TWI819916B

Movatterモバイル変換

Info

Publication number: TWI819916B
Application number: TW111146952A
Authority: TW
Inventors: 鄭廷軒; 許順興; 簡士哲
Original assignee: 中華電信股份有限公司
Priority date: 2022-12-07
Filing date: 2022-12-07
Publication date: 2023-10-21
Also published as: TW202425579A

Abstract

A virtual machine in cloud service disaster recovery system and method based on distributed storage technology, which is applicable to local cloud service, local cloud storage service, backup cloud service and backup cloud storage service. The method includes receiving off-site backup and restoration instruction and setting the backup virtual machine and backup virtual disk in the backup cloud storage service to readable and writable status; searching backup intermediary data to obtain and send restore data and virtual machine specification information to create a backup virtual machine or backup virtual disk to be restored in the backup zone cloud service; starting the backup virtual machine or backup virtual disk in the backup zone cloud service; checking weather the backup virtual machine or backup virtual disk to be restored start normally; and sending an alarm message according to whether the backup virtual machine or backup virtual disk to be restored is started normally.

Description

Translated fromChinese

基於分散式儲存技術之雲服務虛擬機災難備援系統及其方法Cloud service virtual machine disaster recovery system and method based on distributed storage technology

本發明是有關於一種雲服務虛擬機災難備援系統及其方法，且特別是有關於一種基於分散式儲存技術之雲服務虛擬機災難備援系統及其方法。The present invention relates to a cloud service virtual machine disaster recovery system and a method thereof, and in particular, to a cloud service virtual machine disaster recovery system and method based on distributed storage technology.

雲端服務系統OpenStack與分散式儲存技術Ceph已在全球雲端技術扮演重要的角色，進一步推動以通用x86伺服器為底層硬體平台的軟體應用趨勢，Ceph系統符合高可用性且錯誤容忍等需求，不但具備高橫向擴充性，且具備資料保護與高可用特性。The cloud service system OpenStack and the distributed storage technology Ceph have played an important role in global cloud technology, further promoting the trend of software applications using general-purpose x86 servers as the underlying hardware platform. The Ceph system meets the requirements of high availability and error tolerance, and not only has High horizontal scalability, data protection and high availability features.

雖然OpenStack搭配Ceph可提供雲服務能夠確保本地資料被保護不易遺失資料，但是當本地機房發生意外災難時，卻沒有完整的災害備援機制以確保本地機房發生災害時能將雲用戶的虛擬機與虛擬磁碟恢復於異地機房。Although OpenStack and Ceph can provide cloud services to ensure that local data is protected and data is not easily lost, when an unexpected disaster occurs in the local computer room, there is no complete disaster recovery mechanism to ensure that when a disaster occurs in the local computer room, the cloud user's virtual machine can be connected to the virtual machine. The virtual disk is restored in a remote computer room.

本發明提供一種基於分散式儲存技術之雲服務虛擬機災難備援系統及其方法，不僅提供效能評估機制，將大量需要備援服務的虛擬機進行排隊處理，確保本地機房的服務穩定性，而且提出災難復原方法，於災難發生時，快速將客戶備援的服務還原。The present invention provides a cloud service virtual machine disaster backup system and method based on distributed storage technology. It not only provides a performance evaluation mechanism, queues a large number of virtual machines that require backup services, and ensures the service stability of the local computer room, but also Propose disaster recovery methods to quickly restore customer backup services when a disaster occurs.

本發明的一種基於分散式儲存技術之雲服務虛擬機災難備援系統，包括：本地雲服務及本地雲儲存服務，其中本地雲儲存服務包括本地虛擬機及本地虛擬磁碟，本地雲服務用以管理本地虛擬機及本地虛擬磁碟；備援區雲服務及備援區雲儲存服務，其中備援區雲儲存服務包括備援虛擬機及備援虛擬磁碟，備援區雲服務用以管理備援虛擬機及備援虛擬磁碟；以及異地備援調節器，異地備援調節器分別經本地雲服務與本地雲儲存服務通訊連接以及經備援區雲服務與備援區雲儲存服務通訊連接，異地備援調節器包括異地備援控制模組以及效能監控模組，異地備援控制模組包括：多機房儲存系統控制模組，用以接收異地備份還原指令，將備援區雲儲存服務中的備援虛擬機與備援虛擬磁碟設定為可讀寫狀態；多機房雲服務控制模組，與多機房儲存系統控制模組電性連接；中介資料備份模組，與多機房雲服務控制模組電性連接，用以搜尋備份的中介資料以獲取還原資料以及虛擬機規格資訊且傳送給多機房雲服務控制模組，以供多機房雲服務控制模組於備援區雲服務中建立待還原的備援虛擬機或備援虛擬磁碟，並且多機房雲服務控制模組啟動備援區雲服務中的備援虛擬機或備援虛擬磁碟；異地備援虛擬機還原模組，與中介資料備份模組電性連接，用以檢查待還原的備援虛擬機或備援虛擬磁碟是否正常啟動；以及資料還原告警模組，與異地備援虛擬機還原模組電性連接，以依據待還原的備援虛擬機或備援虛擬磁碟是否正常啟動來發送告警訊息。A cloud service virtual machine disaster recovery system based on distributed storage technology of the present invention includes: a local cloud service and a local cloud storage service. The local cloud storage service includes a local virtual machine and a local virtual disk. The local cloud service is used to Manage local virtual machines and local virtual disks; redundant area cloud services and redundant area cloud storage services. The redundant area cloud storage services include redundant virtual machines and redundant virtual disks, and the redundant area cloud services are used for management The redundant virtual machine and the redundant virtual disk; and the remote redundant regulator. The remote redundant regulator communicates with the local cloud storage service through the local cloud service and communicates with the redundant zone cloud storage service through the redundant zone cloud service. Connection, the remote backup regulator includes a remote backup control module and a performance monitoring module. The remote backup control module includes: a multi-machine room storage system control module, which is used to receive remote backup and restore commands and store the backup area in the cloud. The backup virtual machines and backup virtual disks in the service are set to a readable and writable state; the multi-machine room cloud service control module is electrically connected to the multi-machine room storage system control module; the intermediary data backup module is connected to the multi-machine room cloud The service control module is electrically connected to search the backup intermediary data to obtain restoration data and virtual machine specification information and transmit it to the multi-machine room cloud service control module for the multi-machine room cloud service control module to provide cloud services in the backup area. Create a backup virtual machine or backup virtual disk to be restored, and the multi-machine room cloud service control module starts the backup virtual machine or backup virtual disk in the backup area cloud service; the remote backup virtual machine restoration module Group, and intermediary data backup moduleIt is electrically connected to check whether the backup virtual machine or backup virtual disk to be restored is started normally; and the data recovery alarm module is electrically connected to the remote backup virtual machine restoration module to check whether the backup virtual machine or backup virtual disk to be restored is started normally. Alarm messages are sent based on whether the backup virtual machine or backup virtual disk starts normally.

