WO2015167320A1

WO2015167320A1 - A system and method for integrated backup solution in virtualization environments

Info

Publication number: WO2015167320A1
Application number: PCT/MY2015/050022
Authority: WO
Inventors: Bukhary Ikhwan ISMAIL; Mohammad Fairus Khalid; Mohd Nizam MOHD MYDIN; Hong Hoe ONG
Original assignee: Mimos Bhd
Current assignee: Mimos Bhd
Priority date: 2014-04-28
Filing date: 2015-04-03
Publication date: 2015-11-05
Anticipated expiration: 2016-10-28
Also published as: MY186687A

Abstract

A system and method for integrated backup solution in virtualization environments makes use of the existing virtualization infrastructure to handle the backup process and peer with infrastructure at a different site for complying with business continuity requirements. The system (200) of the present invention comprising a Backup Controller (210) which further comprises a Web Service Interface (211) component that acts as an interface to the user; a Backup Engine (212) component that manages and coordinates the backup and restore process; a Host Monitor (214) component that monitors host load; a Replication Manager (213) component that manages the replication of backup images from a site to its peer site; a Backup Database (215) component that stores the backup information; a Backup Datastore (231 ) component that stores the backup information; and a Backup Agent (221) component that manages the backup activity at the node level. The objective of the present invention is achieved through visualization infrastructure to handle the backup process; and peering with infrastructure at a different site for complying with business continuity requirements.

Description

A SYSTEM AND METHOD FOR INTEGRATED BACKUP SOLUTION IN

VISUALIZATION ENVIRONMENTS

FIELD OF INVENTION

The present invention relates to a system and method for integrated backup soiution in virtualization environments, in particuiar, the invention relates to systems and methods that make use of the existing virtualization infrastructure to handle the backup process and peer with infrastructure at a different site for complying with business continuity requirements.

BACKGROUND ART Introducing backup services in cloud computing environments requires a dedicated infrastructure (f.e, server(s) to execute back-up processes and storage to store the backup elements). To compiy with business continuity requirements, backed-up data needs to be replicated off-site for safety, There are various problems associated with this process. For example, various issues may include the fact that the backup workload depends on how many and how frequent users are requesting such services. That is, requests are not consistent. There may be times where the number of backup activities requiring processing by the service provider is high while there are intervals where there is no backup workload to managed. A dedicated resource for backup operation may need to be planned up front. Under commissioning will generally result in long queues of backup operation waiting to be serviced. This will impact the Service Level Agreement (SLA) in place with the end users. Over commissioning will result in wastage through idle resources. United States Patent No. 8,387,054 B1 discloses a method and apparatus for scheduling data movers is described. The method and apparatus comprise monitoring utilization of a resource used by a data mover to process a data flow, determining rank information based on the utilization of the resource and scheduling a task according to the rank information. This document discloses a method in which backup and restore operations are executed on the highest ranked BA.

United States Patent No. 8,046.550 B2 describes backup systems and methods for a virtual computing environment. Certain examples include a system having a backup management server that communicates with a host server having at least one virtual machine. The management server coordinates with the host server to perform backup copies of entire virtual machine disks from outside the guest operating system of the virtual machine. In certain examples, such backup systems further utilize a volume shadow copy service executing on the host server to quiesce virtual machine applications to put data in a consistent state to be backed up. The backup system then utilizes hypervisor snapshot capabilities of the host server to record intended changes to the virtual machine disk fifes while such fifes are being copied (e.g., backed up) by the host server. Such recorded changes can be later committed to the virtual machine disk files once the backup operation has completed. This document discloses systems and methods in which backup and restore operations are executed on a single external machine.

The publication sRQW: An Efficient Live Snapshot System for Virtual Machine Disk (ICPADS), 2012 IEEE 18th international Conference on Parallel and Distributed Systems, Pages 378-383, suggests that the high-availability of mission-critical data and services hosted in a virtual machine (V ) is one of the top concerns in a cloud computing environment. In the IROW publication, for each of backup and restore operation, consistent state is requested. Virtual Machine is snapshot and the backup operation is executed on a singie external machine which improves the copy-and-write snapshot mechanism.

The overall objective of the present invention is achieved through virtuallzation infrastructure to handle the backup process; and peering with infrastructure at a different site for complying with business continuity requirements. SUMMARY OF INVENTION

The present invention relates to a system and method for integrated backup solution in visualization environments. In particular, the present invention relates to systems and methods which distributes backup process workload across nodes and replicates the backup image into peer sites,

One aspect of the present invention provides a system for distributed backup of workload across nodes and replication of backup information into peer sites, the system comprising a Backup Controller comprising a Web Service Interface component that acts as an interface to the user; a Backup Engine component that manages and coordinates the backup and restore process; a Host Monitor component that monitors host load; a Replication Manager component that manages the replication of backup images from a site to its peer site; a Backup Database component that stores the backup information; a Backup Datastore component that stores the backup information; and a Backup Agent component that manages the backup activity at the node level.

