단일 데이터 소스에서 여러 디스크를 만들려면 스냅샷 대신 이미지를 사용하세요. Compute Engine은 대상 영역에서 로컬 캐싱을 수행하므로 이미지에서 디스크 만들기가 스냅샷에서 디스크 만들기보다 빠릅니다.
머신 이미지를 사용하여 인스턴스에 연결된 모든 디스크의 백업 만들기
인스턴스에 연결된 모든 디스크의 백업을 만들려면 머신 이미지를 사용합니다.
머신 이미지를 사용하면 한 번에 여러 디스크를 백업하여 머신 이미지에 캡처된 데이터가 모든 디스크에서 일관성을 유지하도록 할 수 있습니다. 영구 디스크 스냅샷은 한 번에 하나의 디스크만 백업할 수 있습니다. 자세한 내용은 머신 이미지를 사용하는 경우를 참조하세요.
Google Cloud 백업 및 DR 서비스를 사용하여 대규모로 인스턴스 백업 관리
백업 및 DR 서비스를 사용하면 다음과 같은 방식으로 고급 정책, 중앙 집중식 모니터링 및 백업 보고를 사용하여 프로젝트와 환경 전반에서 인스턴스 백업을 관리할 수 있습니다.
[[["이해하기 쉬움","easyToUnderstand","thumb-up"],["문제가 해결됨","solvedMyProblem","thumb-up"],["기타","otherUp","thumb-up"]],[["이해하기 어려움","hardToUnderstand","thumb-down"],["잘못된 정보 또는 샘플 코드","incorrectInformationOrSampleCode","thumb-down"],["필요한 정보/샘플이 없음","missingTheInformationSamplesINeed","thumb-down"],["번역 문제","translationIssue","thumb-down"],["기타","otherDown","thumb-down"]],["최종 업데이트: 2025-07-29(UTC)"],[[["\u003cp\u003eCompute Engine offers multiple backup and replication options for Persistent Disk, Hyperdisk volumes, VM instances, and workloads, including on-premises.\u003c/p\u003e\n"],["\u003cp\u003eGoogle recommends using one of the backup solutions discussed, as deleted compute instances, disks, images, or snapshots are unrecoverable, even if accidentally deleted.\u003c/p\u003e\n"],["\u003cp\u003eBackup and DR Service is a managed option for secure storage and centralized backup management across multiple projects, providing features like backup vaults, backup plans, and security threat event insights.\u003c/p\u003e\n"],["\u003cp\u003eStandard, Archive, and Instant snapshots offer varying levels of cost and access speed for long-term data backup, ranging from geo-redundant storage to rapid data restoration.\u003c/p\u003e\n"],["\u003cp\u003eMachine images offer a method to back up and replicate all the disks attached to an instance with consistency across the disks, whereas persistent disk snapshots can only back up a single disk at a time.\u003c/p\u003e\n"]]],[],null,["# Data protection options\n\n*** ** * ** ***\n\nThis document compares the options available on Compute Engine to\nback up and replicate the following Compute Engine resources:\n\n- Persistent Disk and Google Cloud Hyperdisk volumes\n- Virtual machine (VM) and bare metal instances\n- Workloads running in Compute Engine and on-premises\n\nTo safeguard important data, Google recommends using one of the backup options\ndiscussed in this document. You can't recover a compute instance, disk, image,\nor snapshot if you delete it, even if the deletion was accidental.\n\nOptions\n-------\n\nBest practices for backups\n--------------------------\n\nObserve the following best practices when creating backups for your disks.\n\n### Avoid temporary standard snapshots\n\nTo immediately create a copy of a disk in the same zone for verification or\nexport, use disk clones or instant snapshots instead of standard snapshots.\nCompared to disk clones and instant snapshots, standard snapshots\nhave longer copy times for upload and download.\n\n### Schedule hourly standard snapshots for backup and disaster recovery\n\n[Schedule hourly standard snapshots](/compute/docs/disks/scheduled-snapshots).\nIf you require daily snapshots, consider scheduling snapshots every 6 hours.\n\n### Use images for fast and frequent disk creation across regions\n\nTo create many disks from a single data source,\n[use images](/compute/docs/instances/create-start-instance#startinginstancewithimage)\ninstead of snapshots. Because Compute Engine performs local caching in\ntarget zones, disk creation from images is faster than disk creation from\nsnapshots.\n\n### Use machine images to create backups of all disks attached to an instance\n\nTo create backups of all disks that are attached to an instance,\n[use machine images](/compute/docs/machine-images/create-machine-images#create-image-from-instance).\nA machine image can be used to backup multiple disks at a time to help ensure\nthat the data captured in the machine image is consistent across all disks. A\npersistent disk snapshot can only backup a single disk at a time. For more\ninformation, see [When to use machine images](/compute/docs/machine-images#when-to-use).\n\n### Use Google Cloud Backup and DR Service to manage instance backups at scale\n\nWith Backup and DR Service, you manage backups of your instances across projects and\nenvironments by using advanced policies, centralized monitoring and backup\nreporting in the following way:\n\n- [Create backup vaults](/backup-disaster-recovery/docs/cloud-console/backup-vault-create) that serve as secure storage locations for your backups.\n- [Create backup plans](/backup-disaster-recovery/docs/cloud-console/backup-plan-create) to configure the schedule and rules for your backup.\n- Apply backup plans [to existing instances](/backup-disaster-recovery/docs/cloud-console/compute/compute-instance-backup) or [during instance creation](/compute/docs/instances/create-instance-with-gcbdr-backup-plan).\n- Assign specific permissions for [backup access](/backup-disaster-recovery/docs/configuration/workforce-identity-federation).\n- Proactively and reactively [monitor backup jobs](/backup-disaster-recovery/docs/monitor-reports/monitor-jobs-console).\n- Create [reports for tracking backups](/backup-disaster-recovery/docs/monitor-reports/reports-overview) across resources spanning multiple projects.\n- When needed, [restore an instance from a backup vault](/backup-disaster-recovery/docs/cloud-console/compute/compute-instance-restore).\n\nYou can also integrate\n[Security Command Center](/security-command-center/docs/security-command-center-overview)\nwith Backup and DR Service to additionally:\n\n- Track audit logs for backup access.\n- Monitor for malicious activity on your backups.\n\nWhat's next\n-----------\n\n- Read [Best practices for persistent disk snapshots](/compute/docs/disks/snapshot-best-practices)."]]
