Backup for SAP HANA | Guided by A Guide to Cloud

Protecting HANA data

🔧 Aisha opens the backup console. “Monitoring tells us when something goes wrong. Backup is our insurance policy for when things go very wrong — data corruption, accidental deletion, ransomware. At TradeCorp, Carlos wants daily backups with the ability to restore to any point in the last 30 days.”

Carlos nods. “Our auditors require proof that we can recover data. Not just that we take backups, but that we can actually restore them.”

Simple explanation

Think of HANA backup like saving your progress in a video game.

A full backup is a complete save file — everything captured at that moment. A differential backup saves only what changed since the last full save (smaller, faster). An incremental backup saves only what changed since the last backup of any kind (smallest, fastest). Log backups capture every single action as it happens — so you can replay from a save file to any exact moment. Together, they let you “rewind” to any point in time.

Backup types for HANA

HANA backup types
Type	What it captures	Size	Frequency	Restore speed
Full	Complete database copy	Largest	Weekly (typically Sunday)	Fastest (single restore)
Differential	Changes since last full backup	Medium	Daily	Moderate (full + differential)
Incremental	Changes since last backup of any type	Smallest	Daily or more frequent	Slower (full + chain of incrementals)
Log	Transaction log entries	Continuous stream	Every 15 minutes (configurable)	Used for PITR — replay to exact second

Azure Backup with Backint

Backint is the SAP-certified backup interface. Azure Backup implements Backint so HANA streams backup data directly to the Recovery Services vault:

No local staging disk needed — data flows directly from HANA to the vault
Application-consistent — HANA ensures data integrity during backup
Supports all backup types (full, differential, incremental, log)
Managed from the Azure portal or HANA Studio/cockpit
Pre-registration script configures the HANA system for Azure Backup integration

HANA native backup to Azure Storage

Alternatively, HANA can write native file-based backups to a mounted filesystem that stores data on ANF volumes or Azure Blob-backed NFS:

HANA writes backup files to a mounted NFS file system (HANA does not write directly to Azure Blob Storage — it writes to a POSIX filesystem)
The mounted file system can be backed by ANF volumes or NFS-enabled Azure Blob containers
More manual management than Backint (retention, cleanup)
Useful for specific scenarios like very large databases where Backint streaming is slow

Exam tip: Backint is the Azure-integrated answer

When the exam asks about HANA backup on Azure, the preferred answer is Azure Backup with Backint. It is the fully integrated, SAP-certified approach with centralized management, policy-based retention, and geo-redundant storage. File-based backup to Azure Storage is valid but requires more manual management. If the question mentions “integrated backup management” or “centralized policy,” Backint is the answer.

Backup policies

A well-designed HANA backup policy includes:

Full backup: Weekly (e.g., Sunday at 2 AM)
Differential or incremental: Daily (e.g., daily at 2 AM, except Sunday)
Log backup: Every 15 minutes (HANA default, configurable down to 15 minutes)
Retention: 30 days for daily, 12 weeks for weekly, 12 months for monthly, 10 years for annual (adjust per compliance)

Geo-redundant vaults

Recovery Services vaults can be configured for redundancy:

LRS (Locally Redundant) — 3 copies within a single datacenter. Cheapest.
ZRS (Zone Redundant) — copies across availability zones. Better protection.
GRS (Geo-Redundant) — copies to the paired region. Required for DR backup strategy.
RA-GRS — GRS with read access from the secondary region. Enables cross-region restore.

For SAP, GRS is recommended for production systems to enable backup-based DR.

Restore scenarios

🔧 Aisha runs through the restore options. “Azure Backup gives us four ways to restore HANA:”

Point-In-Time Recovery (PITR): Restore to any specific second within the retention window. Uses the most recent full + differential/incremental + log replay to the target time. Most common for data corruption or accidental deletion.

Restore to a specific backup: Restore a specific full or differential backup without log replay. Faster but less precise than PITR.

Alternate location restore: Restore to a different HANA instance (different VM, different SID). Useful for creating test copies or migrating data.

Cross-region restore: Restore from a GRS vault in the paired region. Used during regional DR when the primary region is unavailable.

ANF snapshot-based backup

For HANA on ANF storage, application-consistent snapshots provide near-instant backup:

ANF snapshots are taken at the volume level
HANA is briefly frozen (quiesced) during the snapshot for consistency
Snapshots are space-efficient (copy-on-write)
Restore from snapshot is near-instant (no data streaming)
ANF cross-region replication can copy snapshots to the DR region

Snapshots complement Backint

ANF snapshots provide fast backup and restore for operational recovery (accidental deletion, quick rollback). Backint provides long-term retention with policy management and cross-region protection. A comprehensive HANA backup strategy often uses both: ANF snapshots for fast operational recovery and Backint for compliance-driven long-term retention.

Question

What is Backint and how does Azure Backup use it?