在本發明的一實施例中，上述的雲服務虛擬機災難備援系統，其中效能監控模組包括：儲存系統效能監控模組，用以監控本地虛擬機、本地虛擬磁碟、備援虛擬機及備援虛擬磁碟的儲存效能，以獲取儲存效能資訊；跨機房網路監控模組，與儲存系統效能監控模組電性連接，用以監控本地雲服務及備援區雲服務的網路效能，以獲取網路效能資訊；以及效能告警模組，分別與跨機房網路監控模組、儲存系統效能監控模組電性連接，用以傳送儲存效能資訊以及網路效能資訊至多機房儲存系統控制模組，多機房儲存系統控制模組依據儲存效能資訊以及網路效能資訊對需進行資料備援的本地虛擬機或本地虛擬磁碟進行排隊處理，並控制進行資料備援的備援虛擬機或備援虛擬磁碟的數量。In an embodiment of the present invention, in the above-mentioned cloud service virtual machine disaster recovery system, the performance monitoring module includes: a storage system performance monitoring module for monitoring local virtual machines, local virtual disks, and backup virtual machines. And the storage performance of the backup virtual disk to obtain storage performance information; the cross-machine room network monitoring module is electrically connected to the storage system performance monitoring module to monitor the network of local cloud services and backup area cloud services. performance to obtain network performance information; and a performance alarm module that is electrically connected to the cross-computer room network monitoring module and the storage system performance monitoring module respectively to transmit storage performance information and network performance information to the multi-computer room storage system. Control module, the multi-machine room storage system control module queues local virtual machines or local virtual disks that need data backup based on storage performance information and network performance information, and controls the backup virtual machines for data backup. Or the number of spare virtual disks.

在本發明的一實施例中，上述的雲服務虛擬機災難備援系統，其中儲存效能資訊包括儲存吞吐量與儲存IOPS(Input/Output Operations Per Second)，網路效能資訊包括網路吞吐量。In an embodiment of the present invention, in the above-mentioned cloud service virtual machine disaster recovery system, the storage performance information includes storage throughput and storage IOPS (Input/Output Operations Per Second), and the network performance information includes network throughput.

在本發明的一實施例中，上述的雲服務虛擬機災難備援系統，其中異地備援控制模組更包括：異地備援演練模組，用以經由本地雲服務控制本地雲儲存服務中將要進行異地備援演練的本地虛擬機或本地虛擬磁碟暫停並進行軟刪除，再經由備援區雲服務控制備援區雲儲存服務中的備援虛擬機或備援虛擬磁碟啟動，以執行異地備援演練。In an embodiment of the present invention, in the above-mentioned cloud service virtual machine disaster recovery system, the remote backup control module further includes: a remote backup drill module for controlling the upcoming events in the local cloud storage service through the local cloud service. Conducting remote backup drillsThe local virtual machine or local virtual disk is suspended and soft-deleted, and then the backup virtual machine or backup virtual disk in the backup cloud storage service is controlled through the backup zone cloud service to start up to perform an off-site backup drill.

在本發明的一實施例中，上述的雲服務虛擬機災難備援系統，其中資料還原告警模組更用以於待還原的備援虛擬機或備援虛擬磁碟正常啟動時，發送異地備援完成的告警訊息；以及資料還原告警模組更用以於待還原的備援虛擬機或備援虛擬磁碟正常啟動失敗時，發送異地備援失敗的告警訊息以及還原失敗的備援虛擬機或備援虛擬磁碟的資訊。In an embodiment of the present invention, in the above-mentioned cloud service virtual machine disaster recovery system, the data recovery alarm module is further used to send an off-site backup when the backup virtual machine or backup virtual disk to be restored is normally started. The alarm message of backup completion; and the data recovery alarm module is also used to send an alarm message of remote backup failure and the backup virtual machine that failed to be restored when the backup virtual machine to be restored or the backup virtual disk fails to start normally. Or backup virtual disk information.