Another aspect of the present invention provides a system wherein the backup information comprises backup images, corresponding virtual machine (VM) images and backup image chunks.

A further aspect of the present invention provides a method for distributed backup of workload across nodes and replication of backup information into peer sites, the method comprising receiving a request to backup a VM from a user; locating the VM to backup; flushing the VM system state into a VM image; taking a snapshot of the VM image and copying it into a Backup Datastore as a backup image; assigning a least loaded node to split the backup image into a plurality of backup image chunks; informing Backup Agents in all nodes to start the backup process; and storing backup information into a Backup Database,

Yet another aspect of the present invention provides a method wherein the backup information comprises the backup image chunks. Sti!l another aspect of the present invention provides a method wherein informing Backup Agents in all nodes to start the backup process further comprises steps of receiving a request to process backup from a Backup Controi!er; checking the node has resource capacity to process a backup image chunk; checking there Is a backup image chunk available to be processed; if available, processing a backup image chuck; and informing the Backup Controller of the outcome of the process.

Another aspect of the present invention provides a method wherein the processing of the backup image chunk comprises but not limited to compression, encryption or deduplication of the backup image chunks.

A further aspect of the present invention provides a method wherein a Replication manager manages the replication of backup images to peer sites, the replication management comprising detecting the availability of processed backup images: identifying a configurable number of nodes with least load; informing a peer site to prepare a similar number of nodes to accept file transfer; parallel transferring the backup images to the peer site; and updating the backup information in the Backup Database.

Another aspect of the present invention is provided with a restoring process which manages the restoration of backed up images, the restore process comprising receiving a restore V request from user; locating the backup image; informing the Backup Agent in aii the nodes to start reversing the backup process; assigning the least loaded node to merge the backup image chunks; copying the backup image into an Image Datastore; informing a Cloud Controller to launch the copied backup image; and updating the backup information in the Backup Database.

Yet another aspect of the present invention provides a method wherein the reversing of the backup process comprises decryption or uncompressing of the backup image chunks.

The present invention consists of features and a combination of parts hereinafter fully described and illustrated in the accompanying drawings, it being understood that various changes In the details may be made without departing from the scope of the invention or sacrificing any of the advantages of the present invention.

BRIEF DESCRIPTION OF ACCOMPANYING DRAWINGS

To further clarify various aspects of some embodiments of the present invention, a more particular description of the invention will be rendered by references to specific embodiments thereof, which are illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail through the accompanying drawings in which: FIG. 1 illustrates a distributed backup context of a system according to an embodiment of the invention.

FIG. 2 illustrates distributed backup component architecture of a system according to an embodiment of the invention.

FIG. 3 illustrates the backup process of a system according to an embodiment of the invention.

FIG. 4 illustrates the architecture of the backup process of the present invention.

FIG. 4a illustrates a backup process simulation of a system according to an embodiment of the invention.

FIG. 5 illustrates a backup agent process of a system according to an embodiment of the invention.

FIG. 6 illustrates a backup replication process of a system according to an embodiment of the invention. FIG. 7 illustrates a backup replication simulation of a system according to an embodiment of the invention. FIG. 8 illustrates a restore process of a system according to an embodiment of the invention.

FIG, 9 iiiustrates a restore process simulation of a system according to an embodiment of the invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The present invention relates to a system and method for integrated backup soiution in visualization environments. In particular, the invention relates to systems and methods that make use of the existing virtuaiization infrastructure to handle the backup process and peer with infrastructure at a different site for complying with business continuity requirements. To integrate the backup service into the virtuaiization infrastructure and minimize the impact of backup/restore workload, the backup data is segmented into smaller predefined sizes and the workload is distributed amongst the nodes. Nodes with insufficient processing capacity are not be used as processing nodes.

Hereinafter, this specification will describe the present invention according to the preferred embodiments. It is to be understood that limiting the description to the preferred embodiments of the invention is merely to facilitate discussion of the present invention and it is envisioned without departing from the scope of the appended claims.

Referring to Figure 1 , a simple distributed backup context design 100 is illustrated. This includes a pair of sites, X and Y, storage 101 , a plurality of nodes 102, a cloud controller 103 and backup controller 104. Further discussion of these components will be described in the following description of the subsequent Figures.

