CN105824571A - Data seamless migration method and device - Google Patents
Data seamless migration method and device Download PDFInfo
- Publication number
- CN105824571A CN105824571A CN201510003665.2A CN201510003665A CN105824571A CN 105824571 A CN105824571 A CN 105824571A CN 201510003665 A CN201510003665 A CN 201510003665A CN 105824571 A CN105824571 A CN 105824571A
- Authority
- CN
- China
- Prior art keywords
- node
- master node
- memory
- shared memory
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013508 migration Methods 0.000 title claims abstract description 56
- 230000005012 migration Effects 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000001514 detection method Methods 0.000 claims description 28
- 238000012423 maintenance Methods 0.000 claims description 20
- 238000012545 processing Methods 0.000 claims description 15
- 238000012546 transfer Methods 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 230000001960 triggered effect Effects 0.000 description 6
- 238000005192 partition Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000000737 periodic effect Effects 0.000 description 4
- 230000001360 synchronised effect Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 238000010998 test method Methods 0.000 description 2
- 206010033799 Paralysis Diseases 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
Landscapes
- Memory System Of A Hierarchy Structure (AREA)
Abstract
本发明公开了一种实现数据无缝迁移的方法，所述方法包括：分别创建主节点和备用节点的共享内存，以及主节点和备用节点各自的内存区域，将主节点内存区域内的内存信息迁移至主节点的共享内存，使主节点的内存信息与备用节点的内存信息一致；检测主节点发生故障时，访问共享内存中的数据，实现数据无缝迁移。本发明还同时公开了一种实现数据无缝迁移的装置。
The invention discloses a method for realizing seamless data migration. The method includes: respectively creating shared memory of a master node and a backup node, and respective memory areas of the master node and the backup node, and transferring the memory information in the memory area of the master node to Migrate to the shared memory of the master node, so that the memory information of the master node is consistent with the memory information of the standby node; when the failure of the master node is detected, the data in the shared memory is accessed to realize seamless data migration. The invention also discloses a device for realizing seamless migration of data.
Description
技术领域technical field
本发明涉及数据传输技术，尤其涉及一种实现数据无缝迁移的方法及装置。The invention relates to data transmission technology, in particular to a method and device for realizing seamless data migration.
背景技术Background technique
双机热备是一种解决服务器的临时故障的备份技术，双机热备技术使用两台服务器，互相备份，共同执行同一服务；当一台服务器出现故障时，可以由另一台服务器承担服务任务，从而在不需要人工干预的情况下，自动保证系统能持续提供服务，避免长时间的服务中断，保证系统长期、可靠的服务。Dual-system hot backup is a backup technology to solve the temporary failure of the server. The dual-system hot backup technology uses two servers to back up each other and perform the same service together; when one server fails, the other server can take over the service Tasks, so that without manual intervention, the system can automatically ensure that the system can continue to provide services, avoid long-term service interruptions, and ensure long-term and reliable services of the system.
随着企业信息化程度越来越高，企业对信息化的依赖越来越高，特别在电信、银行等行业，甚至提出了全年业务不能中断的要求；为避免主机硬件和系统故障对业务应用造成影响，众多企业都采用小型机UNIX系统+HA的硬件架构模式；在这种模式下，首先可以利用小型机硬件高冗余的特性，单个硬件故障并不会影响主机节点的正常运行；其次当备用主机节点监控到主机出现故障时，会自动重新启动业务至备用节点上，从而保证了业务的连续性。With the increasing degree of enterprise informatization, enterprises rely more and more on informatization, especially in telecommunications, banking and other industries, and even put forward the requirement that the business cannot be interrupted throughout the year; in order to avoid host hardware and system failures on business Many enterprises have adopted the hardware architecture model of minicomputer UNIX system + HA; in this mode, first of all, the high redundancy feature of minicomputer hardware can be used, and a single hardware failure will not affect the normal operation of the host node; Secondly, when the standby host node monitors that the host fails, it will automatically restart the business to the standby node, thereby ensuring the continuity of the business.
现有技术中，UNIX系统下数据的切换的方法是在主机/备机之间部署镜像磁盘，即：A主节点将LunA磁盘上的业务数据传输至镜像磁盘分区上，同时镜像至B备用节点上的LunB磁盘上，利用镜像分区工具在主备机上将目标磁盘上开辟镜像磁盘同步分区，用以存储中间业务数据；再在主备机上安装相关应用软件，将中间业务数据部署到镜像同步分区上，主备机之间再通过内部以太网连接传输同步数据；该方案与传统方案相比，使A主节点上的业务数据传输至B备用节点上的速度更快，保证了数据传输的一致性；In the prior art, the method of data switching under the UNIX system is to deploy mirror disks between the main machine and the standby machine, that is: the A master node transmits the business data on the LunA disk to the mirror disk partition, and mirrors it to the B standby node at the same time On the LunB disk on the main machine, use the mirror partition tool to create a mirror disk synchronization partition on the target disk on the master and backup machines to store intermediate business data; then install relevant application software on the master and slave machines to deploy the intermediate business data to the mirror synchronization partition On top of that, the master and standby machines transmit synchronous data through the internal Ethernet connection; compared with the traditional scheme, this scheme enables the business data on the master node A to be transmitted to the standby node B at a faster speed, ensuring the consistency of data transmission sex;
但是，该方案存在如下弊端：第一，当A主节点宕机时，B备用节点读取镜像磁盘分区上的业务数据，还要重新启动宕机时传输过来的内存程序，导致从A主节点到B备用节点的业务切换时间要三至十五分钟，业务切换时间仍旧很长；第二，该方案需要在主机上增加额外的硬件和软件，过程复杂，成本高，故障点多，若一个点出现故障，整个系统瘫痪，甚至数据丢失；第三，该方案传输数据的实时性不强，切换时间长，很容易造成数据的丢失；第四，主备机的同步数据是通过内部以太网进行传输，传输速率慢，当以太网出现故障时，会造成数据传输的停止，中断了主备机之间的切换，致使系统不能正常运行。However, this solution has the following disadvantages: First, when the primary node A goes down, the standby node B reads the business data on the mirrored disk partition, and restarts the memory program transmitted during the downtime, resulting in The service switching time to the standby node B takes three to fifteen minutes, and the service switching time is still very long; second, this solution needs to add additional hardware and software on the host computer, the process is complicated, the cost is high, and there are many failure points. If a point failure, the entire system is paralyzed, and even data is lost; third, the real-time performance of data transmission in this scheme is not strong, and the switching time is long, which is easy to cause data loss; fourth, the synchronization data of the master and backup machines is through the internal Ethernet For transmission, the transmission rate is slow. When the Ethernet fails, it will cause the stop of data transmission, interrupt the switch between the main machine and the standby machine, and cause the system to fail to operate normally.
