Unstructured data and its protection
The data center and storage landscape is changing dramatically year-over-year driving organizations to seek more efficient ways to protect and recover unstructured file data, much of which lives on network-attached storage (NAS). There are too many files, too much data, and insufficient backup time (i.e. shorter backup windows).
Effective NAS data protection has always been challenging. This challenge is increasing with time as data volumes grow to and exceed petabyte levels with no end in sight. User home directories and shares, engineering repositories, and application datasets are a few familiar sources of ever-growing mission-critical unstructured data, growing at an annual estimated average of over 34%.
The rapid adoption of 4G and 5G technologies creates high-volume digital content and IoT data. According to an IDC report, an increase in the sale of connected devices could generate about 79.4 zettabytes of data by 2025.
Change file tracking for incremental backups
In today’s world, data is the new oil. The dramatic growth of unstructured data and increased cyber threats make efficient protection paramount. Yet with so much data and new SLAs due to security concerns, customers must now back up this data incrementally (full backups take too long).
With NAS, scanning and finding the files that have changed since the last backup can be a time-consuming and IO-intensive activity that can consume a large proportion of the available IOPS capacity of NAS filers. The scanning process takes proportionally longer as the number of files increases. Even for a minimal change rate, organizations may see a larger incremental backup time. This becomes an issue when the incremental backup time overshoots the backup window.
NAS backup via native API integration
Snapshot-based integration with NAS solutions and vendor-native APIs accelerates backup operations by using point-in-time snapshots and efficient file tracking for incremental backups. The underlying technology, change list tracking between two snapshots via vendor native APIs, is a fast and reliable mechanism. However, these APIs are not always available from all the vendors, and no standardized approach across vendors makes backup and long-term retention of NAS data expensive and complicated — especially as customers adopt hybrid NAS storage solutions.
Faster NAS backup with advanced smart scan
Druva has expanded its cross-platform NAS backup capabilities with a new feature called Advanced Smart Scan. This patent pending feature, currently available for any NAS using SMB protocol, delivers blazing fast incremental backups and eliminates the need for vendor-specific APIs, providing you with the ability to back up, manage and recover data to and from any platform.
Traditional NAS incremental backup requires traversing through all the directories and files in a share configured for backup to track changes. Advanced Smart Scan uses a deterministic approach and performs file system scan very efficiently. It optimizes three aspects of the scan:
- Reduces network round-trips
- Reduces API call latencies
- Reduces pressure on the NAS filer (consumes fewer IOPS)
These aspects combine to dramatically reduce the overall time consumed during a scan, thereby significantly decreasing the incremental backup job time.
Results at scale
To scan 50 million files, we would previously need to make 50 million calls; with Advanced Smart Scan, this is done with 1 million calls only.