本發明的一種基於分散式儲存技術之雲服務虛擬機災難備援方法，適用於本地雲服務、本地雲儲存服務、備援區雲服務及備援區雲儲存服務中，其中本地雲儲存服務包括本地虛擬機及本地虛擬磁碟，本地雲服務用以管理本地虛擬機及本地虛擬磁碟，備援區雲儲存服務包括備援虛擬機及備援虛擬磁碟，備援區雲服務用以管理備援虛擬機及備援虛擬磁碟，方法包括：接收異地備份還原指令，將備援區雲儲存服務中的備援虛擬機與備援虛擬磁碟設定為可讀寫狀態；搜尋備份的中介資料以獲取並且傳送還原資料以及虛擬機規格資訊，以於備援區雲服務中建立待還原的備援虛擬機或備援虛擬磁碟；啟動備援區雲服務中的備援虛擬機或備援虛擬磁碟；檢查待還原的備援虛擬機或備援虛擬磁碟是否正常啟動；以及依據待還原的備援虛擬機或備援虛擬磁碟是否正常啟動來發送告警訊息。A cloud service virtual machine disaster recovery method based on distributed storage technology of the present invention is suitable for local cloud services, local cloud storage services, backup area cloud services and backup area cloud storage services, wherein the local cloud storage service includes Local virtual machines and local virtual disks. Local cloud services are used to manage local virtual machines and local virtual disks. Cloud storage services in the backup area include backup virtual machines and backup virtual disks. Cloud services in the backup area are used to manage them. Backup virtual machines and backup virtual disks. Methods include: receiving off-site backup and restore instructions, setting the backup virtual machines and backup virtual disks in the cloud storage service in the backup area to readable and writable status; searching for backup intermediaries Data to obtain and transmit restoration data and virtual machine specification information to create a backup virtual machine or backup virtual disk to be restored in the backup area cloud service; activate the backup virtual machine or backup in the backup area cloud service backup virtual disk; check whether the backup virtual machine or backup virtual disk to be restored is started normally; and send an alarm message based on whether the backup virtual machine or backup virtual disk to be restored is started normally.

在本發明的一實施例中，上述的雲服務虛擬機災難備援方法，其中方法更包括：監控本地虛擬機、本地虛擬磁碟、備援虛擬機及備援虛擬磁碟的儲存效能，以獲取儲存效能資訊；監控本地雲服務及備援區雲服務的網路效能，以獲取網路效能資訊；以及接收儲存效能資訊以及網路效能資訊，以依據儲存效能資訊以及網路效能資訊對需進行資料備援的本地虛擬機或本地虛擬磁碟進行排隊處理，並控制進行資料備援的備援虛擬機或備援虛擬磁碟的數量。In an embodiment of the present invention, the above cloud service virtual machine disaster recovery method further includes: monitoring the storage performance of the local virtual machine, the local virtual disk, the backup virtual machine and the backup virtual disk, to Obtain storage performance information; monitor the network performance of local cloud services and backup zone cloud services to obtain network performance information; and receive storage performance information and network performance information to meet needs based on storage performance information and network performance information. The local virtual machines or local virtual disks for data backup are queued for processing, and the number of backup virtual machines or backup virtual disks for data backup is controlled.

在本發明的一實施例中，上述的雲服務虛擬機災難備援方法，其中儲存效能資訊包括儲存吞吐量與儲存IOPS(Input/Output Operations Per Second)，網路效能資訊包括網路吞吐量。In an embodiment of the present invention, in the above cloud service virtual machine disaster recovery method, the storage performance information includes storage throughput and storage IOPS (Input/Output Operations Per Second), and the network performance information includes network throughput.

在本發明的一實施例中，上述的雲服務虛擬機災難備援方法，其中方法更包括：經由本地雲服務控制本地雲儲存服務中將要進行異地備援演練的本地虛擬機或本地虛擬磁碟暫停並進行軟刪除，再經由備援區雲服務控制備援區雲儲存服務中的備援虛擬機或備援虛擬磁碟啟動，以執行異地備援演練。In an embodiment of the present invention, the above-mentioned cloud service virtual machine disaster recovery method further includes: controlling the local virtual machine or local virtual disk in the local cloud storage service that is to be used for remote recovery drills through the local cloud service. Pause and perform soft deletion, and then control the startup of the backup virtual machine or backup virtual disk in the cloud storage service of the backup area through the backup area cloud service to perform an off-site backup drill.

在本發明的一實施例中，上述的雲服務虛擬機災難備援方法，其中依據待還原的備援虛擬機或備援虛擬磁碟是否正常啟動來發送告警訊息的步驟中，更包括：於待還原的備援虛擬機或備援虛擬磁碟正常啟動時，發送異地備援完成的告警訊息；以及於待還原的備援虛擬機或備援虛擬磁碟正常啟動失敗時，發送異地備援失敗的告警訊息以及還原失敗的備援虛擬機或備援虛擬磁碟的資訊。In an embodiment of the present invention, the above cloud service virtual machine disaster recovery method, the step of sending an alarm message based on whether the backup virtual machine or the backup virtual disk to be restored is normally started, further includes: When the backup virtual machine or backup virtual disk to be restored starts normally, an alarm message indicating that the remote backup is completed is sent; and when the backup virtual machine or backup virtual disk to be restored fails to start normally, an exception message is sent.Alarm messages for local backup failures and information about backup virtual machines or backup virtual disks that failed to be restored.

