Snapshot (computer storage)
|This article needs additional citations for verification. (September 2013)|
In computer systems, a snapshot is the state of a system at a particular point in time. The term was coined as an analogy to that in photography. It can refer to an actual copy of the state of a system or to a capability provided by certain systems.
A full backup of a large data set may take a long time to complete. On multi-tasking or multi-user systems, there may be writes to that data while it is being backed up. This prevents the backup from being atomic and introduces a version skew that may result in data corruption. For example, if a user moves a file into a directory that has already been backed up, then that file would be completely missing on the backup media, since the backup operation had already taken place before the addition of the file. Version skew may also cause corruption with files which change their size or contents underfoot while being read.
One approach to safely backing up live data is to temporarily disable write access to data during the backup, either by stopping the accessing applications or by using the locking API provided by the operating system to enforce exclusive read access. This is tolerable for low-availability systems (on desktop computers and small workgroup servers, on which regular downtime is acceptable). High-availability 24/7 systems, however, cannot bear service stoppages.
To avoid downtime, high-availability systems may instead perform the backup on a snapshot—a read-only copy of the data set frozen at a point in time—and allow applications to continue writing to their data. Most snapshot implementations are efficient and can create snapshots in O(1). In other words, the time and I/O needed to create the snapshot does not increase with the size of the data set; by contrast, the time and I/O required for a direct backup is proportional to the size of the data set. In some systems once the initial snapshot is taken of a data set, subsequent snapshots copy the changed data only, and use a system of pointers to reference the initial snapshot. This method of pointer-based snapshots consumes less disk capacity than if the data set was repeatedly cloned.
Read-write snapshots are sometimes called branching snapshots, because they implicitly create diverging versions of their data. Aside from backups and data recovery, read-write snapshots are frequently used in virtualization, sandboxing and virtual hosting setups because of their usefulness in managing changes to large sets of files.
Some Unix systems have snapshot-capable logical volume managers. These implement copy-on-write on entire block devices by copying changed blocks—just before they are to be overwritten within "parent" volumes—to other storage, thus preserving a self-consistent past image of the block device. Filesystems on such snapshot images can later be mounted as if they were on a read-only media.
Some volume managers also allow creation of writable snapshots, extending the copy-on-write approach by disassociating any blocks modified within the snapshot from their "parent" blocks in the original volume. Such a scheme could be also described as performing additional copy-on-write operations triggered by the writes to snapshots.
Some file systems, such as WAFL,[note 1] fossil for Plan 9 from Bell Labs, and ODS-5, internally track old versions of files and make snapshots available through a special namespace. Others, like UFS2, provide an operating system API for accessing file histories. In NTFS, access to snapshots is provided by the Volume Shadow-copying Service (VSS) in Windows XP and Windows Server 2003 and Shadow Copy in Windows Vista. Melio FS provides snapshots via the same VSS interface for shared storage. Snapshots have also been available in the NSS (Novell Storage Services) file system on NetWare since version 4.11, and more recently on Linux platforms in the Open Enterprise Server product.
EMC's Isilon OneFS clustered storage platform implements a single scalable file system that supports read-only snapshots at the file or directory level. Any file or directory within the file system can be snapshotted and the system will implement a copy-on-write or point-in-time snapshot dynamically based on which method is determined to be optimal for the system.
On Linux, the Btrfs and OCFS2 file systems support creating snapshots (cloning) of individual files. Additionally, Btrfs also supports the creation of snapshots of subvolumes. On AIX, JFS2 also support snapshots.
Time Machine, included in Apple's Mac OS X v10.5 operating system, is not a snapshotting scheme but a system-level incremental backup service: it merely watches mounted volumes for changes and copies changed files periodically to a specially-designated volume using hard links.
The SQL specification mandates four levels of transaction isolation. In the highest, SERIALIZABLE, a snapshot is implicitly created at the start of every transaction. The backup utilities for many popular SQL databases use this feature to generate self-consistent dumps of table data.
A database snapshot provides a read-only, static view of a source database as it existed at snapshot creation, minus any uncommitted transactions. Uncommitted transactions are rolled back in a newly created database snapshot because the Database Engine runs recovery after the snapshot has been created (transactions in the database are not affected).
Database snapshots are dependent on the source database. The snapshots of a database must be on the same server instance as the database. Furthermore, if that database becomes unavailable for any reason, all of its database snapshots also become unavailable.
Snapshots can be used for reporting purposes. Also, in the event of a user error on a source database, you can revert the source database to the state it was in when the snapshot was created. Data loss is confined to updates to the database since the snapshot's creation. Also, creating a database snapshot can be useful immediately before making a major change to a database, such as changing the schema or the structure of a table. For more information on the uses of snapshots, see Typical Uses of Database Snapshots.
Understanding how snapshots work is helpful though not essential to using them. Database snapshots operate at the data-page level. Before a page of the source database is modified for the first time, the original page is copied from the source database to the snapshot. This process is called a copy-on-write operation. The snapshot stores the original page, preserving the data records as they existed when the snapshot was created. Subsequent updates to records in a modified page do not affect the contents of the snapshot. The same process is repeated for every page that is being modified for the first time. In this way, the snapshot preserves the original pages for all data records that have ever been modified since the snapshot was taken.
To store the copied original pages, the snapshot uses one or more sparse files. Initially, a sparse file is an essentially empty file that contains no user data and has not yet been allocated disk space for user data. As more and more pages are updated in the source database, the size of the file grows. When a snapshot is taken, the sparse file takes up little disk space. As the database is updated over time, however, a sparse file can grow into a very large file.
Virtualization environments host a guest operating system inside a virtual machine; some of them (including VMware, XenServer, VirtualBox, Parallels Desktop, QEMU and Virtual PC) can perform whole-system snapshots by dumping the entire machine state to a backing file, and redirecting future guest writes to a second file, which then acts as a copy-on-write table.
Software transactional memory is a scheme which applies the same concepts to data structures held only in memory.
- Application checkpointing
- System image
- LVM snapshots (Linux)
- R1Soft Hot Copy (Linux)
- Microsoft Volume Shadow Copy
- Storage Hypervisor
- WAFL is not a file system. WAFL is a file layout that provides mechanisms that enable a variety of file systems and technologies that want to access disk blocks.
- "LVM HOWTO". 3.8. Snapshots. tldp.org. Retrieved 2013-09-29.
- "Optimized Storage Solution for Enterprise Scale Hyper-V Deployments". Microsoft. March 2010. p. 15. Retrieved 25 October 2012.
- Garimella, Neeta (2006-04-26). "Understanding and exploiting snapshot technology for data protection, Part 1: Snapshot technology overview".
- Harwood, Mike (2003-09-24). "Storage Basics: Backup Strategies".
- Project web page of rsnapshot