Referring to Figure 2, a system 200 of an embodiment of the invention is illustrated in more detail below. The system 200 comprises a backup controller 210 that includes a web service interface 211. a backup engine 212, a replication manager 213, a host monitor 214 and a backup database 215. The system 200 also includes at least one backup agent 221 associated with at least one node 220. Each node comprises a hypervisor 222 and is designated, as illustrated, with associated virtual machines VM,, VM₂,.....VM_n. Storage 230 is also provided and includes a backup datastore 231 and a VM image datastore 232.

The backup methodology according to an embodiment of the invention is illustrated by way of flowchart in Figure 3, According to the backup method 300, a backup controller (BC) receives a request 301 from a user to backup a virtual machine (VM). Prior to the initiation of the backup, the user must ensure proper handling of application data or application transactions to be properly flush or stored in the disk at the time snapshot is taken, The backup engine (BE) locates the VM to backup 302 and sends a command to the hypervisor to flush the system states into the VM image 303, The BE then sends a command to the hypervisor to take a snapshot of the VM 304 and copies the snapshot delta into the backup datastore 305. The duration of the copying process is subjected to the size of the snapshot delta.

The host monitor then identifies the node with least load 306 and the BE sends a command to the backup agent (BA) of the identified node to split the backup delta 307. The delta is split into predefined chunks 308 and the BA informs the BE on completion of this process 309. The BE then broadcasts to all BA with "sufficient resources" to start processing the chunks of backup delta 310. "Sufficient resources" is a server with enough CPU, memory, storage I/O, etc resources. The data process may involve which is not limited to deduplication, compression and encryption, as previously described. Finally, the BE stores the backup information in the backup database 311.

The methodology of Figure 3 is illustrated further in Figure 4 and Figure 4a. Figure 4 illustrates the architecture of the backup process while Figure 4a illustrates a backup process simulation according to this embodiment of the invention. In this illustration, a backup request is received 401. The backup engine 412 locates the requested VM 402 on node 420'. The BE 412 requests that the hypervisor 422^! flush the system state 403. The BE 412 then sends a command to the hypervisor 422' to snapshot the VM 404 and copies the snapshot delta 405 into the backup store 431.

The host monitor 414 finds 406 the least load node 420". The BE 412 sends a command to the backup agent 421" of the node 420" to split the delta 407 he BA 421" informs the BE 412 on completed status 409. The BE 412 broadcasts to all BA 421" with sufficient resources to process the chunks of backup delta 410.

Referring to Figure 5, a backup agent process 500 is illustrated. The backup agent process 500 comprises initial receipt by the BA of a request to start the backup process from the BE 501 , followed by checking the node has resource capacity to process a backup image chunk 502. if not, the BA informs the BE 503. If the node does have capacity, a check is conducted to determine if there is a backup image chunk available to be processed 504. If available, the backup image chuck is marked for processing 505 and processed 508, Once the process is completed, the chunk is marked as processed. 507,

The methodology for distributed backup of workload across nodes and replication of backup information into peer sites further comprising managing replication of backup images to peer sites. The backup replication methodoiogy 800 is illustrated in Figure 6. Referring to this figure, the replication manager (RM) detects the availability of processed backup images 601. It then identifies a configurable number of nodes with "least load" 802 and informs the peer backup controller (BC) 803, , "Least load" nodes is a server in which its CPU, memory, storage I/O, etc is not under heavy load. The peer BC prepares a similar number of nodes with least load to accept replication and inform the RM 804, The RM then informs the BA of the identified nodes to start parallel transferring of the backup images to the peer site 605. Subsequently, the RM updates the backup information in the backup database 606 and the peer BC updates the backup information in the Backup Database 607. The methodology of Figure 6 is illustrated further in Figure 7, which illustrates a backup replication simulation 700 according to this embodiment of the invention. As previously described, the process involves initial detection of availability of processed backup data 701 by the RM 713. The RM 713 then identifies local nodes with least load 702 and informs BC 710 for the replication process 703. The peer RM 713' finds a similar number of nodes with least load to accept replication 704 and informs the BA to start the parallel transfer of data 705. The source RM 713 updates the backup information 708 in the backup database 715 and the peer RM 713' updates the backup information in the peer backup database 715'. Referring to Figure 8, a restoring process 800 is illustrated. The restoring process is a further step from the methodology for distributed backup of workload across nodes and replication of backup information into peer sites further comprising managing replication of backup images to peer sites. The process 800 generally includes receiving a restore VM request from a user 801 , The BE then finds the backup information from the backup database 802 and identifies whether the information is found 803, if not, the BE queries the peer BC 804 and determines whether information is found 805, if not, the peer BC replies to the query accordingiy 806, If so, the peer BC initiates the transfer process 807 and replies to the query and updates both backup databases 808.