基於上述，本發明的一種基於分散式儲存技術之雲服務虛擬機災難備援系統及其方法，不僅提供效能評估機制，同時監控與控制多個機房的Ceph系統，並對Ceph進行效能評估與RBD-Mirror之相關操作，雲管人員能夠透過使用者介面清楚各機房Ceph系統效能狀態，並操作各機房Ceph系統的備援對象，選擇各機房須執行備援的本地虛擬機或本地虛擬磁碟，將大量需要備援服務的本地虛擬機進行排隊處理，確保本地機房的服務穩定性，而且提出災難復原方法，使得兩個或是多個區域的雲服務互相作為彼此的備援區，透過定期備份之中介資料關聯出備援虛擬磁碟與備援虛擬機之間的關聯，將備援虛擬機與備援虛擬磁碟建出還原服務，於災難發生時，快速將客戶備援的服務還原。Based on the above, the present invention provides a cloud service virtual machine disaster recovery system and method based on distributed storage technology. It not only provides a performance evaluation mechanism, but also monitors and controls Ceph systems in multiple computer rooms at the same time, and performs performance evaluation and RBD on Ceph. -Mirror-related operations, cloud administrators can understand the performance status of the Ceph system in each computer room through the user interface, and operate the backup objects of the Ceph system in each computer room, and select the local virtual machines or local virtual disks that need to be redundant in each computer room. Queue up a large number of local virtual machines that require backup services to ensure the service stability of the local computer room, and propose disaster recovery methods so that cloud services in two or more regions can serve as each other's backup areas through regular backups. The intermediary data associates the association between the backup virtual disk and the backup virtual machine, and creates a restoration service between the backup virtual machine and the backup virtual disk. When a disaster occurs, the customer's backup service can be quickly restored.

為讓本發明的上述特徵和優點能更明顯易懂，下文特舉實施例，並配合所附圖式作詳細說明如下。In order to make the above-mentioned features and advantages of the present invention more obvious and easy to understand, embodiments are given below and described in detail with reference to the accompanying drawings.

100:異地備援調節器100: Remote backup regulator

200:本地雲服務200:Local cloud service

300:本地雲儲存服務300:Local cloud storage service

400:備援區雲服務400: Backup area cloud service

500:備援區雲儲存服務500: Cloud storage service in backup area

110:異地備援控制模組110: Remote backup control module

120:效能監控模組120:Performance monitoring module

111:多機房儲存系統控制模組111:Multi-computer room storage system control module

112:多機房雲服務控制模組112:Multiple computer room cloud service control module

113:異地備援虛擬機還原模組113: Remote backup virtual machine restoration module

114:異地備援演練模組114: Remote backup drill module

115:中介資料備份模組115: Intermediary data backup module

116:資料還原告警模組116: Information returned to the police model team

121:儲存系統效能監控模組121:Storage system performance monitoring module

122:跨機房網路監控模組122: Cross-machine room network monitoring module

123:效能告警模組123: Performance alarm module

S201、S202、S203、S204、S205、S206、S207、S208、S209、S301、S302、S303、S304、S305、S306、S307:步驟S201, S202, S203, S204, S205, S206, S207, S208, S209, S301, S302, S303, S304, S305, S306, S307: Steps

圖1是依照本發明的一實施例的雲服務虛擬機災難備援系統的示意圖。Figure 1 is a schematic diagram of a cloud service virtual machine disaster recovery system according to an embodiment of the present invention.

圖2是依照本發明的一實施例的雲服務虛擬機災難備援方法的示意圖。Figure 2 is a schematic diagram of a cloud service virtual machine disaster recovery method according to an embodiment of the present invention.

圖3是依照本發明的一實施例的異地備援演練方法的示意圖。Figure 3 is a schematic diagram of a remote backup drill method according to an embodiment of the present invention.

本發明的部份實施例接下來將會配合附圖來詳細描述，以下的描述所引用的元件符號，當不同附圖出現相同的元件符號將視為相同或相似的元件。這些實施例只是本發明的一部份，並未揭示所有本發明的可實施方式。更確切的說，這些實施例只是本發明的專利申請範圍中的方法、電子裝置以及電腦可讀取儲存媒體的範例。Some embodiments of the present invention will be described in detail with reference to the accompanying drawings. The component symbols cited in the following description will be regarded as the same or similar components when the same component symbols appear in different drawings. These embodiments are only part of the present invention and do not disclose all possible implementations of the present invention. Rather, these embodiments are only examples of methods, electronic devices, and computer-readable storage media within the scope of the patent application of the present invention.

請參照圖1，基於分散式儲存技術之雲服務虛擬機災難備援系統包括異地備援調節器100、本地雲服務200、本地雲儲存服務300、備援區雲服務400以及備援區雲儲存服務500。Please refer to Figure 1. The cloud service virtual machine disaster recovery system based on distributed storage technology includes aremote backup regulator 100, alocal cloud service 200, a localcloud storage service 300, a backuparea cloud service 400, and a backup area cloud storage. Serves 500.

其中本地雲儲存服務300包括本地虛擬機及本地虛擬磁碟，本地雲服務200用以管理本地虛擬機及本地虛擬磁碟。The localcloud storage service 300 includes local virtual machines and local virtual disks, and thelocal cloud service 200 is used to manage local virtual machines and local virtual disks.

備援區雲儲存服務500包括備援虛擬機及備援虛擬磁碟，備援區雲服務400用以管理備援虛擬機及備援虛擬磁碟。The backup zonecloud storage service 500 includes backup virtual machines and backup virtual disks, and the backupzone cloud service 400 is used to manage the backup virtual machines and backup virtual disks.

異地備援調節器100分別經本地雲服務200與本地雲儲存服務300通訊連接以及經備援區雲服務400與備援區雲儲存服務500通訊連接。異地備援調節器100以微服務的方式執行，且可以於不同實體伺服器(即虛擬機)部署多個作高可用性(High Availability)的Active-Standby架構，使得運行於兩個伺服器的異地備援調節器100享有備援效果，且當一台伺服器意外暫停或故障，另外一台伺服器中的微服務能接手處理故障這台正在處理的備援任務，以確保備援服務不中斷。Theremote backup regulator 100 is communicated with the localcloud storage service 300 via thelocal cloud service 200 and communicates with the backup zonecloud storage service 500 via the backupzone cloud service 400. Theremote backup regulator 100 is executed in a microservice manner, andMultiple Active-Standby architectures for high availability (High Availability) can be deployed on different physical servers (i.e. virtual machines), so that theremote redundancy regulator 100 running on two servers can enjoy the redundancy effect and act as a If a server unexpectedly suspends or fails, the microservice in another server can take over the backup task being processed by the failed server to ensure that the backup service is not interrupted.

