[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-08-18。"],[[["\u003cp\u003eInstances, which are virtual machines on Google's cloud, store data on persistent disks (PD), and Backup and DR uses snapshots to back up this data incrementally at the instance level.\u003c/p\u003e\n"],["\u003cp\u003eAfter a snapshot is created, the backup image can be used to mount the image to create a new instance, mount the image to a new disk, or restore the image to the source instance.\u003c/p\u003e\n"],["\u003cp\u003eSubsequent snapshots only include new or modified data, referencing previous snapshots for unchanged data, ensuring efficient storage and backups.\u003c/p\u003e\n"],["\u003cp\u003eWhen creating a backup plan, the snapshot storage location can be specified as either a multi-regional or a regional location, affecting availability and networking costs, and it's important to consider organization policies when selecting.\u003c/p\u003e\n"],["\u003cp\u003eSnapshots are useful for creating backups, but differ from custom images and machine images, which are used for creating instance boot disks.\u003c/p\u003e\n"]]],[],null,["# Backup and DR Service for Compute Engine instance backups\n\nAn instance is a virtual machine (VM) hosted on Google's cloud infrastructure.\nEach instance stores data on persistent disks (PD) that are attached to the\ninstance.\n\nBackup and DR uses snapshots to incrementally back up data from your\nPersistent Disks at the instance level. After Backup and DR creates a\nsnapshot to back up the current state of some or all Persistent Disks\nattached to an instance, you can use the backup image to either:\n\n- [Mount the image to create a new instance](/backup-disaster-recovery/docs/access-data/mount-snapshot-images-of-cloud-instances).\n\n- [Mount the image to a new disk attached to the source instance or a different instance](/backup-disaster-recovery/docs/access-data/mount-snapshot-images-of-cloud-instances).\n\n- [Restore the image to the source instance](/backup-disaster-recovery/docs/restore-data/restore-instance).\n\nCompute Engine stores multiple copies of each snapshot across multiple\nlocations with automatic checksums to ensure the integrity of your data.\nTo learn more, see [Snapshots](/compute/docs/disks/snapshots).\n\nNote that snapshots are different from\n[custom images](/compute/docs/images/create-delete-deprecate-private-images) and\n[machine images](/compute/docs/machine-images), which are useful for creating\ninstance boot disks. To learn more, see the\n[table comparing the use of images, snapshots, and instance templates](/compute/docs/machine-images#when-to-use).\n\nHow it works: Backup and DR Persistent Disk snapshots\n-----------------------------------------------------\n\nData backup with Persistent Disk snapshots follows these steps:\n\n1. The first successful snapshot of an instance creates a snapshot of\n each Persistent Disk. For each disk this is a full snapshot that contains\n all the data on the Persistent Disk. Metadata relating to the instance is\n also backed up to a OnVault pool that is automatically created.\n\n | **Note:** Protection of Compute Engine instances using Backup and DR Service does not consume any space in the backup/recovery appliance Snapshot pool.\n2. The second snapshot only contains any new data or modified data since the\n first snapshot. Data that hasn't changed since snapshot 1 isn't included.\n Instead, snapshot 2 contains references to snapshot 1 for any unchanged data.\n\n3. The third snapshot contains any new or changed data since snapshot 2 but won't\n contain any unchanged data from snapshot 1 or 2. Instead, snapshot 3 contains\n references to blocks in snapshot 1 and snapshot 2 for any unchanged data.\n\nThis repeats for all subsequent snapshots of each protected instances Persistent Disk.\nSnapshots are always created based on the last successful snapshot taken.\nIf an additional disk is added to the instance, this disk is automatically\nincluded in the next snapshot of the instance. You can also use include and exclude\nrules to control which persistent disks are included in each backup.\n\nData restore with Persistent Disk snapshots follows these steps:\n\n1. Select the instance and point in time that they want to work with.\n You should take note also of the storage type and regional location.\n\n2. Select if you want to mount to an existing instance, create a\n new instance or restore the disks of the source instance.\n\n3. If creating a new instance, select the location variables such as in\n which project or zone the instance is created.\n\n4. Backup and DR requests Compute Engine to create new disks from the snapshots\n When these disks are created they are attached to the new or existing instance.\n\n| **Note:** If you create application consistent snapshots of Windows Compute Engine instances then on mount to existing VM or mount as new VM, you need to [mark your disk as available for read and write](/compute/docs/instances/windows/creating-windows-persistent-disk-snapshot#read-only) to make them usable.\n\nSnapshot storage location\n-------------------------\n\nWhen you create a backup plan for a Compute Engine instance, you can specify\nwhere the backup is stored. The location of a snapshot affects its\n[availability](/storage/sla) and can incur [networking costs](/storage/pricing#network-pricing)\nwhen creating the snapshot or restoring it to a new disk.\n\nSnapshots can be stored in either one [Cloud Storage multi-regional location](/storage/docs/bucket-locations#location-mr), such as asia, or one [Cloud Storage regional location](/storage/docs/bucket-locations#location-r), such as asia-south1.\n\nA multi-regional storage location provides the highest availability and resilience.\nA regional storage location gives you more control over the physical location of\nyour data because you specify a single region.\n\nA snapshot can be used to create a new instance or disk in any region and zone,\nregardless of the storage location of the snapshot.\n\nIf you have an [organization policy](/resource-manager/docs/organization-policy/overview)\nthat includes the resource locations [constraint](/resource-manager/docs/organization-policy/understanding-constraints), any snapshot storage location that you specify must\nbe in the set of locations defined by the constraint. See\n[Compute Engine resource locations](/resource-manager/docs/organization-policy/defining-locations-supported-services#compute-engine) for more information.\n\nIf you don't specify a storage location for a snapshot, Backup and DR uses\nthe [default location](/compute/docs/disks/snapshots#default_location), which stores\nyour snapshot in a Cloud Storage multi-regional location closest to the region of\nthe source disk.\n\nThe Backup and DR Compute Engine guide\n--------------------------------------\n\n- [Check for the cloud credentials](/backup-disaster-recovery/docs/configuration/create-cloud-credentials)\n\n- [Discover and protect Compute Engine instances](/backup-disaster-recovery/docs/configuration/discover-and-protect-ce-inst)\n\n- [Mount backup images of Compute Engine instances](/backup-disaster-recovery/docs/access-data/mount-snapshot-images-of-cloud-instances)\n\n- [Restore a Compute Engine instance](/backup-disaster-recovery/docs/restore-data/restore-instance)\n\n- [Import Persistent Disk snapshot images](/backup-disaster-recovery/docs/configuration/import-pdsnapshot-images)"]]