WO2017101642A1

WO2017101642A1 - Method and apparatus for upgrading data node of distributed system

Info

Publication number: WO2017101642A1
Application number: PCT/CN2016/106804
Authority: WO
Inventors: 许瑞亮
Original assignee: 北京奇虎科技有限公司; 奇智软件（北京）有限公司
Priority date: 2015-12-17
Filing date: 2016-11-22
Publication date: 2017-06-22
Also published as: CN105610903B; CN105610903A

Abstract

Disclosed are a method and apparatus for upgrading a data node of a distributed system. In a distributed system, with regard to one piece of data, a plurality of data nodes are respectively used for backing up primary copy data and multiple secondary copy data of this piece of data. The method comprises: receiving an upgrade instruction, and selecting, from a plurality of data nodes, a first data node to be upgraded; transferring a service load of the first data node to at least one second data node, and the second data node backing up secondary copy data corresponding to primary copy data in the first data node; and performing offline processing on the first data node, so as to complete the upgrade of the first data node offline.

Description

Data node upgrade method and device for distributed system

Cross-reference to related applications

The present application claims priority to Chinese Patent Application No. 201510954639.8, filed on Jan. 17, 2015, the entire content of

Technical field

The present invention relates to the field of computer network technologies, and in particular, to a data node upgrade method and apparatus for a distributed system.

Background technique

Distributed storage is the decentralized storage of data on multiple independent devices. A distributed system is a cluster system with many data nodes, and each data node can be understood as a data storage server. When doing data storage, storing several pieces of data into corresponding data nodes is the core concept of a distributed system.

In the prior art, when the data node of the distributed system needs to be upgraded, the generally adopted method is to shut down the data node and stop the external service service of the data node, that is, the data node is offline, offline. The data node is upgraded. After the data node is upgraded, the data node is started, that is, the data node is brought online, so that the upgraded data node provides the external service service. However, the data node of the distributed system generally has a backup. Different master copy data, the data node needs to be restarted after the upgrade, and the master copy data of the data node is re-decisive. The decision process takes a long time, and the start and stop of the data node causes the service service delay or error of the data node. Etc., thus affecting the business services of distributed systems.

Summary of the invention

In view of the above problems, the present invention has been made in order to provide a data node upgrade method for a distributed system that overcomes the above problems or at least partially solves the above problems, and a data node upgrade device for a corresponding distributed system.

According to an aspect of the present invention, a data node upgrade method for a distributed system is provided. In a distributed system, for a piece of data, a plurality of data nodes are respectively used to back up the primary copy data of the data and multiple From copy data, methods include:

Receiving an upgrade instruction, selecting a first data node to be upgraded from a plurality of data nodes;

Transferring the traffic load of the first data node to the at least one second data node, the second data node backing up the slave replica data corresponding to the primary replica data in the first data node;

Performing offline processing on the first data node to complete the upgrade of the first data node offline.

According to another aspect of the present invention, a data node upgrading apparatus for a distributed system is provided. In a distributed system, for a piece of data, a plurality of data nodes are respectively used to back up the primary copy data of the data and From the copy data, the device includes:

a selection module, configured to receive an upgrade instruction, and select a first data node to be upgraded from the plurality of data nodes;

a transfer module, configured to transfer a service load of the first data node to the at least one second data node, where the second data node backs up the slave copy data corresponding to the primary copy data in the first data node;

The upgrade module is configured to perform offline processing on the first data node to complete the upgrade of the first data node offline.

According to still another aspect of the present invention, a computer program is provided comprising computer readable code that, when run on a computing device, causes the computing device to perform the data node upgrade method of the distributed system described above.

According to still another aspect of the present invention, a computer readable medium storing the above computer program is provided.

According to the solution provided by the present invention, the upgrade instruction is received, and the first data node to be upgraded is selected from the plurality of data nodes; the traffic load of the first data node is transferred to the at least one second data node, and the second data node backup has a corresponding The slave data of the primary replica data in the first data node; performing offline processing on the first data node to complete the upgrade of the first data node offline, thereby implementing a smooth upgrade of the distributed system and avoiding the data node The impact of shutting down and starting business services on distributed systems.