发明内容Contents of the invention
有鉴于此，本发明实施例期望提供一种实现数据无缝迁移的方法及装置，能够在主节点异常宕机时，备用节点在极短的时间内接管主节点的业务，实现主节点和备用节点之间的无扰切换和数据的无缝迁移，保证业务的连续性。In view of this, the embodiments of the present invention expect to provide a method and device for realizing seamless data migration. When the master node is abnormally down, the backup node can take over the business of the master node in a very short time, and realize the master node and the backup. Undisturbed switching between nodes and seamless migration of data ensures business continuity.
本发明实施例的技术方案是这样实现的：The technical scheme of the embodiment of the present invention is realized like this:
本发明实施例提供一种实现数据无缝迁移的方法，所述方法包括：分别创建主节点和备用节点的共享内存，以及主节点和备用节点各自的内存区域，将主节点内存区域内的内存信息迁移至主节点的共享内存，使主节点的内存信息与备用节点的内存信息一致；检测主节点发生故障时，访问共享内存中的数据，实现数据无缝迁移；其中，An embodiment of the present invention provides a method for realizing seamless data migration. The method includes: respectively creating the shared memory of the master node and the backup node, and the respective memory areas of the master node and the backup node, and converting the memory in the memory area of the master node to The information is migrated to the shared memory of the master node, so that the memory information of the master node is consistent with the memory information of the backup node; when the failure of the master node is detected, the data in the shared memory is accessed to realize seamless data migration; among them,
使主节点的内存信息与备用节点的内存信息一致，包括：主节点的快速缓冲贮存区读取主节点的内存信息，并将所述主节点的内存信息同步至所述主节点的共享内存；将主节点共享内存的数据同步至备用节点共享内存；备用节点的快速缓冲贮存区读取所述备用节点的共享内存的信息，并将所述共享内存的信息同步至备用节点的内存区域。Make the memory information of the master node consistent with the memory information of the standby node, comprising: reading the memory information of the master node by the fast buffer storage area of the master node, and synchronizing the memory information of the master node to the shared memory of the master node; Synchronize the data of the shared memory of the master node to the shared memory of the standby node; the fast buffer storage area of the standby node reads the information of the shared memory of the standby node, and synchronizes the information of the shared memory to the memory area of the standby node.
在一实施例中，所述将所述主节点的内存信息同步至所述主节点的共享内存，包括：实时地复制所述主节点的内存信息，并同步传递至所述主节点的共享内存；对所述主节点的内存信息进行页面快照，将变化的内存页面发送至所述主节点的共享内存，更新所述主节点的共享内存中所述主节点发生变化的内存页面对应的内存信息。In an embodiment, the synchronizing the memory information of the master node to the shared memory of the master node includes: copying the memory information of the master node in real time, and synchronously transferring the memory information to the shared memory of the master node ; Take a page snapshot of the memory information of the master node, send the changed memory page to the shared memory of the master node, and update the memory information corresponding to the changed memory page of the master node in the shared memory of the master node .
在一实施例中，所述访问共享内存中的数据，包括：主节点的快速缓冲贮存区加载所述主节点的内存区域中未发生变化的数据，备用节点的快速缓冲贮存区加载所述备用节点共享内存的数据，并重新挂载时控。In one embodiment, the accessing the data in the shared memory includes: loading the unchanged data in the memory area of the master node into the cache area of the master node, and loading the data in the memory area of the master node into the cache area of the backup node. Nodes share memory data and remount time-controlled.
在一实施例中，所述主节点与备用节点通过心跳线连接；相应的，所述检测主节点发生故障，包括：动态检测主节点的运行状态，并周期性的优先从所述心跳线向所述备用节点发送所述主节点的状态维护报，所述备用节点在预设的阈值时间内未收到所述主节点的状态维护报，则检测主节点发生故障。In one embodiment, the master node is connected to the backup node through a heartbeat line; correspondingly, the detection of a failure of the master node includes: dynamically detecting the running status of the master node, and periodically preferentially sending the heartbeat line to The backup node sends the status maintenance report of the master node, and if the backup node does not receive the status maintenance report of the master node within a preset threshold time, it detects that the master node fails.