其中，異地備援調節器100包括異地備援控制模組110以及效能監控模組120。其中異地備援控制模組110包括多機房儲存系統控制模組111、多機房雲服務控制模組112、異地備援虛擬機還原模組113、異地備援演練模組114、中介資料備份模組115以及資料還原告警模組116。Among them, theremote backup regulator 100 includes a remotebackup control module 110 and aperformance monitoring module 120 . The remotebackup control module 110 includes a multi-computer room storagesystem control module 111, a multi-computer room cloudservice control module 112, a remote backup virtualmachine restoration module 113, a remotebackup drill module 114, and an intermediary data backup module. 115 and the information was returned to thepolice model team 116.

多機房儲存系統控制模組111用以接收異地備份還原指令，將備援區雲儲存服務500中的備援虛擬機與備援虛擬磁碟設定為可讀寫狀態。The multi-machine room storagesystem control module 111 is used to receive off-site backup and restore instructions, and set the backup virtual machine and backup virtual disk in the backup areacloud storage service 500 to a readable and writable state.

多機房雲服務控制模組112與多機房儲存系統控制模組111電性連接。The multi-machine room cloudservice control module 112 is electrically connected to the multi-machine room storagesystem control module 111.

中介資料備份模組115與多機房雲服務控制模組112電性連接，用以搜尋備份的中介資料以獲取還原資料以及虛擬機規格資訊且傳送給多機房雲服務控制模組112，以供多機房雲服務控制模組112於備援區雲服務400中建立待還原的備援虛擬機或備援虛擬磁碟，並且多機房雲服務控制模組112可啟動備援區雲服務400中的備援虛擬機或備援虛擬磁碟。The intermediarydata backup module 115 is electrically connected to the multi-machine room cloudservice control module 112 for searching the backed-up intermediary data to obtain restoration data and virtual machine specification information and transmit it to the multi-machine room cloudservice control module 112 for multi-machine room cloudservice control module 112. The computer room cloudservice control module 112 creates a backup virtual machine or a backup virtual disk to be restored in the backuparea cloud service 400, and the multi-machine room cloudservice control module 112 can activate the backup in the backuparea cloud service 400. backup virtual machine or backup virtual disk.

異地備援虛擬機還原模組113與中介資料備份模組115電性連接，用以檢查待還原的備援虛擬機或備援虛擬磁碟是否正常啟動。Remote backup virtualmachine restoration module 113 and intermediarydata backup module 115Electrical connection to check whether the backup virtual machine or backup virtual disk to be restored starts normally.

資料還原告警模組116與異地備援虛擬機還原模組113電性連接，以依據待還原的備援虛擬機或備援虛擬磁碟是否正常啟動來發送告警訊息。當待還原的備援虛擬機或備援虛擬磁碟正常啟動時，資料還原告警模組116可發送異地備援完成的告警訊息，當待還原的備援虛擬機或備援虛擬磁碟正常啟動失敗時，資料還原告警模組116可發送異地備援失敗的告警訊息以及還原失敗的備援虛擬機或備援虛擬磁碟的資訊。換言之，在RBD-Mirrorr進行資料備援時，若發生任何異常或是資料傳輸失敗，異地備援調節器100中的資料還原告警模組116遇到備份異常事件時會透過雲服務管理平台發出告警訊息，其備份異常事件可例如是備援區域連線問題、儲存空間滿載等等問題，從而確保備援資料能夠確實地進行資料備份，避免讓備援服務因異常暫停而雲管人員無法得知，造成無法挽回的錯誤。The datarecovery alarm module 116 is electrically connected to the remote backup virtualmachine restoration module 113 to send an alarm message based on whether the backup virtual machine or backup virtual disk to be restored is started normally. When the backup virtual machine or backup virtual disk to be restored is started normally, the datarecovery alarm module 116 can send an alarm message that remote backup is completed. When the backup virtual machine or backup virtual disk to be restored is started normally, When it fails, the datarecovery alarm module 116 can send an alarm message of the remote backup failure and the information of the backup virtual machine or the backup virtual disk that failed to be restored. In other words, when RBD-Mirrorr performs data backup, if any exception occurs or data transmission fails, the data in theremote backup regulator 100 will be restored. Thealarm module 116 will issue an alarm through the cloud service management platform when encountering a backup abnormal event. Messages, the backup abnormal events can be, for example, backup area connection problems, storage space full, etc., to ensure that the backup data can be accurately backed up, and to avoid the backup service being abnormally suspended and the cloud management personnel not being able to know , causing irreparable errors.

異地備援演練模組114用以經由本地雲服務200控制本地雲儲存服務300中將要進行異地備援演練的本地虛擬機或本地虛擬磁碟暫停並進行軟刪除，再經由備援區雲服務400控制備援區雲儲存服務500中的備援虛擬機或備援虛擬磁碟啟動，以執行異地備援演練。The remotebackup drill module 114 is used to control the local virtual machine or local virtual disk in the localcloud storage service 300 that will be used for the remote backup drill via thelocal cloud service 200 to suspend and soft delete it, and then use the backuparea cloud service 400 Control the startup of the backup virtual machine or backup virtual disk in thecloud storage service 500 in the backup area to perform a remote backup drill.

