Shulou(Shulou.com)06/02 Report--

You can deploy System Center Data Protection Manager (DPM) for

Application-aware backups: application-aware backups for Microsoft workloads, including SQL Server, Exchange, and SharePoint.

File backup: back up files, folders, and volumes on computers running Windows Server and Windows client operating systems.

System backup: back up the system state or completely naked the physical computer running the Windows Server or Windows client operating system

Machine backup.

Hyper-V backup: back up a Hyper-V virtual machine (VM) running Windows or Linux. You can back up the entire VM or run

Application-aware backups that run Microsoft workloads on Windows's Hyper-V VM.

DPM stores backup data to

Disk: for short-term storage, DPM backs up the data to the disk pool.

Azure: for short-term and long-term external storage, Azure backup service can be used to back up DPM data stored in disk pools to Microsoft

Azure Cloud.

Tape: for long-term storage, data can be backed up to tape, which can then be stored off-site.

DPM backup overall architecture diagram

Disk-based protection process

The method used by System Center Data Protection Manager (DPM) to protect data is based on the type of protected data and the protection method chosen

It's different. To provide disk-based data protection, the DPM server creates and maintains a copy of the data on the protected server. This copy is stored in a storage pool, which consists of a set of disks located on a DPM server or custom volume. The following figure shows the basic relationship between a protected volume and its copy.

Copy of the data source

Whether you are protecting file data or application data, you must first create a copy of the data source. Copies are synchronized or updated periodically according to the configured settings. The method DPM uses to synchronize replicas depends on the type of data being protected. If it is determined that the replica is in an inconsistent state, DPM performs a consistency check, which verifies the replica block by block against the data source.

Protection agent

The protection agent tracks changes to protected data and transfers the changes to the DPM server. The protection agent also identifies data on the computer that can be protected and involved in the recovery process. Protection agents must be installed on each computer that you want to protect by using DPM. The protection agent can be installed by DPM or manually by using an application such as Systems Management Server (SMS).

Protection group

Protection groups are used to manage the protection of data sources on your computer. A protection group is a set of data sources that share the same protection configuration. A protection configuration is a set of settings common to a protection group, such as protection group name, protection policy, disk allocation, and replica creation method.

File data synchronization process

In DPM, for file volumes or shares on the server, the protection agent uses volume filters and change logs to determine which files have changed, and then performs a checksum process on those files to synchronize only the changed blocks. During synchronization, these changes are transferred to the DPM server and then applied to the replica to synchronize the replica with the data source. The following figure illustrates the file synchronization process.

If the copy becomes inconsistent with its data source, DPM generates an alert indicating which computers and which data sources are affected. To resolve this problem, the administrator fixes the replica by initiating a synchronization on the replica that includes a consistency check (also referred to as a consistency check for short). During the consistency check, DPM checks the copy block by block and repairs the copy to make it consistent with the data source again.

You can schedule a daily consistency check for the protection group, or you can manually initiate a consistency check.

DPM will create recovery points for members of the protection group on a regular basis (specific time is up to you). The recovery point is the version of the data from which the data can be recovered.

Application data synchronization process

For application data, after DPM creates a copy, the volume filter tracks changes to volume blocks that are part of the application file. How these changes are transferred to the DPM server depends on the type of application and synchronization. In the DPM administrator console, the operation marked synchronization is similar to an incremental backup and, when used in conjunction with a replica, creates an accurate image of the application data.

When you perform a synchronization type marked as Quick full backup in the DPM Administrator console, a full Volume Shadow copy Service (VSS) snapshot is created, but only the changed blocks are transferred to the DPM server.

Each quick full backup creates a recovery point for the application data. If your application supports incremental backups, each synchronization also creates a recovery point. The following outlines the types of synchronization supported by each application data:

For protected Exchange data, synchronization uses the Exchange VSS writer to transfer incremental VSS snapshots. Each synchronization and quick full backup creates a recovery point.

SQL Server databases where logs are delivered, are in read-only mode, or use a simple recovery model do not support incremental backups. Restore points are created only on each quick full backup. For all other SQL Server databases, synchronization transfers transaction log backups, and each incremental synchronization and fast full backup creates a recovery point. The transaction log is a continuous record of all transactions performed on the database since the last backup of the transaction log.

Windows SharePoint Services and Microsoft Virtual Server do not support incremental backups. Restore points are created only on each quick full backup.