本发明实施例还提供一种实现数据无缝迁移的装置，所述装置包括：创建模块、处理模块、检测模块和迁移模块；其中，The embodiment of the present invention also provides a device for seamless data migration, the device includes: a creation module, a processing module, a detection module and a migration module; wherein,
所述创建模块，用于分别创建主节点和备用节点主节点和备用节点的共享内存，以及主节点和备用节点各自的内存区域，将主节点内存区域内的内存信息迁移至主节点的共享内存；The creation module is used to create the shared memory of the master node and the backup node respectively, and the respective memory areas of the master node and the backup node, and migrate the memory information in the memory area of the master node to the shared memory of the master node ;
所述处理模块，用于读取主节点的内存信息，并将所述主节点的内存信息同步所述主节点的共享内存；将主节点共享内存的数据同步至备用节点共享内存；读取所述备用节点的共享内存的信息，并将所述共享内存的信息同步至备用节点的内存区域；The processing module is used to read the memory information of the master node, and synchronize the memory information of the master node with the shared memory of the master node; synchronize the data of the master node's shared memory to the standby node's shared memory; read the Describe the information of the shared memory of the standby node, and synchronize the information of the shared memory to the memory area of the standby node;
所述检测模块，用于检测主节点是否发生故障；The detection module is used to detect whether the master node fails;
所述迁移模块，用于在检测模块检测主节点发生故障时，访问共享内存中的数据，实现数据无缝迁移。The migration module is used to access the data in the shared memory when the detection module detects that the master node fails, so as to realize seamless data migration.
在一实施例中，所述处理模块，具体用于实时地复制所述主节点的内存信息，并同步传递至所述主节点的共享内存；对所述主节点的内存信息进行页面快照，将变化的内存页面发送至所述主节点的共享内存，更新所述主节点的共享内存中所述主节点发生变化的内存页面对应的内存信息。In one embodiment, the processing module is specifically configured to copy the memory information of the master node in real time, and transfer it to the shared memory of the master node synchronously; take a page snapshot of the memory information of the master node, and The changed memory page is sent to the shared memory of the master node, and the memory information corresponding to the changed memory page of the master node in the shared memory of the master node is updated.
在一实施例中，所述迁移模块，具体用于加载所述主节点的内存区域中未发生变化的数据，备用节点的快速缓冲贮存区加载所述备用节点共享内存的数据，并重新挂载时控。In an embodiment, the migration module is specifically configured to load the unchanged data in the memory area of the master node, and the fast buffer storage area of the standby node loads the data in the shared memory of the standby node, and remounts time control.
在一实施例中，所述主节点与备用节点通过心跳线连接；相应的，所述检测模块，具体用于动态检测主节点的运行状态，并周期性的优先从所述心跳线向所述备用节点发送所述主节点的状态维护报，所述备用节点在预设的阈值时间内未收到所述主节点的状态维护报，则检测主节点发生故障。In one embodiment, the master node is connected to the standby node through a heartbeat line; correspondingly, the detection module is specifically configured to dynamically detect the running status of the master node, and periodically preferentially connect the heartbeat line to the The backup node sends the status maintenance report of the master node, and if the backup node does not receive the status maintenance report of the master node within a preset threshold time, it detects that the master node is faulty.
本发明实施例所提供的实现数据无缝迁移的方法及装置，分别创建主节点和备用节点的共享内存，以及主节点和备用节点各自的内存区域，将主节点内存区域内的内存信息迁移至主节点的共享内存，使主节点的内存信息与备用节点的内存信息一致；检测主节点发生故障时，访问共享内存中的数据，实现数据无缝迁移；其中，使主节点的内存信息与备用节点的内存信息一致，包括：主节点的快速缓冲贮存区读取主节点的内存信息，并将所述主节点的内存信息同步至所述主节点的共享内存；将主节点共享内存的数据同步至备用节点共享内存；备用节点的快速缓冲贮存区读取所述备用节点的共享内存的信息，并将所述共享内存的信息同步至备用节点的内存区域。如此，通过将UNIX操作系统从本地系统磁盘迁移至共享存储，主节点和备用节点通过共享存储实现主节点的内存信息与备用节点的内存信息一致；在主节点异常宕机时，备用节点在极短的时间内接管主节点的业务，实现主节点和备用节点之间的无扰切换和数据的无缝迁移，保证业务的连续性。The method and device for realizing seamless data migration provided by the embodiments of the present invention respectively create the shared memory of the master node and the backup node, and the respective memory areas of the master node and the backup node, and migrate the memory information in the memory area of the master node to The shared memory of the master node makes the memory information of the master node consistent with the memory information of the backup node; when the failure of the master node is detected, the data in the shared memory is accessed to realize seamless data migration; among them, the memory information of the master node is consistent with that of the backup node The memory information of the nodes is consistent, including: the fast buffer storage area of the master node reads the memory information of the master node, and synchronizes the memory information of the master node to the shared memory of the master node; synchronizes the data of the shared memory of the master node to the standby node shared memory; the fast buffer storage area of the standby node reads the information of the shared memory of the standby node, and synchronizes the information of the shared memory to the memory area of the standby node. In this way, by migrating the UNIX operating system from the local system disk to the shared storage, the master node and the backup node realize the consistency of the memory information of the master node with the memory information of the backup node through the shared storage; Take over the business of the master node in a short period of time, realize the undisturbed switching and seamless migration of data between the master node and the backup node, and ensure the continuity of the business.
