Backup Options — SAP on GCP
Google Cloud has an ongoing partnership with SAP to provide SAP-certified infrastructure for all of your SAP systems. Whether you are responsible for the core, mission-critical SAP applications in a large-scale enterprise or for similarly critical SAP applications in a small enterprise, Google Cloud can help you leverage your SAP data in innovative ways, all while running your SAP applications more reliably, securely, and cost-effectively.
SAP Application System
In SAP application server system binaries and metadata are installed on the “root” file system. During instance creation, the operating system gets installed on base disk. “root” & “/dev” file systems are installed as default directories with the operating system.
SAP Database
Database server has /, /root, /dev directories installed with the operating system. It contains OS binaries and metadata. Directories /usr/sap has SAP specific binaries and log files. Database binaries get installed in separate file systems (/data and /log) during installation. /Shared directory is used during multi-host/instance configuration.
Backup and Recovery can be performed via the following methods on GCP.
- Backup and Recovery using GCP Native tools (i.e. Snapshot, Image)
- Backup and Recovery using Database Native tools
- Backup and Recovery using BACKINT
Backup and Recovery using GCP Native tools
Snapshot
Snapshot can be performed from the Google Cloud Console or using gcloud SDK.
Snapshot can be scheduled immediately or for the future using google scheduler. Snapshot doesn’t require an instance to be down. The recommended approach is to plan snapshot during low read/write load on instances. Disk snapshot can also be taken when it is not attached to any instance. Snapshot protects the disk failover or corruption issues.
Snapshots are incremental and automatically compressed. Snapshots can be created regularly much faster and at a lower cost.
Image
Operating system images are used to create boot disks for compute instances. Google regularly updates the operating system images and release them for users.
For recovering VM from scratch image of the boot disk is mandatory which stores binary and metadata of instances.
For image backup, instance needs to be freezed (shutdown) to make the consistency of files. Image and snapshot backup together perform recovery of the VM point in time or the latest state. Image backup and recovery can also trigger from the GCP console or via Google SDK.
Backup and Recovery using BACKINT
As of now, GCP supports “BACKINT” agent only for the SAP HANA database. With “BACKINT” agent backups can be stored and retrieved directly from Cloud Storage by using the native SAP backup and recovery functions.
To use the “BACKINT” agent for the SAP HANA database on GCP, it should be deployed first on the compute instance where HANA database is running.
To recover a SAP HANA database using BACKINT-based data backup, use <backup_definition_for_recovery> parameter in combination with USING CATALOG BACKINT. “USING CATALOG BACKINT” specifies the location of the backup catalog which is 3rd party tool.
Refer this for more details https://cloud.google.com/solutions/sap/docs/sap-hana-backint-overview.
Backup and Recovery using Third-party tools
Some of the third-party tools for Backup and Recovery are as follows:
- Commvault
- Actifio
- Cohesity
