and if so, determining that the data state is consistent with the data state of the main server.

In another embodiment of the present application, the data update time in the read request is obtained by any one of the following manners:

By the data processing method and the server, after the processing server receives the reading request, when the data state of the processing server is determined to be consistent with that of the main server, the processing server can directly process the reading request. The read request processing method provided by the application can reduce the read request processing burden of the main server, utilize the idle time of the processing server, and simultaneously, the main server is not required to work in the key processing process of the read request, so that the processing delay is not increased, and the read request processing efficiency of the high-consistency communication system is improved. In addition, the method of the embodiment can enable a plurality of processing servers in the high-consistency communication system to process the read request simultaneously, the maximum processing capacity of the communication system can be expanded to the sum of the processing capacities of all the processing servers, and the read request processing capacity of the high-consistency communication system is expanded.

Based on the above data processing method, another embodiment of the read request processing method is provided from the perspective of the client, fig. 9 is a method flowchart of another embodiment of the read request processing method provided in the present application, and as shown in fig. 9, the method may include:

s91: sending a read request to a processing server;

s92: and receiving a processing result of the processing server on the read request.

In an embodiment of the present application, the processing result may include:

when the data states of the processing server and the main server are consistent, processing the read request and then returning an output result;

or, when the data states of the processing server and the main server are inconsistent, rejecting to process the read request and returning an error request report.

Correspondingly, fig. 10 is a flowchart of a method of an embodiment of a processing method for processing a read request failure provided in the present application, and as shown in fig. 10, after receiving a processing result of the read request by the processing server, the method may further include:

s101: and counting the total number of the error request reports.

S102: and if the total number is less than or equal to a preset threshold value, reselecting a processing server from the server set to process the read request.

Further, as shown in fig. 10, in another embodiment of the present application, after the counting the total number of the error request reports, the method may further include:

s103: and if the total number is larger than the preset threshold value, sending the reading request to the main server.

Based on the read request processing method proposed from the perspective of the client, the present application also proposes a client, fig. 11 is a schematic diagram of a module structure of an embodiment of the client provided by the present application, and as shown in fig. 11, the client 110 may include:

a read request sending unit 111 for sending a read request to the processing server;

a processing result receiving unit 112, configured to receive a processing result of the read request by the processing server.

In another embodiment of the present application, the processing server may comprise a processing server selected from a provided set of servers.

In another embodiment of the present application, the processing result may include:

Correspondingly, in another embodiment of the present application, the client may further process the read request failure, fig. 12 is a schematic diagram of a module structure of another embodiment of the client provided in the present application, and as shown in fig. 12, the client 120 may further include:

a report counting unit 121, configured to count the total number of error request reports;

and the secondary processing unit 122 is configured to reselect a processing server from the server set to process the read request if the total number is less than or equal to a preset threshold.

Further, in an embodiment of the present application, as shown in fig. 12, the client 120 may further include:

a main server requesting unit 123, configured to send the read request to the main server if the total number is greater than the preset threshold.

By the data processing method and the client side, the client side can select the slave servers from the server set and send the read requests to the selected slave servers, and the client side can also receive the processing results of the slave servers on the read requests. Compared with the prior art, the client does not need to send a read request to the master server, and the read request is distributed to the slave servers by the master server for processing. The read request processing method provided by the application can reduce the read request processing burden of the master server, utilize the idle time of the slave server, and simultaneously, the master server is not required to work in the key processing process of the read request, so that the processing delay is not increased, and the read request processing efficiency of the high-consistency communication system is improved.

Although the present application refers to the data transmission, processing, and determination descriptions such as sending a read request, determining a binding condition, sending a synchronous binding request, and the like in the embodiments, the present application is not limited to the case where the data transmission and processing described in the embodiments are necessarily completely satisfied. The same, equivalent or similar, or variations of the embodiments described above may be accomplished using custom or certain communication standards/protocols or with slightly modified embodiments as described in the examples set forth herein. Of course, even if the data processing and determining manner is not adopted, the same application can still be realized as long as the information transmission and information determination feedback manner according to the above embodiments of the present application is met, and details are not described herein again.

Although the present application provides method steps as described in an embodiment or flowchart, more or fewer steps may be included based on conventional or non-inventive means. The order of steps recited in the embodiments is merely one manner of performing the steps in a multitude of orders and does not represent the only order of execution. When an actual apparatus or client product executes, it may execute 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 units and systems illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. For convenience of description, the above devices are described as being divided into various modules by functions, and are described separately. Of course, the functionality of the various modules may be implemented in the same one or more software and/or hardware implementations as the present application. Of course, a module that implements the same function may also be implemented by multiple sub-modules or a combination of sub-modules in a unit module described in this application.

Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may therefore be considered as a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.