附图说明Description of drawings
图1为实现数据迁移的双机架构示意图；Figure 1 is a schematic diagram of a dual-machine architecture for data migration;
图2为数据迁移的实现过程示意图；Figure 2 is a schematic diagram of the implementation process of data migration;
图3为本发明实施例实现数据无缝迁移的方法的处理流程示意图；3 is a schematic diagram of a processing flow of a method for realizing seamless data migration according to an embodiment of the present invention;
图4为实现本发明实施例所应用的双机架构拓扑图；FIG. 4 is a topological diagram of a dual-machine architecture applied in an embodiment of the present invention;
图5为本发明实施例实现数据无缝迁移的装置的组成结构示意图。FIG. 5 is a schematic diagram of the composition and structure of a device for realizing seamless data migration according to an embodiment of the present invention.
具体实施方式detailed description
本发明实施例中，分别创建主节点和备用节点的共享内存，以及主节点和备用节点各自的内存区域，将主节点内存区域内的内存信息迁移至主节点的共享内存，使主节点的内存信息与备用节点的内存信息一致；检测主节点发生故障时，访问共享内存中的数据，实现数据无缝迁移；其中，使主节点的内存信息与备用节点的内存信息一致，包括：主节点的快速缓冲贮存区读取主节点的内存信息，并将所述主节点的内存信息同步至所述主节点的共享内存；将主节点共享内存的数据同步至备用节点共享内存；备用节点的快速缓冲贮存区读取所述备用节点的共享内存的信息，并将所述共享内存的信息同步至备用节点的内存区域。In the embodiment of the present invention, the shared memory of the master node and the backup node, and the respective memory areas of the master node and the backup node are respectively created, and the memory information in the memory area of the master node is migrated to the shared memory of the master node, so that the memory of the master node The information is consistent with the memory information of the standby node; when the failure of the primary node is detected, the data in the shared memory is accessed to realize seamless data migration; among them, the memory information of the primary node is consistent with the memory information of the standby node, including: the primary node The fast buffer storage area reads the memory information of the primary node, and synchronizes the memory information of the primary node to the shared memory of the primary node; synchronizes the data of the shared memory of the primary node to the shared memory of the standby node; the fast buffer of the standby node The storage area reads the information of the shared memory of the standby node, and synchronizes the information of the shared memory to the memory area of the standby node.
本发明实施例中，在创建主节点和备用节点的共享内存之后，主节点通过NPIV协议将存储在本地系统磁盘上的UNIX操作系统迁移至共享内存；由于一台主机可以划分为多个虚拟UNIX系统，因此本申请能够提高主机中央处理器内存资源利率。In the embodiment of the present invention, after creating the shared memory of the master node and the standby node, the master node migrates the UNIX operating system stored on the local system disk to the shared memory through the NPIV protocol; since a host can be divided into multiple virtual UNIX system, so the application can improve the memory resource rate of the host CPU.
为更好地理解本发明技术方案，下面简单介绍现有技术中基于磁盘阵列的方式实现数据迁移的实现过程；现有技术中实现数据迁移的双机架构，如图1所示，包括：A主节点、B备用节点和存储装置，双机热备系统的用户数据存储在磁盘阵列中，UNIX系统通过SAN网络访问磁盘阵列的资源。For a better understanding of the technical solution of the present invention, the following briefly introduces the implementation process of data migration based on disk arrays in the prior art; the dual-machine architecture for realizing data migration in the prior art, as shown in Figure 1, includes: A The master node, B backup node and storage device, the user data of the dual-machine hot standby system are stored in the disk array, and the UNIX system accesses the resources of the disk array through the SAN network.
现有技术中数据迁移的实现过程，如图2所示，当A主节点发生故障时，业务中断，B备用节点从存储装置上接管存储资源和IP资源，并恢复相关的应用；但是，B备用节点监控到A主节点发生故障后，在B备用节点上重启业务，根据承载的业务不同，进行业务恢复一般需要三至十分钟，甚至更长时间；在进行业务恢复的这段时间内，业务无法提供对外服务；因此，B备用节点的启用和A主节点的恢复无法实现无扰切换，使得业务不具备连续性。The implementation process of data migration in the prior art, as shown in Figure 2, when the master node A fails, the service is interrupted, and the backup node B takes over storage resources and IP resources from the storage device, and restores related applications; however, B After the standby node monitors that the primary node A fails, it restarts the business on the standby node B. Depending on the business carried, it generally takes three to ten minutes or even longer to recover the business; during the period of business recovery, The business cannot provide external services; therefore, the activation of the B standby node and the recovery of the A master node cannot achieve non-disruptive switching, making the business non-continuous.