The above description is only an overview of the technical solution of the present invention, in order to understand the technology of the present invention more clearly. The above and other objects, features and advantages of the present invention will be apparent from the description of the appended claims.

BRIEF abstract

Various other advantages and benefits will become apparent to those skilled in the art from a The drawings are only for the purpose of illustrating the preferred embodiments and are not to be construed as limiting. Throughout the drawings, the same reference numerals are used to refer to the same parts. In the drawing:

1 is a flow chart showing a data node upgrade method of a distributed system according to an embodiment of the present invention;

2 is a schematic flow chart of a data node upgrade method of a distributed system according to another embodiment of the present invention;

3 is a functional block diagram of a data node upgrading apparatus of a distributed system in accordance with one embodiment of the present invention;

4 is a functional block diagram of a data node upgrading apparatus of a distributed system according to another embodiment of the present invention;

FIG. 5 is a block diagram schematically showing a computing device for performing a data node upgrade method of a distributed system according to an embodiment of the present invention;

Fig. 6 schematically shows a storage unit for holding or carrying program code implementing a data node upgrade method of a distributed system according to an embodiment of the present invention.

Preferred embodiment of the invention

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While the embodiments of the present invention have been shown in the drawings, the embodiments Rather, these embodiments are provided so that this disclosure will be more fully understood and the scope of the disclosure will be fully disclosed.

In a distributed system, for a piece of data, a plurality of data nodes are used to back up the primary copy data and the plurality of secondary copy data of the data, that is, the primary copy data of the data is more The copy data is stored separately on a plurality of different data nodes. For different data, both the primary replica data and the secondary replica data may be stored on each data node of the distributed system. In general, the service service is provided by the primary replica data of the data node.

FIG. 1 is a flow chart showing a data node upgrade method of a distributed system according to an embodiment of the present invention. As shown in Figure 1, the method includes the following steps:

Step S100: Receive an upgrade instruction, and select a first data node to be upgraded from the plurality of data nodes.

Specifically, after receiving the upgrade instruction, starting to upgrade the data node of the distributed system, in order to implement a smooth upgrade of the distributed system, selecting a data node to be upgraded from multiple data nodes of the distributed system, specifically, The data node to be upgraded may be randomly selected from a plurality of data nodes, or the data node to be upgraded may be selected according to the load condition of the plurality of data nodes, as the first data node to be upgraded, where the first to be upgraded is selected. A data node selection is not specifically limited.

Step S101: Transfer the traffic load of the first data node to the at least one second data node, and the second data node backs up the slave replica data corresponding to the primary replica data in the first data node.

In the embodiment of the present invention, the traffic load of the first data node is transferred to the at least one second data node, where the service access service is provided by the at least one second data node instead of the first data node, and the primary data node is The business access of the replica data is transferred to the slave replica data of at least one second data node.

In a distributed system, the first data node may store the primary replica data of different data. Therefore, when performing the traffic load transfer, the service access provided by all the primary replica data stored by the first data node needs to be transferred to the primary. Copy data from the copy data.

For one piece of data, there will be multiple slave replica data, and multiple slave replica data are stored in different data nodes respectively. Therefore, it is necessary to find the slave replica data backed up with all the master replica data corresponding to the first data node. A data node from which to select at least one second data node that is responsible for the traffic load of the first data node. For example, for one piece of data, three data nodes can be used to back up the primary copy data and the two secondary copy data of the data. If the first data node backs up the primary copy data of three copies of the data, the three data are The backup data is backed up at least on two data nodes. When performing business load transfer, it is generally transferred to a slave copy data, so it needs to be from at least two. Among the data nodes, a data node that assumes the traffic load of the first data node is selected as at least one second data node.

Here, the traffic load of the first data node is transferred to the at least one second data node, so that the corresponding service service of the first data node can be continuously provided without affecting the service service of the distributed system, thereby avoiding the data node activation. Stopping caused the business service to time out.