效能監控模組120包括儲存系統效能監控模組121、跨機房網路監控模組122以及效能告警模組。Theperformance monitoring module 120 includes a storage systemperformance monitoring module 121, a cross-machine roomnetwork monitoring module 122, and a performance alarm module.

其中，儲存系統效能監控模組121用以監控本地虛擬機、本地虛擬磁碟、備援虛擬機及備援虛擬磁碟的儲存效能，以獲取儲存效能資訊。跨機房網路監控模組122與儲存系統效能監控模組121電性連接，用以監控本地雲服務及備援區雲服務的網路效能，以獲取網路效能資訊。效能告警模組123分別與跨機房網路監控模組122、儲存系統效能監控模組121電性連接，用以傳送儲存效能資訊以及網路效能資訊至多機房儲存系統控制模組111，多機房儲存系統控制模組111依據儲存效能資訊以及網路效能資訊對需進行資料備援的本地虛擬機或本地虛擬磁碟進行排隊處理，並控制進行資料備援的備援虛擬機或備援虛擬磁碟的數量。在一實施例中，儲存效能資訊可包括儲存吞吐量與儲存IOPS(Input/Output Operations Per Second)，網路效能資訊包括網路吞吐量。本發明並不以此為限。Among them, the storage systemperformance monitoring module 121 is used to monitor the storage performance of local virtual machines, local virtual disks, backup virtual machines, and backup virtual disks to obtain storage performance information. The cross-computer roomnetwork monitoring module 122 is electrically connected to the storage systemperformance monitoring module 121 to monitor the network performance of local cloud services and backup area cloud services to obtain network performance information. Theperformance alarm module 123 is electrically connected to the cross-machine roomnetwork monitoring module 122 and the storage systemperformance monitoring module 121 respectively, and is used to transmit storage performance information and network performance information to the multi-machine room storagesystem control module 111. Multi-machine room storage Thesystem control module 111 queues the local virtual machines or local virtual disks that require data backup based on the storage performance information and the network performance information, and controls the backup virtual machines or backup virtual disks that need to perform data backup. quantity. In one embodiment, the storage performance information may include storage throughput and storage IOPS (Input/Output Operations Per Second), and the network performance information may include network throughput. The present invention is not limited thereto.

在一實施例中，雲用戶能夠自行決定需要備援的本地虛擬機與本地虛擬磁碟，並透過中介資料備份模組115將本地虛擬機與本地虛擬磁碟關聯中介資料，將實際位於本地雲儲存服務300中的本地虛擬機與本地虛擬磁碟資料關聯並交由多機房儲存系統控制模組111執行異地備援排程，多機房儲存系統控制模組111可以同時控制本地雲儲存服務300與備援區雲儲存服務500進行虛擬機與虛擬磁碟的資料備援，當然，雲儲存服務300與備援區雲儲存服務500可以互相作為彼此的備援儲存服務。In one embodiment, cloud users can decide on their own which local virtual machines and local virtual disks need to be backed up, and associate the local virtual machines and local virtual disks with intermediary data through the intermediarydata backup module 115, so that the local virtual machines and local virtual disks that are actually located in the local cloud can be associated with intermediary data. The local virtual machines in thestorage service 300 are associated with the local virtual disk data and are handed over to the multi-machine room storagesystem control module 111 to execute the remote backup schedule. The multi-machine room storagesystem control module 111 can simultaneously control the localcloud storage service 300 and The backup areacloud storage service 500 performs data backup of virtual machines and virtual disks. Of course, thecloud storage service 300 and the backup areacloud storage service 500 can serve as each other's backup storage services.

當本地雲服務機房發生自然災害時，若是本地雲服務200與本地雲儲存服務300全毀，多機房儲存系統控制模組111會控制備援區雲儲存服務500將備援虛擬機與備援虛擬磁碟設定為可讀寫狀態，並控制備援區雲服務400立即啟動位於備援區雲儲存服務500的備援虛擬機，啟動前，會透過中介資料備份模組115備份的中介資料將所有備援虛擬機與備援虛擬磁碟作關聯，並將本地虛擬機與本地虛擬磁碟服務還原到發生災害前的配置。When a natural disaster occurs in the local cloud service computer room, if thelocal cloud service 200With the localcloud storage service 300 completely destroyed, the multi-machine room storagesystem control module 111 will control the backup areacloud storage service 500 to set the backup virtual machine and backup virtual disk to a read-write state, and control the backuparea cloud Service 400 immediately starts the backup virtual machines located in thecloud storage service 500 in the backup area. Before starting, all backup virtual machines will be associated with the backup virtual disks through the intermediary data backed up by the intermediarydata backup module 115, and will Local virtual machines and local virtual disk services are restored to their pre-disaster configurations.

請參照圖2，於步驟S201中，儲存系統效能監控模組121監控本地虛擬機、本地虛擬磁碟、備援虛擬機及備援虛擬磁碟的儲存效能，以獲取儲存效能資訊。Referring to FIG. 2 , in step S201 , the storage systemperformance monitoring module 121 monitors the storage performance of the local virtual machine, the local virtual disk, the backup virtual machine, and the backup virtual disk to obtain storage performance information.

於步驟S202中，跨機房網路監控模組122監控本地雲服務及備援區雲服務的網路效能，以獲取網路效能資訊。In step S202, the cross-machine roomnetwork monitoring module 122 monitors the network performance of the local cloud service and the backup area cloud service to obtain network performance information.

於步驟S203中，效能告警模組123依據接收的儲存效能資訊以及網路效能資訊對需進行資料備援的本地虛擬機或本地虛擬磁碟進行排隊處理，並控制進行資料備援的備援虛擬機或備援虛擬磁碟的數量。In step S203, theperformance alarm module 123 queues the local virtual machines or local virtual disks that need data backup based on the received storage performance information and network performance information, and controls the backup virtual machines for data backup. The number of machines or spare virtual disks.