本发明实施例中，实现数据无缝迁移的方法的处理流程，如图3所示，包括以下步骤：In the embodiment of the present invention, the processing flow of the method for realizing seamless data migration, as shown in FIG. 3 , includes the following steps:
步骤101，分别创建主节点和备用节点的共享内存，以及主节点和备用节点各自的内存区域，将主节点内存区域内的内存信息迁移至主节点的共享内存，使主节点的内存信息与备用节点的内存信息一致；Step 101, respectively create the shared memory of the master node and the backup node, and the respective memory areas of the master node and the backup node, and migrate the memory information in the memory area of the master node to the shared memory of the master node, so that the memory information of the master node is consistent with that of the backup node. The memory information of the node is consistent;
具体地，实现数据无缝迁移的装置中的创建模块创建主节点和备用节点的共享内存；主节点通过NPIV协议将存储在本地系统磁盘上的UNIX操作系统迁移至共享内存；实现数据无缝迁移的装置中的处理模块主节点的快速缓冲贮存区读取主节点的内存信息，并将所述主节点的内存信息同步至所述主节点的共享内存；将主节点共享内存的数据同步至备用节点共享内存；备用节点的快速缓冲贮存区读取所述备用节点的共享内存的信息，并将所述共享内存的信息同步至备用节点的内存区域，使主节点的内存信息与备用节点的内存信息一致；Specifically, the creation module in the device for seamless data migration creates the shared memory of the master node and the standby node; the master node migrates the UNIX operating system stored on the local system disk to the shared memory through the NPIV protocol; realizes seamless data migration The fast buffer storage area of the processing module master node in the device reads the memory information of the master node, and the memory information of the master node is synchronized to the shared memory of the master node; the data of the master node shared memory is synchronized to the standby Node shared memory; the fast buffer storage area of the standby node reads the information of the shared memory of the standby node, and synchronizes the information of the shared memory to the memory area of the standby node, so that the memory information of the master node is consistent with the memory information of the standby node consistent information;
其中，所述将所述主节点的内存信息同步至所述主节点的共享内存，具体为：Wherein, the synchronizing the memory information of the master node to the shared memory of the master node is specifically:
将共享内存分为两部分，一部分为页面标识区，用于标识变化的页面，每一位对应一个大小为4KB的数据页面；另一部分为数据区，用于存储应用的关键接续数据；所述关键接续数据为在发生主节点与备用节点切换时，保证业务不中断、接续计算机系统运行所需要的数据；在主节点的内存信息发生变化时，将变化的内存页面发送至共享内存，更新共享内存中所述主节点发生变化的内存页面对应的内存信息；之后，将更新的内存信息对应的页面标识区置为1，表示该数据区进行了更新。The shared memory is divided into two parts, one part is the page identification area, which is used to identify the changed page, and each bit corresponds to a data page with a size of 4KB; the other part is the data area, which is used to store the key connection data of the application; The key connection data is the data needed to ensure uninterrupted business and continue the operation of the computer system when the master node and the standby node switch; when the memory information of the master node changes, the changed memory page is sent to the shared memory and the shared memory is updated. The memory information corresponding to the memory page where the master node has changed in the memory; after that, the page identification area corresponding to the updated memory information is set to 1, indicating that the data area has been updated.
步骤102，检测主节点发生故障时，访问共享内存中的数据，实现数据无缝迁移；Step 102, when detecting failure of the master node, accessing the data in the shared memory to realize seamless data migration;
这里，主节点和备用节点建立基于C/S的系统架构，即实现本发明实施例所应用的双机架构拓扑图，如图4所示，主节点与备用节点分别至少配置两个网口，其中，一个网口连接公网，另一个网口配置为主节点与备用节点之间的私有网络，即心跳线；通常为公网配置两个网口，采用双网主备冗余的方式进行绑定，可以增加系统的可靠性；私有网络端口的数量可以根据数据量情况增加一个私有网络端口，采用负载均衡的方式绑定，用于增加私有网络的带宽和可靠性；Here, the master node and the backup node establish a C/S-based system architecture, that is, implement the topology diagram of the dual-machine architecture applied in the embodiment of the present invention. As shown in FIG. 4, the master node and the backup node are respectively configured with at least two network ports. Among them, one network port is connected to the public network, and the other network port is configured as a private network between the master node and the backup node, that is, the heartbeat line; usually two network ports are configured for the public network, and dual-network master-standby redundancy is adopted. Binding can increase the reliability of the system; the number of private network ports can be increased by one private network port according to the amount of data, and bound in a load balancing manner to increase the bandwidth and reliability of the private network;
具体地，采用定时检测和动态检测相结合的方式进行双机状态的检测；其中，定时检测是指主节点和备用节点周期性的发送自身的状态维护报，优先从心跳线发送状态维护报，在心跳线发送失败时，从公网发送状态维护报；在主节点或备用节点中的任意一方在预设的阈值时间内未收到对方的状态维护报，则检测对方发生故障；采用定时检测的方式进行状态检测，不仅具有ping测试方式的优点，还能在一定程度上冗余软件故障。动态检测是指主动探测主节点系统或备用节点系统软件的运行状态，如软件超时、软件死循环、软件一场退出等；当检测到软件故障时，停止发送状态维护报，进行状态告警并通知对方进行切换，再重启故障软件。Specifically, a combination of timing detection and dynamic detection is used to detect the dual-machine state; wherein, timing detection means that the master node and the backup node periodically send their own status maintenance reports, and the status maintenance reports are sent from the heartbeat line first. When the heartbeat line fails to be sent, a status maintenance report is sent from the public network; if either of the master node or the backup node does not receive the other party’s status maintenance report within the preset threshold time, it will detect that the other party has failed; use timing detection The method of state detection not only has the advantage of ping test method, but also can redundant software faults to a certain extent. Dynamic detection refers to the active detection of the operating status of the master node system or backup node system software, such as software timeout, software infinite loop, software exit, etc.; when a software failure is detected, stop sending status maintenance reports, issue status alarms and notify The other party switches, and then restarts the faulty software.
