Advantages of Data Backup
The advantages of data backup are many:
- Minimize the risk of loss and have a distance protection tackling problems of theft, fire, flood, etc.;
- Storage spaces are not limited in size;
- Completely free;
- A fully automated backup and periodic;
The major drawback of this technique is that it is open only to individuals but do not meet the needs of companies and it is important to ensure that data is fully encrypted to make it unreadable on other storage spaces. This technique must register within trusted group.
The most common methods (Types) of backup are:
The simplest method is the full backup or full (also called “full backup”); it is to copy all data to back that they are recent, old, modified or not.
This method is also more reliable but it is long and very costly in terms of disk space, making it difficult to use in practice for all backups to be performed. To gain speed and time saving, there are methods that perform the backup of only modified data and / or added between two full backups. We identified two:
- Differential backup
- The incremental backup
Restoring a disk with one of these methods is longer and more tedious because in addition to restoring the differential backup or incremental backups, you must also restore the last full backup. Files deleted in the meantime will also be restored.
To understand the difference between the two methods, we take the example of a backup plan as shown below:
- A full backup on D-Day (Sunday night for example)
- A backup of changed files or new D-Day +1 to Day 6 (Monday evening to Saturday evening included)
- A full backup on D-Day 7 (Sunday night following)
Data Backup Mechanism
To differentiate these different methods of backup / archiving (full, incremental, differential), the mechanism is implemented using a marker for archiving. Each file has a marker for archiving, which is set to “true” when creating or modifying a file. One can understand this position as “I have just been changed or created, I am ready to be archived, so I position my marker to true. This marker is also called archive attribute (or archive bit). On Windows, this attribute is editable and can be viewed by the ATTRIB command (attribute A for archive). The backup system may also be a database containing the definitions files and use an internal marking.
Backup – complete
During a full backup, it will reset to “0″ attribute of the file to store the fact that the file was saved. When working with date, it stores the date of the last backup so you can differentiate the files that were backed others (date of last modification).
Backup – Technicality
During a full backup, all files are saved, regardless of the position of the marker (true or false). Once the archive file, it is assigned the position of the marker (or bit) to “false” (or “0″).
The differential backup makes a copy of the files created or modified since the last full backup, regardless of the intermediate backups. In other words, the full backup of D-Day serves as a reference to identify files created, modified or added, and thus only save them from Day 1 to Day 6.
The restoration made from this type of backup disk mirroring requires the last full backup and differential backup most recent.
With our example, if the restoration is on a full disk that was backed up on Day 2, then it must be copied to disk complete backup of D-Day and differential backup of Day 2 to get the latest version of data.
But when it comes to restoring a file or folder that was saved on D-Day 2 only the last backup, the differential here is useful.
Data Backup – Technical Details
When a differential backup, all files whose tag is “true” are saved. Once the archive file, it retains the position of his marker as he had before the backup.
Incremental backup or incremental
This method involves backing up files created or modified since the last backup regardless of its type (full, incremental or differential).
Example: A full backup is performed on D-Day D +1, incremental backup is performed by reference to D-Day, D-Day +2; the incremental backup is performed by reference to a D-Day. And so on.
If the restoration is on a full disk that was backed up on D-Day 4, you must copy the backup to disk of D-Day and incremental backups of days D + 1, D + 2, D + 3 and D + 4 to obtain the latest version of all data. But when it comes to restoring a file or folder that was saved on D-Day 3, only the last backup, the incremental here, is useful.
Data Backup – Technical Details
In an incremental backup, all files whose tag is “true” are saved. Once the archive file, it is assigned the position of his marker to “false”.
Backup, archiving and retention
The retention can be the difference between backup and archiving.
Retention is the time during which the stored data is kept intact. A short work retention is considered a classic work of backup: the data is protected against the loss / alteration. Work retention time (one or more years) is considered a work of archiving and will aim to recover the data to a specific date, express request.
Example: retention of four weeks implies that the instance of data stored on a specific date will still be available until 28 days after backup. After these 28 days, a logical point of view, the data do not exist in the backup system and are considered not found. Physically, the tracks used to store this backup can be deleted.
More retention is longer and the number of pending saved for the same object file or folder is important. More retention is longer and tends to a backup mechanism that requires an archive search system and appropriate indexing. More retention is longer and the space required to store the backup jobs will be important.
Formula for calculation of backup space needed
This formula allows to resize a library of backup (tape or disk VTL).
In the case of backups, that is to say full backup on weekends (Friday night) and incremental backups other days of the week, Monday to Thursday (not Friday) or four days:
- D is the space of useful data to back up,
- R is retention of desired work, expressed in weeks,
- T is the rate of change per day files to save space, the following formula is obtained: D x R + (D x T%) 4 = x storage capacity.
Example: 100 GB total backup with retention of three weeks and change rate of 20% gives 100 per day x 3 + (100 x 20%) x 4 = 380 GB 380 GB will be necessary to safeguard our 100 GB of data with a retention of three weeks, and a change of 20% per day.
Technological innovations such as snapshots and de-duplication can reduce the value of a very interesting way.
Backup – Additional technical information
The backup data can be performed using techniques more or less sophisticated. The easiest way is to browse directories and files from a workstation or server, but it is soon limited by the number of files and volumes of data, which have a direct impact on the Backup time. To circumvent these limitations, several approaches are possible:
- Stored data compression, used by most backup solutions
- Technical Snapshot: taking snapshot of a disk, especially in a SAN (see Logical Volume Management)
- Backup block mode (NDMP for NAS in particular)
- De-duplication technology to reduce the volume of backups by eliminating duplication
- A combination of these techniques.
Study: From Wikipedia, the free encyclopedia. The text is available under the Creative Commons.