The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, classes, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

From the above description of the embodiments, it is clear to those skilled in the art that the present application can be implemented by software plus necessary general hardware platform. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, or the like, and includes several instructions for enabling a computer device (which may be a personal computer, a mobile terminal, a server, or a network device) to execute the method according to the embodiments or some parts of the embodiments of the present application.

The embodiments in the present specification are described in a progressive manner, and the same or similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. The application is operational with numerous general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.

While the present application has been described with examples, those of ordinary skill in the art will appreciate that there are numerous variations and permutations of the present application without departing from the spirit of the application, and it is intended that the appended claims encompass such variations and permutations without departing from the spirit of the application.

Claims

1. A method of data processing, the method comprising:

a client sends a reading request to a processing server;

the processing server receives the reading request, and processes the reading request when the processing server determines that the data state of the processing server is consistent with that of the main server; wherein the processing server is subordinate to the main server, and the main server is selected from a system comprising N processing servers; the data state comprises a data logic state of a data storage unit in the server, and the determining by the processing server that the data state is consistent with the data state of the main server comprises: when the same data are input into the processing server and the main server, the same data processing result is obtained;

wherein, after the processing server receives the read request, the method further comprises: and when the data state of each processing server with a preset threshold value is determined to be inconsistent with the data state of the main server, the client sends the reading request to the main server.

2. The method of claim 1, wherein the processing server comprises a processing server selected by the client from a set of provided servers.

3. The method of claim 2, wherein after the processing server receives the read request, the method further comprises:

when the processing server determines that the data state of the processing server is inconsistent with the data state of the main server, the processing server refuses to process the read request of the client and returns an error request report;

the client receives the error request report and counts the total number of the error request reports;

and if the total number is less than or equal to the preset threshold value, the client reselects a processing server from the server set to process the read request.

4. The method according to claim 3, wherein when a preset threshold number of processing servers all determine that the data state of the processing servers is inconsistent with the data state of the main server, the sending, by the client, the read request to the main server includes:

and if the total number is larger than the preset threshold value, the client sends the reading request to the main server.

5. The method of claim 1, wherein the processing server determining that the data state of the primary server is consistent comprises:

when the processing server determines that the processing server is in a binding state with a main server, the data states of the processing server and the main server are consistent.

6. The method of claim 5, wherein determining that the master server is in a binding state comprises:

acquiring the moment when the processing server receives a reply message last time, wherein the reply message comprises a message that the processing server is successfully bound with the main server;

judging whether the time interval between the moment of receiving the reply message and a reference moment is smaller than a preset time interval, wherein the reference moment is the moment of executing the judging step;

and if the time interval is smaller than the preset time interval, determining that the processing server and the main server are in a binding state.

7. The method according to claim 6, wherein said primary server sends said reply message to said processing server when at least the following conditions are satisfied:

8. The method according to claim 1, wherein the read request includes a data state update time of the main server that is obtained by the client last time;

the processing server acquires the latest data state updating time of the main server and judges whether the latest data state updating time is equal to or later than the data state updating time in the reading request;

9. The method of claim 8, wherein the data state update time in the read request is obtained by any one of the following methods:

10. The method of claim 1, wherein after the processing server determines that the data state of the processing server is consistent with the data state of the primary server, the method further comprises:

and marking the processing server as a priority processing server for processing the read request of the client so that the client preferentially selects the priority processing server in the server set.

11. A method of data processing, the method comprising:

a processing server receives a reading request of a client;

when the processing server determines that the data state of the processing server is consistent with that of the main server, processing the read request; wherein the processing server is subordinate to the main server, and the main server is selected from a system comprising N processing servers; the data state comprises a data logic state of a data storage unit in the server, and the determining by the processing server that the data state is consistent with the data state of the main server comprises: when the same data are input into the processing server and the main server, the same data processing result is obtained; wherein, after the processing server receives the read request, the method further comprises: and when the data state of the processing server is determined to be inconsistent with that of the main server by the preset threshold value, turning to the main server to send the reading request.

12. The method of claim 11, wherein the read request comprises a read request sent by a client to a processing server selected from a set of provided servers.

13. The method of claim 11, wherein determining that the data state of the primary server is consistent comprises:

14. The method of claim 13, wherein determining that the master server is in a binding state comprises:

acquiring the moment of receiving a reply message last time, wherein the reply message comprises a reply message successfully bound with the main server;

and if the time interval is smaller than the preset time interval, determining that the server is in a binding state with the main server.

15. The method of claim 14, wherein the reply message is received when at least the following conditions are met:

the data state is consistent with the data state of the main server;

and the response speed meets the preset binding requirement.