於步驟S204中，多機房儲存系統控制模組111接收異地備份還原指令，將備援區雲儲存服務500中的備援虛擬機與備援虛擬磁碟設定為可讀寫狀態。In step S204, the multi-machine room storagesystem control module 111 receives the remote backup and restore command and sets the backup virtual machine and the backup virtual disk in the backup areacloud storage service 500 to a readable and writable state.

於步驟S205中，中介資料備份模組115搜尋備份的中介資料以獲取並且傳送還原資料以及虛擬機規格資訊，以於備援區雲服務400中建立待還原的備援虛擬機或備援虛擬磁碟。In step S205, the intermediarydata backup module 115 searches for the backed-up intermediary data to obtain and transmit restoration data and virtual machine specification information to the backup area.Create a backup virtual machine or backup virtual disk to be restored in thecloud service 400 .

於步驟S206中，多機房雲服務控制模組112啟動備援區雲服務中的備援虛擬機或備援虛擬磁碟。In step S206, the multi-machine room cloudservice control module 112 activates the backup virtual machine or backup virtual disk in the backup area cloud service.

於步驟S207中，異地備援虛擬機還原模組113檢查待還原的備援虛擬機或備援虛擬磁碟是否正常啟動。In step S207, the remote backup virtualmachine restoration module 113 checks whether the backup virtual machine or backup virtual disk to be restored is started normally.

於步驟S208中，當待還原的備援虛擬機或備援虛擬磁碟正常啟動時，資料還原告警模組116發送異地備援完成的告警訊息In step S208, when the backup virtual machine or the backup virtual disk to be restored is started normally, the data returnalarm module 116 sends an alarm message that remote backup is completed.

於步驟S209中，當待還原的備援虛擬機或備援虛擬磁碟正常啟動失敗時，資料還原告警模組116發送異地備援失敗的告警訊息以及還原失敗的備援虛擬機或備援虛擬磁碟的資訊。In step S209, when the backup virtual machine or backup virtual disk to be restored fails to start normally, the datarecovery alarm module 116 sends an alarm message of the remote backup failure and the backup virtual machine or backup virtual disk that fails to be restored. Disk information.

上述實施例為雲服務虛擬機災難備援方法，不僅能夠控制本地端之Ceph系統(即本地雲儲存服務300中的本地虛擬機與本地虛擬磁碟)，將本地雲儲存服務300中的本地虛擬機與本地虛擬磁碟以RBD-Mirror的方式以非同步的方式備份至異地機房Ceph系統(即備援區雲儲存服務500中的備援虛擬機與備援虛擬磁碟)，且為了能夠預防雲服務因災難造成資料庫與相關中介資料遺失，可通過評估當前本地儲存效能並依照效能的狀況針對需進行備援之虛擬機與虛擬硬碟進行排隊處理及排程，將雲服務中重要的資料庫與中介資料備份至異地儲存空間，用以確保本地機房的服務穩定性。The above embodiment is a cloud service virtual machine disaster recovery method, which can not only control the local Ceph system (that is, the local virtual machine and the local virtual disk in the local cloud storage service 300), but also control the local virtual machine in the localcloud storage service 300. The machine and the local virtual disk are backed up asynchronously to the Ceph system in the remote computer room (i.e., the backup virtual machine and the backup virtual disk in the backup area cloud storage service 500) in an RBD-Mirror manner, and in order to prevent If a cloud service loses its database and related intermediary data due to a disaster, important data in the cloud service can be saved by evaluating the current local storage performance and queuing and scheduling the virtual machines and virtual hard disks that need to be backed up according to the performance status. The database and intermediary data are backed up to off-site storage space to ensure the service stability of the local computer room.

圖3是依照本發明的一實施例的異地備援演練方法的示意圖。Figure 3 is an illustration of a remote backup drill method according to an embodiment of the present invention.intention.

請參照圖3，於步驟S301中，經由本地雲服務200控制本地雲儲存服務300中將要進行異地備援演練的本地虛擬機或本地虛擬磁碟暫停並進行軟刪除，再經由備援區雲服務400控制備援區雲儲存服務500中的備援虛擬機或備援虛擬磁碟啟動。Referring to Figure 3, in step S301, the local virtual machine or local virtual disk that is to be used for remote backup drills in the localcloud storage service 300 is controlled through thelocal cloud service 200 to be paused and soft deleted, and then the local virtual machine or local virtual disk is soft deleted through the backup area cloud service. 400 controls the startup of the backup virtual machine or backup virtual disk in thecloud storage service 500 in the backup area.

於步驟S302中，多機房儲存系統控制模組111接收異地備份還原指令，將備援區雲儲存服務500中的備援虛擬機與備援虛擬磁碟設定為可讀寫狀態。In step S302, the multi-machine room storagesystem control module 111 receives the remote backup and restore command and sets the backup virtual machine and the backup virtual disk in the backup areacloud storage service 500 to a readable and writable state.

於步驟S303中，中介資料備份模組115搜尋備份的中介資料以獲取並且傳送還原資料以及虛擬機規格資訊，以於備援區雲服務400中建立待還原的備援虛擬機或備援虛擬磁碟。In step S303, the intermediarydata backup module 115 searches for the backed up intermediary data to obtain and transmit restoration data and virtual machine specification information to create a backup virtual machine or backup virtual disk to be restored in the backupzone cloud service 400. dish.