Step S102: Perform offline processing on the first data node to complete the upgrade of the first data node offline.

After transferring the traffic load of the first data node to the at least one second data node by using step S101, the first data node only backs up the slave replica data with the data. As mentioned above, the service service is generally provided by the master replica data. That is to say, the first data node is disconnected from the service of the distributed system, and the first data node can be offlined, the first data node is taken offline, and the first data node is upgraded offline.

According to the method provided by the foregoing embodiment of the present invention, receiving an upgrade instruction, selecting a first data node to be upgraded from a plurality of data nodes, and transferring a service load of the first data node to at least one second data node, the second data node The backup has the slave copy data corresponding to the master copy data in the first data node; the first data node is processed offline to complete the upgrade of the first data node offline, thereby implementing a smooth upgrade of the distributed system and avoiding The impact of data node shutdown and startup on the business services of the distributed system.

2 is a flow chart showing a data node upgrade method of a distributed system according to another embodiment of the present invention. As shown in FIG. 2, the method includes the following steps:

Step S200: Receive an upgrade instruction, analyze a service load of a data node of the distributed system, and select a data node with a minimum service load pressure as the first data node to be upgraded.

In the embodiment of the present invention, the traffic load refers to the amount of access to the primary replica data of the data node. In a distributed system, the service services provided by each data node are not the same. Some data nodes have high traffic load pressure, and some data nodes have low traffic load pressure. After receiving the upgrade command, the data of the distributed system is received. The service load of the node is analyzed, and the data node with the lowest traffic load pressure is selected as the first data node to be upgraded. The present invention is to upgrade all the data nodes of the distributed system. Therefore, the first data node to be upgraded is the data node with the least traffic load pressure among the un-upgraded data nodes, and the data node with the lowest traffic load pressure can be avoided. Data Node upgrades, transferring business services to other data nodes and affecting distributed systems.

Step S201: Search, according to the primary replica data identifier of the first data node, a data node that backs up the secondary replica data corresponding to the primary replica data in the first data node in the data node of the distributed system.

In a distributed system, a piece of data is used by multiple data nodes to back up the primary copy data and the plurality of secondary copy data of the data, and the same identification is set for the primary copy data and the copy data, so as to facilitate According to the identifier, the primary copy data or the secondary data is searched, and different identifiers are set for different data, so that the primary copy data of the data or the copy data can be accurately found, which saves the search time and improves the upgrade efficiency.

After the first data node to be upgraded is selected according to step S200, the service service of the primary data of the first data node needs to be transferred to the secondary data of the other data node. Of course, the primary copy of the first data node is not The business service of the data is transferred to the slave copy data of any data node, but to the data node backed up with the slave copy data corresponding to the master copy data in the first data node. In addition, the first data node may back up the data with different data. The primary replica data, if the first data node is to be upgraded, needs to transfer the business services of all the primary replica data in the first data node to the secondary replica data of the other data nodes. Therefore, it is necessary to look up, in the data node of the distributed system, the data node backed up with the replica data corresponding to all the primary replica data in the first data node according to the primary replica data identifier of the first data node.

Step S202, analyzing a service load of the data node from the replica data corresponding to the primary replica data in the first data node, and selecting at least one second data node according to the load balancing policy.

The data node that backs up the slave copy data corresponding to the master copy data in the first data node is found by using step S201, however, the traffic load of the first data node is not transferred to all the found data nodes, for example, for one The data has one data node backed up with the master copy data and two data node backups with the slave copy data. When the business load of the data of the first data node is transferred, only the backup data can be transferred to the backup. One of the data nodes.

In the embodiment of the present invention, by analyzing the service load of the data node that backs up the replica data corresponding to the primary replica data in the first data node, the traffic load pressure of the data node is analyzed, and the traffic load policy is selected according to the load balancing policy. Out at least one second data node, wherein the load The equalization strategy refers to the load balancing of the traffic load of the data nodes of the distributed system, without the pressure of the traffic load of some data nodes is large, and the traffic load pressure of some data nodes is very small, here not only consider that the backup has corresponding to The traffic load of the data node of the replica data from the primary replica data in the first data node also takes into account the traffic load of the data node after the traffic load is transferred. The at least one second data node selected in this step is a data node with a small traffic load pressure. Here, selecting at least one second data node according to the load balancing policy can overcome the defect that the traffic load of the data node with high traffic load pressure is caused by the arbitrary transfer of the traffic load.