16. The method according to claim 11, wherein the read request includes a data state update time of the primary server that is last acquired by the client;

17. The method of claim 16, wherein the data state update time in the read request is obtained by any one of the following methods:

18. A method of data processing, the method comprising:

the method comprises the steps that a client sends a reading request to a processing server, so that the processing server processes the reading request when determining that the data state of the processing server is consistent with that of a main server; wherein the processing server is subordinate to the main server, and the main server is selected from a system comprising N processing servers; the data state comprises a data logic state of a data storage unit in the server, and the determining by the processing server that the data state is consistent with the data state of the main server comprises: when the same data are input into the processing server and the main server, the same data processing result is obtained; wherein, still include: when a preset threshold value of processing servers determine that the data states of the processing servers are inconsistent with the data states of the main server, the client sends the reading request to the main server;

and the client receives the processing result of the processing server to the read request.

19. The method of claim 18, wherein the processing server comprises a processing server selected from a set of provided servers.

20. The method of claim 19, wherein the processing results comprise:

21. The method of claim 20, wherein after receiving a result of processing the read request by the processing server, the method further comprises:

counting the total number of the error request reports;

and if the total number is less than or equal to a preset threshold value, reselecting a processing server from the server set to process the read request.

22. The method of claim 21, wherein after said counting the total number of error request reports, the method further comprises:

and if the total number is larger than the preset threshold value, sending the reading request to the main server.

23. A server, characterized in that the server comprises:

the receiving unit is used for receiving a reading request of a client;

the processing unit is used for processing the reading request when the data state of the processing unit is consistent with that of a main server, wherein the server belongs to the main server, and the main server is selected from a system comprising N servers; the data state comprises a data logic state of a data storage unit in the server, and the determining that the data state is consistent with the data state of the main server comprises: when the same data are input into the server and the main server, the same data processing result is obtained; wherein, still include: and when the data state of the processing server is determined to be inconsistent with that of the main server by the preset threshold value, turning to the main server to send the reading request.

24. The server according to claim 23, wherein the read request comprises a read request sent by a client to a processing server selected from a set of provided servers.

25. The server according to claim 23, wherein the processing unit is further configured to:

26. The server according to claim 25, wherein the processing unit comprises:

a reply message obtaining unit, configured to obtain a time when a reply message is received last time, where the reply message includes a reply message that is successfully bound with the master server;

a time interval determination unit, configured to determine whether a time interval between the time when the reply message is received and a reference time is smaller than a preset time interval, where the reference time is the time when the determination step is executed;

and the binding relation determining unit is used for determining that the main server is in a binding state if the time interval is smaller than the preset time interval.

27. The server according to claim 26, wherein the reply message is received when at least the following conditions are met:

the data state is consistent with the data state of the main server;

and the response speed meets the preset binding requirement.

28. The server according to claim 23, wherein the read request includes a data state update time of the main server that is last acquired by the client;

accordingly, the processing unit is further configured to:

29. The server according to claim 28, wherein the data status update time in the read request is obtained by any one of the following methods:

30. A client, the client comprising:

a read request sending unit, configured to send a read request to a processing server in the system, so that when the processing server determines that the data state of the processing server is consistent with that of the main server, the read request is processed; wherein the processing server is subordinate to the main server, and the main server is selected from a system comprising N processing servers; the data state comprises a data logic state of a data storage unit in the server, and the determining by the processing server that the data state is consistent with the data state of the main server comprises: when the same data are input into the processing server and the main server, the same data processing result is obtained; wherein, still include: when a preset threshold value of processing servers determine that the data states of the processing servers are inconsistent with the data states of the main server, the client sends the reading request to the main server;

31. The client of claim 30, wherein the processing server comprises a processing server selected from a set of servers provided.

32. The client of claim 30, wherein the processing result comprises:

33. The client of claim 32, further comprising:

the report counting unit is used for counting the total number of the error request reports;

and the secondary processing unit is used for reselecting the processing server from the server set to process the read request if the total number is less than or equal to a preset threshold value.

34. The client of claim 33, further comprising:

and the main server request unit is used for sending the reading request to the main server if the total number is greater than the preset threshold value.

35. A data processing system, characterized in that the system comprises:

the processing server is used for receiving a reading request of the client; and, when it is determined that the data state of the primary server is consistent, processing the read request;

the system also comprises a main server, wherein the main server is selected from N processing servers; the data state comprises a data logic state of a data storage unit in the server, and the determining by the processing server that the data state is consistent with the data state of the main server comprises: when the same data are input into the processing server and the main server, the same data processing result is obtained; wherein, still include: and when the data state of the processing server is determined to be inconsistent with that of the main server by the preset threshold value, turning to the main server to send the reading request.

36. The system of claim 35, further comprising:

37. The system according to claim 36, wherein the read request includes a data state update time of the main server that the client has last acquired; correspondingly, the processing server is further configured to obtain a latest data state updating time of the main server, and determine whether the latest data state updating time is equal to or later than the data state updating time in the read request; and if so, determining that the data state is consistent with the data state of the main server.