访问共享内存中的数据具体包括：在备用节点升级为主节点后，实现数据无缝迁移的装置中的迁移模块通知各应用从备用节点读取数据，立即加载共享区的关键接续数据，主节点的快速缓冲贮存区加载所述主节点的内存区域中未发生变化的数据，备用节点的快速缓冲贮存区加载所述备用节点共享内存的数据；再根据所述关键接续数据重新挂载应用的时控，重新挂载时控包括：1)对于周期性循环的时控，直接挂载即可；2)对于由某类信息触发或人工干预触发的单次或多次循环的时控，将触发条件或干预命令等信息作为关键数据存储于共享内存，备用节点切换为主节点后，根据所述触发条件或干预命令等信息重新解算并挂起时控；3)对于由周期性的消息触发的单次时控，可由新的消息再次触发，无需做特殊的加载处理。Accessing data in the shared memory specifically includes: after the backup node is upgraded to the master node, the migration module in the device for realizing seamless data migration notifies each application to read data from the backup node, and immediately loads the key connection data in the shared area, and the master node The fast buffer storage area of the primary node loads the data that has not changed in the memory area of the primary node, and the fast buffer storage area of the standby node loads the data of the shared memory of the standby node; and when the application is remounted according to the key connection data Time control, remounting time control includes: 1) For periodic cycle time control, just mount it directly; 2) For single or multiple cycle time control triggered by certain types of information or manual intervention, trigger Conditions or intervention commands and other information are stored in the shared memory as key data. After the standby node switches to the master node, it will recalculate and suspend time control according to the trigger conditions or intervention commands and other information; 3) For periodic messages triggered The single time control can be triggered again by new messages without special loading process.
本发明应用与SAN网络环境时，主节点与备用节点进行数据迁移时，可直接访问共享内存的数据，提高数据迁移的速率。When the present invention is applied to the SAN network environment, when data migration is performed between the master node and the standby node, the data in the shared memory can be directly accessed, and the speed of data migration is improved.
为实现上述实现数据无缝迁移的方法，本发明实施例还提供一种实现数据无缝迁移的装置，所述装置的组成结构，如图5所示，包括：创建模块11、处理模块12、检测模块13和迁移模块14；其中，In order to realize the above-mentioned method for realizing seamless data migration, an embodiment of the present invention also provides a device for realizing seamless data migration. The composition structure of the device, as shown in FIG. 5 , includes: a creation module 11, a processing module 12, Detection module 13 and migration module 14; Wherein,
所述创建模块11，用于分别创建主节点和备用节点主节点和备用节点的共享内存，以及主节点和备用节点各自的内存区域，将主节点内存区域内的内存信息迁移至主节点的共享内存；The creation module 11 is used to create the shared memory of the master node and the backup node respectively, and the respective memory areas of the master node and the backup node, and migrate the memory information in the memory area of the master node to the shared memory of the master node. Memory;
所述处理模块12，用于读取主节点的内存信息，并将所述主节点的内存信息同步至所述主节点的共享内存；将主节点共享内存的数据同步至备用节点共享内存；读取所述备用节点的共享内存的信息，并将所述共享内存的信息同步至备用节点的内存区域；The processing module 12 is configured to read the memory information of the master node, and synchronize the memory information of the master node to the shared memory of the master node; synchronize the data of the master node's shared memory to the backup node's shared memory; read Obtaining the information of the shared memory of the standby node, and synchronizing the information of the shared memory to the memory area of the standby node;
所述检测模块13，用于检测主节点是否发生故障；The detection module 13 is used to detect whether the master node fails;
所述迁移模块14，用于在检测模块检测主节点发生故障时，访问共享内存中的数据，实现数据无缝迁移。The migration module 14 is configured to access data in the shared memory when the detection module detects that the master node fails, so as to realize seamless data migration.
本发明实施例中，所述处理模块12，具体用于实时地复制所述主节点的内存信息，并同步传递至所述主节点的共享内存；对所述主节点的内存信息进行页面快照，将变化的内存页面发送至所述主节点的共享内存，更新所述主节点的共享内存中所述主节点发生变化的内存页面对应的内存信息。In the embodiment of the present invention, the processing module 12 is specifically configured to copy the memory information of the master node in real time, and transfer it to the shared memory of the master node synchronously; take a page snapshot of the memory information of the master node, Send the changed memory page to the shared memory of the master node, and update the memory information corresponding to the changed memory page of the master node in the shared memory of the master node.
本发明实施例中，所述迁移模块14，具体用于加载所述主节点的内存区域中未发生变化的数据，备用节点的快速缓冲贮存区加载所述备用节点共享内存的数据，并重新挂载时控。In the embodiment of the present invention, the migration module 14 is specifically used to load the data that has not changed in the memory area of the master node, and the fast buffer storage area of the backup node loads the data in the shared memory of the backup node, and remounts load time control.
本发明实施例中，所述主节点与备用节点通过心跳线连接；相应的，所述检测模块13，具体用于动态检测主节点的运行状态，并周期性的优先从所述心跳线向所述备用节点发送所述主节点的状态维护报，所述备用节点在预设的阈值时间内未收到所述主节点的状态维护报，则检测主节点发生故障。In the embodiment of the present invention, the master node is connected to the backup node through a heartbeat line; correspondingly, the detection module 13 is specifically used to dynamically detect the running status of the master node, and periodically preferentially transfer from the heartbeat line to the The backup node sends the status maintenance report of the master node, and the backup node detects that the master node fails if it does not receive the status maintenance report of the master node within a preset threshold time.
本发明实施例中，在创建主节点和备用节点的共享内存之后，主节点通过NPIV协议将存储在本地系统磁盘上的UNIX操作系统迁移至共享内存。In the embodiment of the present invention, after creating the shared memory of the master node and the standby node, the master node migrates the UNIX operating system stored on the local system disk to the shared memory through the NPIV protocol.