Since the first data node may back up multiple primary replica data, there may be multiple data nodes selected according to the load balancing policy. Here, the traffic load of the first data node is transferred to the at least one second data node, so that the corresponding service service of the first data node can be continuously provided without affecting the service service of the distributed system, thereby avoiding the data node activation. Stopping caused the business service to time out.

Step S203, converting the copy data of the at least one second data node into the main copy data, and converting the corresponding main copy data of the first data node into the slave copy data.

After selecting at least one second data node according to step S202, the traffic load of the first data node needs to be transferred to the at least one second data node. In the distributed system, the service load is generally provided by the primary replica data, and at least one The second data node is a data node that backs up the replica data corresponding to the primary replica data of the first data node, and therefore, the replica data of the at least one second data node needs to be converted into the primary replica data, and the first data node is The corresponding primary copy data is converted into secondary replica data, thereby transferring the traffic load to the at least one second data node.

Step S204: Perform offline processing on the first data node to complete the upgrade of the first data node offline.

After the service load of the first data node is transferred to the at least one second data node by using step S203, the first data node will no longer provide the service service. At this time, the first data node can be offline, and the first line is offline. The data node is upgraded.

Although it is mentioned above that the service service is generally provided by the primary replica data of the data node, there may also be cases where the secondary data of the data node is accessed. Therefore, the method further includes:

Step S205, relocating the access request from the replica data in the first data node to the primary replica data/from the replica data corresponding to the replica data in the other data nodes.

Specifically, after the user initiates an access request from the replica data in the first data node, the access request from the replica data in the first data node is relocated to the primary replica data corresponding to the replica data in the other data node. The copy data is used to provide a service service from the primary data corresponding to the copy data from the other data nodes after the positioning, wherein the first data node may be the first data node before the copy data may be the traffic load transfer. The backup copy data can also be obtained from the copy data after the business load is transferred.

Step S206, after the first data node completes the upgrade, performing online processing on the upgraded first data node.

After the upgrade of the first data node is completed online, the first data node needs to be brought online, so that the first data node continues to provide service services.

After the upgraded first data node is online, the service load of the at least one third data node needs to be transferred to the upgraded first data node, so that the upgraded first data node is restored to provide service services, where after the upgrade The first data node backup has slave data corresponding to the primary copy data in the third data node.

In the embodiment of the present invention, the traffic load of the at least one third data node is transferred to the first data node, where the first data node replaces the at least one third data node to provide a service service, and the at least one third data node is the master. The business service of the replica data is transferred to the slave replica data of the first data node. This can be achieved by the following method steps:

Step S207: Searching, in the data node of the distributed system, the data node backed up with the primary replica data corresponding to the replica data in the first data node after the upgrade according to the upgraded data identifier of the first data node.

After the first data node is brought online, the traffic load of other data nodes needs to be transferred to the first data node, and the first data node can provide the service service.

When the data is backed up on the data node, the same identification is set for the primary replica data and the replica data. Therefore, the backup data may be searched for in the data node of the distributed system according to the secondary data identifier of the first data node. A data node of a data node corresponding to the primary copy data corresponding to the replica data.

Step S208, analyzing a service load of the data node corresponding to the primary replica data corresponding to the replica data in the upgraded first data node, and selecting from the load balancing policy according to the load balancing policy One less third data node.

After searching for the data nodes corresponding to the primary replica data of the replica data in the upgraded first data node according to step S207, analyzing the traffic load of the data nodes, determining the traffic load of the data nodes, according to the load. The equalization strategy selects a data node with a large traffic load as the third data node, thereby reducing the traffic load of the third data node, and balancing the traffic load of the distributed system.