於步驟304中，多機房雲服務控制模組112啟動備援區雲服務中的備援虛擬機或備援虛擬磁碟。In step 304, the multi-machine room cloudservice control module 112 starts the backup virtual machine or backup virtual disk in the backup area cloud service.

於步驟S305中，異地備援虛擬機還原模組113檢查待還原的備援虛擬機或備援虛擬磁碟是否正常啟動。In step S305, the remote backup virtualmachine restoration module 113 checks whether the backup virtual machine or backup virtual disk to be restored is started normally.

於步驟S306中，當待還原的備援虛擬機或備援虛擬磁碟正常啟動時，資料還原告警模組116發送異地備援完成的告警訊息。In step S306, when the backup virtual machine or backup virtual disk to be restored is started normally, the data returnalarm module 116 sends an alarm message indicating that remote backup is completed.

於步驟S307中，當待還原的備援虛擬機或備援虛擬磁碟正常啟動失敗時，資料還原告警模組116發送異地備援失敗的告警訊息以及還原失敗的備援虛擬機或備援虛擬磁碟的資訊。In step S307, when the backup virtual machine or backup virtual disk to be restored fails to start normally, the datarecovery alarm module 116 sends an alarm message of the remote backup failure and the backup virtual machine or backup virtual disk that fails to be restored. Disk information.

在一實施例中，雲用戶或是雲管人員亦能夠於適當時間透過異地備援控制模組110進行異地備援演練，異地備援演練模組114可控制本地雲服務200將要進行異地備援演練的本地虛擬機與本地虛擬磁碟暫停並進行軟刪除，再透過同樣的方式將備援虛擬機與備援虛擬磁碟於備援區雲服務400與備援區雲儲存服務500啟動，以模擬災害發生後用戶服務於兩地機房切換之場景。In one embodiment, cloud users or cloud administrators can alsoA remote backup drill is performed through the remotebackup control module 110. The remotebackup drill module 114 can control thelocal cloud service 200 to suspend and soft-delete the local virtual machine and local virtual disk that are to be used for the remote backup drill, and then through In the same way, the backup virtual machine and the backup virtual disk are started in the backuparea cloud service 400 and the backup areacloud storage service 500 to simulate a scenario in which user services are switched between computer rooms in two places after a disaster occurs.

上述實施例為異地備援演練方法，不僅可提供快速穩定的災難復原功能，雲管人員能夠於雲服務使用者介面查看所有虛擬機與虛擬磁碟的備援的狀態，雲管人員能夠點選指定標的，並針對選擇的標的進行災害復原演練，其備援資料的復原皆以自動化方式進行還原，並且可定期透過備份中介資料，關聯出備援虛擬磁碟與虛擬機之間的關聯，當真正的災害發生時，異地備援調節器100能夠根據備份中介資料自動化地將備援虛擬機與備援虛擬磁碟建出，還原用戶於雲平台所建之服務。The above embodiment is an off-site backup drill method, which not only provides fast and stable disaster recovery functions, but also allows cloud administrators to view the backup status of all virtual machines and virtual disks on the cloud service user interface. Cloud administrators can click Specify the target and conduct disaster recovery drills for the selected target. The recovery of the backup data is restored in an automated manner, and the association between the backup virtual disk and the virtual machine can be associated regularly through backup intermediary data. When When a real disaster occurs, theremote backup regulator 100 can automatically create backup virtual machines and backup virtual disks based on the backup intermediary data, and restore services created by users on the cloud platform.

基於上述，本發明的一種基於分散式儲存技術之雲服務虛擬機災難備援系統及其方法，不僅提供效能評估機制，同時監控與控制多個機房的Ceph系統，並對Ceph進行效能評估與RBD-Mirror之相關操作，雲管人員能夠透過使用者介面清楚各機房Ceph系統效能狀態，並操作各機房Ceph系統的備援對象，選擇各機房須執行備援的本地虛擬機或本地虛擬磁碟，將大量需要備援服務的本地虛擬機進行排隊處理，確保本地機房的服務穩定性，而且提出災難復原方法，使得兩個或是多個區域的雲服務互相作為彼此的備援區，透過定期備份之中介資料關聯出備援虛擬磁碟與備援虛擬機之間的關聯，將備援虛擬機與備援虛擬磁碟建出還原服務，於災難發生時，快速將客戶備援的服務還原。Based on the above, the present invention provides a cloud service virtual machine disaster recovery system and method based on distributed storage technology. It not only provides a performance evaluation mechanism, but also monitors and controls Ceph systems in multiple computer rooms at the same time, and performs performance evaluation and RBD on Ceph. -Mirror-related operations, cloud administrators can understand the performance status of the Ceph system in each computer room through the user interface, and operate the backup objects of the Ceph system in each computer room, and select the local virtual machines or local virtual disks that need to be redundant in each computer room. Queue up a large number of local virtual machines that require backup services to ensure the service stability of the local computer room, and propose disaster recovery methods so that cloud services in two or more regions can serve as each other's backup areas through regular backups. The intermediary information is associated with the backup virtualThe association between the disk and the backup virtual machine creates a restoration service for the backup virtual machine and the backup virtual disk, so that when a disaster occurs, the customer's backup service can be quickly restored.

雖然本揭露已以實施例揭露如上，然其並非用以限定本揭露，任何所屬技術領域中具有通常知識者，在不脫離本揭露的精神和範圍內，當可作些許的更動與潤飾，故本揭露的保護範圍當視後附的申請專利範圍所界定者為準。Although the disclosure has been disclosed above through embodiments, they are not intended to limit the disclosure. Anyone with ordinary knowledge in the technical field may make slight changes and modifications without departing from the spirit and scope of the disclosure. Therefore, The scope of protection of this disclosure shall be determined by the scope of the appended patent application.