Understanding Data Backup Techniques: File‑Level, Block‑Level, Remote Copy, Snapshots and Volume Clone
This article explains the fundamentals and classifications of data backup technologies—including file‑level and block‑level protection, remote file copy, remote volume imaging, snapshot mechanisms, CoFW vs RoFW strategies, and volume clone methods—while also covering backup destinations, paths, and common backup strategies.
File‑Level Backup
File‑level backup reads files through the file‑system interface and stores them on another medium, preserving only the file data without metadata because the new medium can reconstruct the file system.
Block‑Level Backup
Block‑level backup copies every block of a device regardless of the file system, offering higher speed but also copying unused "zombie" blocks and potentially creating fragmentation.
Remote File Copy
Remote file copy (e.g., rsync) transfers changed files to a disaster‑recovery site asynchronously, providing incremental replication.
Remote Volume Imaging
Remote volume imaging backs up block data to a remote site and can be performed synchronously or asynchronously.
Synchronous copy: waits for data to reach the remote site before reporting success.
Asynchronous copy: reports success immediately, then transfers data, sacrificing consistency for speed.
Snapshot Technology
Snapshots capture the state of a volume at a specific point in time. They require the original volume to be quiesced or its writes recorded in a bitmap (BitMap) so that the snapshot remains consistent.
File‑System Snapshots
File‑system snapshots rely on metadata such as inode tables, block bitmaps, and B‑trees. The snapshot saves the mapping tables at a moment, allowing the system to restore the exact state without interrupting I/O.
Physical‑Volume Snapshots
Physical‑volume snapshots copy only the LBA mapping information, which is simpler but still needs CoFW or RoFW handling for subsequent writes.
CoFW vs RoFW
Copy‑on‑First‑Write (CoFW) copies the original block before overwriting, incurring one read and two writes for the first overwrite; Redirect‑on‑First‑Write (RoFW) redirects the write to a new location with a single write, reducing I/O but increasing CPU and bitmap lookup overhead.
Snapshot Significance
Snapshots enable online, low‑impact backups, protect against logical data loss, shorten backup windows, and simplify consistency checks after restoration.
Volume Clone Technology
A clone is a writable snapshot; a virtual clone shares unchanged blocks with the source, while a split clone creates an independent copy.
Backup Destinations
Local disk
SAN storage
NAS directory
Virtual tape library
Backup Paths
Data can flow through local buses, front‑end networks, back‑end SAN/HBA connections, or use LAN‑free and Server‑free techniques that bypass the host CPU and memory.
Backup Strategies
The backup engine (software running on a backup server) defines what, when, and how to back up, communicating with agents on protected hosts. Media servers manage shared devices such as tape libraries, coordinated by a central backup server.
Backup Types
Full backup: copies all selected data.
Differential backup: copies data changed since the last full backup.
Incremental backup: copies data changed since the last backup of any type.
For databases, third‑party backup tools must invoke the database’s own backup commands because internal file changes are not externally visible.
Architects' Tech Alliance
Sharing project experiences, insights into cutting-edge architectures, focusing on cloud computing, microservices, big data, hyper-convergence, storage, data protection, artificial intelligence, industry practices and solutions.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.