Step S209: Convert the primary copy data of the at least one third data node into the secondary copy data, and convert the corresponding copy of the upgraded first data node from the copy data to the primary copy data.

After selecting at least one third data node according to step S208, the traffic load of the at least one third data node needs to be transferred to the first data node. In the distributed system, the service load is generally provided by the primary replica data, and at least one The third data node is a data node backed up with the primary copy data corresponding to the copy data in the first data node, and therefore, the primary copy data of the at least one third data node needs to be converted into the secondary copy data, and the upgraded The correspondence of a data node is converted from the replica data to the primary replica data, thereby transferring the traffic load to the first data node, so that the upgraded data node is restored to provide the service service.

According to the method provided by the foregoing embodiment of the present invention, after analyzing the service load of the data node of the distributed system, the data node with the lowest traffic load pressure is selected as the first data node to be upgraded, and the distributed system can be avoided. The service service is affected. According to the primary copy data identifier of the first data node, the data node in the distributed system is searched for the data node that has the secondary copy data corresponding to the primary copy data in the first data node, so that the data can be accurately found. The primary copy data or the secondary copy data saves the search time and improves the upgrade efficiency, and analyzes the service load of the data node from the duplicate data corresponding to the primary copy data in the first data node, according to the load balancing policy. Selecting at least one second data node can overcome the defect of arbitrarily transferring the traffic load of the data node with heavy traffic load pressure caused by the traffic load, and can continue to provide the corresponding service service of the first data node without being distributed. The system's business services have an impact, thus avoiding The data node starts and stops to cause the service service to time out. After the first data node completes the upgrade, the first data node is processed online. According to the secondary data identifier of the first data node, the backup of the data node of the distributed system has a corresponding correspondence. The data node of the primary replica data of the replica data is analyzed in the first data node, and the service load corresponding to the data node corresponding to the primary replica data of the replica data in the first data node is analyzed, and the traffic load is selected according to the load balancing policy. At least one A third data node, thereby reducing the traffic load of the third data node, and achieving a smooth upgrade of the distributed system.

3 shows a functional block diagram of a data node upgrade device of a distributed system in accordance with one embodiment of the present invention. As shown in FIG. 3, the device includes: a selection module 300, a transfer module 310, and an upgrade module 320.

The selecting module 300 is adapted to receive an upgrade instruction, and select a first data node to be upgraded from the plurality of data nodes.

The transferring module 310 is adapted to transfer the traffic load of the first data node to the at least one second data node, and the second data node backs up the slave replica data corresponding to the primary replica data in the first data node.

The upgrade module 320 is adapted to perform offline processing on the first data node to complete the upgrade of the first data node offline.

The device according to the above embodiment of the present invention receives an upgrade instruction, selects a first data node to be upgraded from a plurality of data nodes, and transfers a service load of the first data node to at least one second data node, the second data node The backup has the slave copy data corresponding to the master copy data in the first data node; the first data node is processed offline to complete the upgrade of the first data node offline, thereby implementing a smooth upgrade of the distributed system and avoiding The impact of data node shutdown and startup on the business services of the distributed system.

4 is a functional block diagram of a data node upgrading apparatus of a distributed system in accordance with another embodiment of the present invention. As shown in FIG. 4, the device includes: a selection module 400, a transfer module 410, and an upgrade module 420.

The selecting module 400 is adapted to receive an upgrade instruction, and select a first data node to be upgraded from the plurality of data nodes.

The transfer module 410 is adapted to transfer the traffic load of the first data node to the at least one second data node, and the second data node backs up the slave replica data corresponding to the primary replica data in the first data node.

The upgrade module 420 is adapted to perform offline processing on the first data node to complete the upgrade of the first data node offline.

Optionally, the device further includes: a processing module 430, configured to: after the first data node completes the upgrade, Performing online processing on the upgraded first data node;

The transfer module 410 is further adapted to: transfer the traffic load of the at least one third data node to the upgraded first data node, so that the upgraded first data node resumes providing the service service, wherein the upgraded first data node The backup has secondary copy data corresponding to the primary copy data in the third data node.