If the information is found, the BE informs all Backup Agents in a!! the nodes to start reversing the backup process by uncompressing and decrypting the chunks backup data 809. The host monitor then assigns the least loaded node 810 to merge the backup image chunks 81 1. The BA merges the backup image into an Image Datastore 812, and the BE copies the backup image from the backup datastore to the image datastore 813. The BE then informs the Cloud Controller to launch the copied backup image 814.

The methodology of Figure 8 is illustrated further in Figure 9, which illustrates a backup replication simulation 900 according to this embodiment of the invention. As illustrated in Figure 9, BA receives request to restore data 902. Thereafter, BE locates backup information from Backup Database 904 and broadcast to all BAs to process the chunks by decrypting and uncompressing the information. The Host Monitor finds out the host with the least node 908 and further sends command to the BA of the identified node to merge the chunks of the backup image 910. Subsequently, BE copies the backup image to Image Datastore 912.

The present invention manages backup workload by enabling a Cloud Computing operator to optimize the usage of existing infrastructure. The present invention further provides backup services to cloud users by making use of the existing virtuaiization infrastructure to handle the backup process; and peering with infrastructure at a different site for complying with business continuity requirements.

Unless the context requires otherwise or specifically stated to the contrary, integers, steps or elements of the invention recited herein as singular integers, steps or elements clearly encompass both singular and piurai forms of the recited integers, steps or elements. Throughout this specification, unless the context requires otherwise, the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated step or element or integer or group of steps or elements or integers, but not the exclusion of any other step or element or integer or group of steps, elements or integers. Thus, in the context of this specification, the term "comprising" is used In an inclusive sense and thus should be understood as meaning "including principally, but not necessarily solely".

It will be appreciated that the foregoing description has been given by way of illustrative example of the invention and that all such modifications and variations thereto as would be apparent to persons of skill in the art are deemed to fall within the broad scope and ambit of the invention as herein set forth.

Claims

1. A system (200) for distributed backup of workload across nodes and replication of backup information into peer sites, the system comprising:

a Backup Controller (210) comprising:

a Web Service interface (211 ) component that acts as an interface to the user;

a Backup Engine (212) component that manages and coordinates the backup and restore process;

a Host Monitor (214) component that monitors host load;

a Replication Manager (213) component that manages the replication of backup images from a site to its peer site;

a Backup Database (215) component that stores the backup information:

a Backup Datastore (231) component that stores the backup information; and

a Backup Agent (221) component that manages the backup activity at the node level.

2. A system (200) as claimed in claim 1 , wherein said backup Information comprises backup images, corresponding virtual machine (VM) images and backup image chunks.

3. A method (300) for distributed backup of workload across nodes and replication of backup information into peer sites, the method comprising:

receiving a request to backup a VM from a user (301);

locating the VM to backup (302);

flushing the VM system state into a VM image (303);

taking a snapshot of the VM image (304) and copying it into a Backup

Datastore as a backup image (305);

assigning a least load node to split the backup image into a plurality of backup image chunks (306); informing Backup Agents in aii nodes to start the backup process (307); and

storing backup information into a Backup Database (311)

A method (300) as claimed in claim 3, wherein said backup information comprises said backup image chunks.

A method (300) as ciaimed in claim 3, wherein informing Backup Agents in ail nodes to start the backup process further comprising steps of:

receiving a request to process backup from a Backup Engine (501 ); checking the node has resource capacity to process a backup image chunk (502);

checking there is a backup image chunk available to be processed (504); if available, processing a backup image chuck (505); and

informing the Backup Controller of the outcome of the process (503).

A method (300) as claimed in claim 5, wherein said processing of said backup image chunk comprises compression, encryption or deduplication of said backup image chunks.

A method (300) as claimed in claim 3, further comprising managing replication of backup images to peer sites, the replication process further comprising steps of: detecting the availability of processed backup images (601);

identifying a configurable number of nodes with least load (802);

informing a peer site to prepare a similar number of nodes to accept file transfer (603);

parallel transferring the backup images to the peer site (604); and updating the backup information in the Backup Database (806, 607),

A method as claimed in claim 3, is provided with a restoring process which manages the restoration of backed up images, the restoring process comprising steps of:

receiving a restore V request from user (801 ); locating the backup image (802);

informing the Backup Agent in all the nodes to start reversing the backup process (809);

assigning the least load node to merge the backup image chunks (811 ); copying the backup Image into an Image Datastore (813);

informing a Cloud Controller to launch the copied backup image (814); and

updating the backup Information in the Backup Database.

A method as ciaimed in claim 8, wherein said reversing of the backup process (809) comprises decryption or uncompressing of said backup image chunks.