Advantages of DPM's fast full backup (compared to incremental backup)

Incremental synchronization takes less time than a fast full backup. However, as the number of synchronizations increases, so does the time it takes to recover data. This is because DPM must restore the last full backup and then restore and apply all incremental synchronizations until the selected restore point in time is reached. To enable faster recovery time, DPM periodically performs fast full backups, which are synchronized and update copies to include changed blocks.

During a fast full backup, DPM takes a snapshot of the copy and then updates the copy with the changed blocks. "to achieve the goal of creating recovery points more frequently, and to reduce the period of data loss, DPM also performs incremental synchronization between two fast full backups."

Differences between file data and application data

The data that exists on the file server and needs to be protected as a flat file is called file data, such as Microsoft Office files, text files, batch files, and so on.

Data that exists on the application server and requires DPM to know about this application is called application data, such as Exchange storage groups, SQL Server databases, Windows SharePoint Services farms, and virtual servers.

Maximum number of recovery points and incremental backups supported by DPM

DPM can store up to 64 recovery points for each file member of the protection group. For application data sources, DPM can store up to 448 fast full backups and up to 96 incremental backups for each fast full backup. The protection job will fail when the storage area limit is reached but the retention period of the existing recovery point has not been reached.

The difference between file recovery point and application recovery point

The process of creating a recovery point for file data is different from that of creating a recovery point for application data. DPM can create recovery points for file data by using a shadow copy of the copy on the configured schedule. For application data, each synchronization and quick full backup creates a recovery point.

Recovery goal

The recovery goal quantifies the organization's data protection requirements. In DPM, recovery goals are defined jointly by retention period, data loss fault tolerance, recovery point schedule, and fast full backup schedule (for database applications).

Retention period

The retention period refers to how long the data you need to back up remains available. For example, do you want today's data to be available a week from now? Is it available two weeks from now? Is it available one year from now?

Data loss fault tolerance

Is the most acceptable big data loss in time for business requirements, and determines how often DPM should synchronize with protected servers by collecting data changes from the protected server. You can change the synchronization frequency to any period of time between 15 minutes and 24 hours. You can also choose to synchronize just before the recovery point is created, rather than according to a specified time schedule.

Recovery point plan

Determine how many recovery points for this protection group should be created. For file protection, you can select the date and time when you want to create the recovery point. For application data protection that supports incremental backups, the synchronization frequency determines the recovery point schedule. For application data protection that does not support incremental backups, the schedule for fast full backups determines the recovery point schedule.

The automatic Discovery process of AD computer added by DPM

Auto-discovery refers to the daily process that DPM uses to automatically detect new or deleted computers in the network. Once a day, DPM sends a small packet (less than 10 KB) to the nearest domain controller at a time you can plan. This domain controller responds to LDAP requests with computers in the domain, and DPM recognizes new and deleted computers. The automatic discovery process generates very little network traffic.

Autodiscovery does not discover new and deleted computers in other domains. To install a protection agent on a computer in another domain, you must use the fully qualified domain name of that computer to identify it.

DPM directory structure

When you start using DPM to protect your data, you will notice that the installation path of DPM contains three folders in the Volumes directory:

\ Microsoft DPM\ DPM\ Volumes\ DiffArea

\ Microsoft DPM\ DPM\ Volumes\ Replica

\ Microsoft DPM\ DPM\ Volumes\ ShadowCopy

The DiffArea folder contains mounted shadow copy volumes that store recovery points for the data source.

The Replica folder contains mounted copy volumes.

The ShadowCopy folder contains a local backup copy of the DPM database. In addition, when you use DPMBackup.exe to create backup shadow copies of copies for archiving by third-party backup software, these backup shadow copies are stored in the ShadowCopy folder.

DPM backup support Matrix (details what can be backed up)

Https://docs.microsoft.com/zh-cn/system-center/dpm/dpm-protection-matrix?view=sc-dpm-1807

Cluster support

DPM protects data in the following clustered applications:

File server

SQL Server

Hyper-V-however, if you use scale-out DPM protection to protect Hyper-V clusters, you cannot add secondary protection for protected Hyper-V workloads.

Exchange Server-for supported versions of Exchange Server, DPM protects clusters of non-shared disks (cluster continuous replication) and also protects Exchange Server configured for local continuous replication.

SQL Server? note that DPM does not support the protection of SQL Server databases hosted on cluster shared volumes (CSV).

DPM protects cluster workloads that are in the same domain as the DPM server and in subdomains or trusted domains. If you want to protect data sources in untrusted domains or workgroups, you need to use NTLM or certificate authentication for a single server, or certificate authentication only for the cluster.

For more study notes, please refer to the official Wechat account.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.