Optionally, the transferring module 410 is further adapted to: convert the copied data of the at least one second data node into the primary copy data, and convert the corresponding primary copy data of the first data node into the secondary copy data.

Optionally, the device further includes: a searching module 440 and an analyzing module 450.

The searching module 440 is adapted to search, in the data node of the distributed system, the data node that is backed up with the secondary copy data corresponding to the primary copy data in the first data node according to the primary copy data identifier of the first data node.

The analyzing module 450 is adapted to analyze a service load of the data node from the replica data corresponding to the primary replica data in the first data node, and select at least one second data node from the load balancing policy.

Optionally, the transferring module 410 is further adapted to: convert the primary copy data of the at least one third data node into the secondary copy data, and convert the corresponding copy of the upgraded first data node from the copy data to the primary copy data.

Optionally, the searching module 440 is further adapted to: according to the upgraded data identifier of the upgraded first data node, searching for a backup in the data node of the distributed system and corresponding to the copy data from the first data node after the upgrade Data node of the primary replica data;

The analyzing module 450 is further configured to: analyze a service load that is backed up with a data node corresponding to the primary replica data of the replica data in the upgraded first data node, and select at least one third data node from the load balancing policy according to the load balancing policy .

Optionally, the selecting module 400 is further adapted to: analyze the traffic load of the data node of the distributed system, and select the data node with the smallest traffic load as the first data node to be upgraded.

Optionally, the apparatus further includes: a positioning module 460, configured to relocate the access request from the replica data in the first data node to the primary replica data/from the replica data corresponding to the replica data in the other data nodes.

According to the foregoing embodiment of the present invention, after analyzing the service load of the data node of the distributed system, the data node with the lowest traffic load pressure is selected as the first data node to be upgraded, and the distributed system can be avoided. The service service is affected. According to the primary copy data identifier of the first data node, the data node in the distributed system is searched for the data node that has the secondary copy data corresponding to the primary copy data in the first data node, so that the data can be accurately found. The primary copy data or the secondary copy data saves the search time and improves the upgrade efficiency, and analyzes the service load of the data node from the duplicate data corresponding to the primary copy data in the first data node, according to the load balancing policy. Selecting at least one second data node can overcome the defect of arbitrarily transferring the traffic load of the data node with heavy traffic load pressure caused by the traffic load, and can continue to provide the corresponding service service of the first data node without being distributed. The system's business services have an impact, thus avoiding The data node starts and stops to cause the service service to time out. After the first data node completes the upgrade, the first data node is processed online. According to the secondary data identifier of the first data node, the backup of the data node of the distributed system has a corresponding correspondence. The data node of the primary replica data of the replica data is analyzed in the first data node, and the service load corresponding to the data node corresponding to the primary replica data of the replica data in the first data node is analyzed, and the traffic load is selected according to the load balancing policy. At least one third data node is eliminated, thereby reducing the traffic load of the third data node, and implementing a smooth upgrade of the distributed system.

The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general purpose systems can also be used with the teaching based on the teachings herein. The structure required to construct such a system is apparent from the above description. Moreover, the invention is not directed to any particular programming language. It is to be understood that the invention may be embodied in a variety of programming language, and the description of the specific language has been described above in order to disclose the preferred embodiments of the invention.

In the description provided herein, numerous specific details are set forth. However, it is understood that the embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of the description.

Similarly, the various features of the invention are sometimes grouped together into a single embodiment, in the above description of the exemplary embodiments of the invention, Figure, or a description of it. However, the method disclosed is not to be interpreted as reflecting the intention that the claimed invention requires more features than those recited in the claims. Rather, as reflected in the claims below As such, the inventive aspects reside in less than all features of the single embodiments disclosed above. Therefore, the claims following the specific embodiments are hereby explicitly incorporated into the embodiments, and each of the claims as a separate embodiment of the invention.