本发明实施例中，将所述主节点的内存信息同步至共享内存，具体为：In the embodiment of the present invention, the memory information of the master node is synchronized to the shared memory, specifically:
将共享内存分为两部分，一部分为页面标识区，用于标识变化的页面，每一位对应一个大小为4KB的数据页面；另一部分为数据区，用于存储应用的关键接续数据；所述关键接续数据为在发生主节点与备用节点切换时，保证业务不中断、接续计算机系统运行所需要的数据；在主节点的内存信息发生变化时，将变化的内存页面发送至共享内存，更新共享内存中所述主节点发生变化的内存页面对应的内存信息；之后，将更新的内存信息对应的页面标识区置为1，表示该数据区进行了更新。The shared memory is divided into two parts, one part is the page identification area, which is used to identify the changed page, and each bit corresponds to a data page with a size of 4KB; the other part is the data area, which is used to store the key connection data of the application; The key connection data is the data needed to ensure uninterrupted business and continue the operation of the computer system when the master node and the standby node switch; when the memory information of the master node changes, the changed memory page is sent to the shared memory and the shared memory is updated. The memory information corresponding to the memory page where the master node has changed in the memory; after that, the page identification area corresponding to the updated memory information is set to 1, indicating that the data area has been updated.
本发明实施例中，采用定时检测和动态检测相结合的方式进行双机状态的检测；其中，定时检测是指主节点和备用节点周期性的发送自身的状态维护报，优先从心跳线发送状态维护报，在心跳线发送失败时，从公网发送状态维护报；在主节点或备用节点中的任意一方在预设的阈值时间内未收到对方的状态维护报，则检测对方发生故障；采用定时检测的方式进行状态检测，不仅具有ping测试方式的优点，还能在一定程度上冗余软件故障。动态检测是指主动探测主节点系统或备用节点系统软件的运行状态，如软件超时、软件死循环、软件一场退出等；当检测到软件故障时，停止发送状态维护报，进行状态告警并通知对方进行切换，再重启故障软件。In the embodiment of the present invention, the combination of timing detection and dynamic detection is used to detect the dual-machine state; wherein, timing detection means that the master node and the backup node periodically send their own status maintenance reports, and the status is sent from the heartbeat line first. Maintenance report, when the heartbeat line fails to send, a status maintenance report is sent from the public network; if either of the master node or the backup node does not receive the other party's status maintenance report within the preset threshold time, it will detect that the other party has failed; The state detection by means of timing detection not only has the advantages of the ping test method, but also has redundant software faults to a certain extent. Dynamic detection refers to the active detection of the operating status of the master node system or backup node system software, such as software timeout, software infinite loop, software exit, etc.; when a software failure is detected, stop sending status maintenance reports, issue status alarms and notify The other party switches, and then restarts the faulty software.
本发明实施例中，访问共享内存中的数据具体包括：在备用节点升级为主节点后，实现数据无缝迁移的装置中的迁移模块通知各应用从备用节点读取数据，立即加载共享区的关键接续数据，主节点的快速缓冲贮存区加载所述主节点的内存区域中未发生变化的数据，备用节点的快速缓冲贮存区加载所述备用节点共享内存的数据；再根据所述关键接续数据重新挂载应用的时控，重新挂载时控包括：1)对于周期性循环的时控，直接挂载即可；2)对于由某类信息触发或人工干预触发的单次或多次循环的时控，将触发条件或干预命令等信息作为关键数据存储于共享内存，备用节点切换为主节点后，根据所述触发条件或干预命令等信息重新解算并挂起时控；3)对于由周期性的消息触发的单次时控，可由新的消息再次触发，无需做特殊的加载处理。In the embodiment of the present invention, accessing the data in the shared memory specifically includes: after the backup node is upgraded to the master node, the migration module in the device for realizing seamless data migration notifies each application to read data from the backup node, and immediately loads the data in the shared area Key connection data, the fast buffer storage area of the main node loads the data that has not changed in the memory area of the main node, and the fast buffer storage area of the backup node loads the data of the shared memory of the backup node; then according to the key connection data The time control of remounting the application includes: 1) For the time control of the periodic cycle, just mount it directly; 2) For the single or multiple cycles triggered by certain types of information or manual intervention time control, store information such as trigger conditions or intervention commands as key data in the shared memory, and after the standby node switches to the master node, recalculate and suspend time control according to the trigger conditions or intervention commands; 3) for A single time control triggered by a periodic message can be triggered again by a new message without special loading processing.
需要说明的是，在实际应用中，所述创建模块11、处理模块12、检测模块13和迁移模块14的功能可由位于实现数据无缝迁移的装置上的中央处理器(CPU)、或微处理器(MPU)、或数字信号处理器(DSP)、或可编程门阵列(FPGA)、快速缓冲贮存区(Cache)实现。It should be noted that, in practical applications, the functions of the creation module 11, the processing module 12, the detection module 13 and the migration module 14 may be provided by a central processing unit (CPU) or a microprocessor located on the device for realizing seamless data migration. device (MPU), or digital signal processor (DSP), or programmable gate array (FPGA), fast buffer storage area (Cache) implementation.