Those skilled in the art will appreciate that the modules in the devices of the embodiments can be adaptively changed and placed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and further they may be divided into a plurality of sub-modules or sub-units or sub-components. In addition to such features and/or at least some of the processes or units being mutually exclusive, any combination of the features disclosed in the specification, including the accompanying claims, the abstract and the drawings, and any methods so disclosed, or All processes or units of the device are combined. Each feature disclosed in this specification (including the accompanying claims, the abstract and the drawings) may be replaced by alternative features that provide the same, equivalent or similar purpose.

In addition, those skilled in the art will appreciate that, although some embodiments described herein include certain features that are included in other embodiments and not in other features, combinations of features of different embodiments are intended to be within the scope of the present invention. Different embodiments are formed and formed. For example, in the following claims, any one of the claimed embodiments can be used in any combination.

The various component embodiments of the present invention may be implemented in hardware, or in a software module running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that some or all of some or all of the data node upgrade devices of the distributed system in accordance with embodiments of the present invention may be implemented in practice using a microprocessor or digital signal processor (DSP). Features. The invention can also be implemented as a device or device program (e.g., a computer program and a computer program product) for performing some or all of the methods described herein. Such a program implementing the invention may be stored on a computer readable medium or may be in the form of one or more signals. Such signals may be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

For example, Figure 5 illustrates a computing device that can implement a method of acquiring application information in accordance with the present invention. The computing device conventionally includes a processor 510 and a computer program product or computer readable medium in the form of a storage device 520. Storage device 520 can be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read Only Memory), EPROM, hard disk, or ROM. Save Storage device 520 has a storage space 530 that stores program code 531 for performing any of the method steps described above. For example, storage space 530 storing program code may include various program code 531 for implementing various steps in the above methods, respectively. The program code can be read from or written to one or more computer program products. These computer program products include program code carriers such as a hard disk, a compact disk (CD), a memory card, or a floppy disk. Such a computer program product is typically a portable or fixed storage unit such as that shown in FIG. The storage unit may have storage segments, storage spaces, and the like that are similarly arranged to storage device 520 in the computing device of FIG. The program code can be compressed, for example, in an appropriate form. Typically, the storage unit comprises computer readable code 531 ' for performing the steps of the method according to the invention, ie code that can be read by a processor such as 510, which when executed by the computing device causes the computing device Perform the various steps in the method described above.

It is to be noted that the above-described embodiments are illustrative of the invention and are not intended to be limiting, and that the invention may be devised without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as a limitation. The word "comprising" does not exclude the presence of the elements or steps that are not recited in the claims. The word "a" or "an" The invention can be implemented by means of hardware comprising several distinct elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means can be embodied by the same hardware item. The use of the words first, second, and third does not indicate any order. These words can be interpreted as names.

Claims

A data node upgrade method for a distributed system, in which, for a piece of data, a plurality of data nodes are respectively used for backing up primary copy data and a plurality of secondary copy data of the data, the method include:

Receiving an upgrade instruction, selecting a first data node to be upgraded from a plurality of data nodes;

Transferring the traffic load of the first data node to at least one second data node, the second data node backing up with the slave copy data corresponding to the primary replica data in the first data node;

Performing offline processing on the first data node to complete upgrade of the first data node offline.
The method of claim 1 wherein the method further comprises:

After the first data node completes the upgrade, the upgraded first data node is processed online, and the service load of the at least one third data node is transferred to the upgraded first data node, so that the upgrade is performed. The first data node is restored to provide a service service, wherein the upgraded first data node backs up the slave copy data corresponding to the primary copy data in the third data node.
The method of claim 2, wherein the transferring the traffic load of the first data node to the at least one second data node further comprises:

Converting the copy data of the at least one second data node into primary copy data, and converting the corresponding primary copy data of the first data node into the secondary copy data.
The method of claim 2 or 3, wherein before the transferring the traffic load of the first data node to the at least one second data node, the method further comprises:

Determining, in the data node of the distributed system, a data node backed up with the secondary replica data corresponding to the primary replica data in the first data node according to the primary replica data identifier of the first data node;

And analyzing, by the backup, a service load of the data node from the replica data corresponding to the primary replica data in the first data node, and selecting at least one second data node from the backup according to the load balancing policy.
The method of any of claims 2-4, wherein said at least one third number Transferring the traffic load of the node to the upgraded first data node further includes:

Converting the primary replica data of the at least one third data node into the secondary replica data, and converting the corresponding correspondence of the upgraded first data node from the replica data to the primary replica data.
The method of claim 4, wherein before the transferring the traffic load of the at least one third data node to the upgraded first data node, the method further comprises:

And searching, in the data node of the distributed system, a data node that is backed up with the primary replica data corresponding to the replica data in the first data node after the upgrade according to the upgraded data identifier of the first data node;

And analyzing, for the backup, a traffic load corresponding to the data node of the primary data of the replica data in the upgraded first data node, and selecting at least one third data node according to the load balancing policy.
The method of any one of claims 1-6, wherein the selecting the first data node to be upgraded from the plurality of data nodes further comprises:

The service load of the data node of the distributed system is analyzed, and the data node with the smallest traffic load is selected as the first data node to be upgraded.
The method according to any one of claims 1 to 7, wherein before the upgraded first data node is brought online, the method further comprises:

The access request from the replica data in the first data node is relocated to the primary copy data/slave copy data corresponding to the secondary copy data in the other data nodes.
A data node upgrading apparatus for a distributed system, in the distributed system, for a piece of data, used by a plurality of data nodes to back up primary copy data and a plurality of secondary copy data of the data, the device include:

a selection module, configured to receive an upgrade instruction, and select a first data node to be upgraded from the plurality of data nodes;

a transfer module, configured to transfer a traffic load of the first data node to at least one second data node, where the second data node is backed up with secondary copy data corresponding to primary copy data in the first data node;

An upgrade module, configured to perform offline processing on the first data node, to complete the online The upgrade of the first data node.
The apparatus of claim 9 wherein said apparatus further comprises:

a processing module, configured to perform online processing on the upgraded first data node after the first data node completes the upgrade;

The transfer module is further adapted to: transfer the service load of the at least one third data node to the upgraded first data node, so that the upgraded first data node resumes providing service services, wherein, after the upgrade The first data node is backed up with slave copy data corresponding to the primary copy data in the third data node.
The apparatus of claim 10, wherein the transfer module is further adapted to: convert the copy data of the at least one second data node into primary copy data, the corresponding primary copy data of the first data node Convert to data from the copy.
The device according to claim 10 or 11, wherein the device further comprises:

a searching module, configured to search, in the data node of the distributed system, a data node that is backed up with the secondary copy data corresponding to the primary copy data in the first data node according to the primary copy data identifier of the first data node;

The analyzing module is configured to analyze, for the backup, a service load of the data node from the replica data corresponding to the primary replica data in the first data node, and select at least one second data node according to the load balancing policy.
The apparatus according to any one of claims 10 to 12, wherein the transfer module is further adapted to: convert main copy data of the at least one third data node into slave copy data, the upgraded The correspondence of a data node is converted from the copy data to the main copy data.
The apparatus according to claim 12, wherein said searching module is further adapted to: look up the backup in the data node of the distributed system according to the upgraded copy data identification of the first data node, corresponding to the post-upgrade a data node of the primary data of the copy data from the first data node;

The analyzing module is further configured to: analyze, by the backup, a service load corresponding to the data node of the primary data of the replica data in the upgraded first data node, and select at least the load according to the load balancing policy A third data node.
The apparatus of any of claims 9-14, wherein the selection module is further adapted to:

The service load of the data node of the distributed system is analyzed, and the data node with the smallest traffic load is selected as the first data node to be upgraded.
The apparatus of any of claims 9-15, wherein the apparatus further comprises: a positioning module adapted to relocate access requests from the copy data to the other data nodes in the first data node The primary copy data/from the copy data corresponding to the copy data.
A computer program comprising computer readable code, when said computer readable code is run on a computing device, causing said computing device to perform data of a distributed system according to any of claims 1-8 Node upgrade method.
A computer readable medium storing the computer program of claim 17.