以上所述仅为本发明的较佳实施例而已，并非用于限定本发明的保护范围。The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510003665.2A CN105824571A (en) | 2015-01-05 | 2015-01-05 | Data seamless migration method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510003665.2A CN105824571A (en) | 2015-01-05 | 2015-01-05 | Data seamless migration method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105824571A true CN105824571A (en) | 2016-08-03 |
Family
ID=56513853
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510003665.2A Pending CN105824571A (en) | 2015-01-05 | 2015-01-05 | Data seamless migration method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105824571A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108092753A (en) * | 2017-12-15 | 2018-05-29 | 北京航天测控技术有限公司 | It is a kind of that synchronous hot backup redundancy system is moved using hardware memory |
| CN109521967A (en) * | 2018-11-15 | 2019-03-26 | 郑州云海信息技术有限公司 | A kind of moving method and device of data |
| CN109547532A (en) * | 2018-10-25 | 2019-03-29 | 交控科技股份有限公司 | A kind of method of data synchronization in multiserver redundant system between each server |
| CN110417584A (en) * | 2019-07-10 | 2019-11-05 | 南京南瑞继保电气有限公司 | A kind of two-shipper main/standby switching method based on multi-link election mechanism |
| WO2021051492A1 (en) * | 2019-09-18 | 2021-03-25 | 招商银行股份有限公司 | Database service node switching method, apparatus and device, and computer storage medium |
| CN115328065A (en) * | 2022-09-16 | 2022-11-11 | 中国核动力研究设计院 | Method for automatically migrating control unit functions applied to industrial control system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1553716A (en) * | 2003-06-04 | 2004-12-08 | 中兴通讯股份有限公司 | Clustering system for utilizing sharing internal memory in mobile communiation system and realizing method thereof |
| CN103199972A (en) * | 2013-03-25 | 2013-07-10 | 成都瑞科电气有限公司 | Double machine warm backup switching method and warm backup system achieved based on SOA and RS485 bus |
| CN103532753A (en) * | 2013-10-11 | 2014-01-22 | 中国电子科技集团公司第二十八研究所 | Double-computer hot standby method based on memory page replacement synchronization |
-
2015
- 2015-01-05 CN CN201510003665.2A patent/CN105824571A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1553716A (en) * | 2003-06-04 | 2004-12-08 | 中兴通讯股份有限公司 | Clustering system for utilizing sharing internal memory in mobile communiation system and realizing method thereof |
| CN103199972A (en) * | 2013-03-25 | 2013-07-10 | 成都瑞科电气有限公司 | Double machine warm backup switching method and warm backup system achieved based on SOA and RS485 bus |
| CN103532753A (en) * | 2013-10-11 | 2014-01-22 | 中国电子科技集团公司第二十八研究所 | Double-computer hot standby method based on memory page replacement synchronization |
Non-Patent Citations (2)
| Title |
|---|
| 张殿明 等: "《网络工程规划与设计》", 30 September 2010, 清华大学出版社 * |
| 陈琛: "基于内存同步双机热备系统", 《指挥信息系统与技术》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108092753A (en) * | 2017-12-15 | 2018-05-29 | 北京航天测控技术有限公司 | It is a kind of that synchronous hot backup redundancy system is moved using hardware memory |
| CN108092753B (en) * | 2017-12-15 | 2021-01-19 | 北京航天测控技术有限公司 | Hot standby redundancy system adopting hardware memory moving synchronization |
| CN109547532A (en) * | 2018-10-25 | 2019-03-29 | 交控科技股份有限公司 | A kind of method of data synchronization in multiserver redundant system between each server |
| CN109521967A (en) * | 2018-11-15 | 2019-03-26 | 郑州云海信息技术有限公司 | A kind of moving method and device of data |
| CN110417584A (en) * | 2019-07-10 | 2019-11-05 | 南京南瑞继保电气有限公司 | A kind of two-shipper main/standby switching method based on multi-link election mechanism |
| WO2021051492A1 (en) * | 2019-09-18 | 2021-03-25 | 招商银行股份有限公司 | Database service node switching method, apparatus and device, and computer storage medium |
| CN115328065A (en) * | 2022-09-16 | 2022-11-11 | 中国核动力研究设计院 | Method for automatically migrating control unit functions applied to industrial control system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US12073091B2 (en) | Low overhead resynchronization snapshot creation and utilization | |
| US9311328B2 (en) | Reference volume for initial synchronization of a replicated volume group | |
| US10949309B2 (en) | Snapshot creation with synchronous replication | |
| CN102761566B (en) | Method and device for migrating virtual machine | |
| CN102662751B (en) | A kind of method improving based on thermophoresis dummy machine system availability | |
| US10379964B2 (en) | Integrating resources at a backup site | |
| CN105824571A (en) | Data seamless migration method and device | |
| CN107911238A (en) | One kind is based on IPSAN servers two unit standby method and system | |
| US20140047263A1 (en) | Synchronous local and cross-site failover in clustered storage systems | |
| CN103532753B (en) | A kind of double hot standby method of synchronization of skipping based on internal memory | |
| US11163655B2 (en) | Systems and methods for resynchronizing mirroring partners in a storage system | |
| US12050558B2 (en) | Facilitating immediate performance of volume resynchronization with the use of passive cache entries | |
| CN102394914A (en) | Cluster brain-split processing method and device | |
| CN117827544B (en) | Hot backup system, method, electronic device and storage medium | |
| CN112783694B (en) | Long-distance disaster recovery method for high-availability Redis | |
| WO2022227719A1 (en) | Data backup method and system, and related device | |
| CN103605616A (en) | Multi-controller cache data consistency guarantee method | |
| CN105389231A (en) | Database dual-computer backup method and system | |
| US20250165167A1 (en) | Non-disruptively transition from asynchronous replication to bi-directional synchronous replication for a multi-site storage system | |
| CN112019601B (en) | Two-node implementation method and system based on distributed storage Ceph | |
| CN118101435B (en) | A high availability service method and system based on dual-machine hot standby | |
| CN117201507A (en) | Cloud platform switching method and device, electronic equipment and storage medium | |
| CN115499296A (en) | Cloud desktop hot standby management method, device and system | |
| CN114035999B (en) | A data migration method for virtual machines | |
| JP5870174B1 (en) | Data transmission system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160803 |