|
|
|
General Backup Concepts
|
|
|
|
Data loss is probably the worst problem that can ever happen to IT related individuals
or companies. Hard drives' failure due to bad sectors, OS crashes while accessing
a file, etc are simple examples that might happen anytime causing data loss. Backup
solutions are, then, made for this purpose; that is to be ready to recover anything
important if such problems occur. And to provide some sort of protection against
any unauthorized people from accessing the contents of the backups.
I will try to get you involved with some backup solutions related terms. Techniques
and plans often exist to make the backup procedure more handy and automated, yet
powerful and straight forward.
|
|
|
|
Backup Job
|
|
You probably want to make backups for different files and probably want to categorize
them in such a way that you can execute and/or restore only related objects or files.
So in this case, every set of selections can be saved as one block we call it the
backup job. For example you might want to backup your documents, well in this case
I'll assume that you have personal items and work items at the same time. It is
a good choice to even split the entire backup to 'personal docs backup' and 'work
docs backup'. And update each Backup Set with modifications whenever they are made
on its own. Hence each set of files has its own backup job.
|
|
|
|
|
|
Backup Set and Backup Run
|
|
Each backup job can be executed several times, every single execution is called
a Backup Run. The entire series of backup runs for a single job is called a Backup
Set.
|
|
|
|
Backup Type
|
There are types for backup plans, depending on how often or how the data is going
to be changed. I will try to mention most known backup plans, and give examples
on how to choose the proper backup plan for different occasions.
-
Normal Backup AKA Full Backup: This is a very
basic backup plan. It is simply "all data selected are going to backed up". This
is often the starting point of every other backup plan. Yes, it backs up everything
at the time it is executed. If this type of backup is performed regularly the amount
of storage required to hold the data might be very high depending on how large your
data is.
-
Incremental Backup: This is the kind of backup where only modified
files are backed up. The first run of this backup will include all the selected
data into the backup set. And then comparisons are made on the next runs, and if
a file is not modified, it will not be added to this run. And at the time you restore
the x run, the data are collected from the previous backups to give you the latest
versions of the files that were modified. This is a good choice when you have files
that are going to be modified. To keep versions and to track changes up to a specific
date.
- Mirror Backup: This kind of a plan behaves the same as the incremental
backup plan except that in mirror backups the selected data for a run might differ
from the original selection. Mirror backup plan states that modified files will
be backed up, newly added files will be backed up, and that deleted files will be
excluded from the backup run. If an item was modified, it will be included in the
backup run, if it was newly added it will be backed up too (preference). And finally
if a file was deleted or removed it will be removed from the backup. If it was added
again, it will be considered as a new file, and will be added.
|
|
Rollback
|
|
Rollback is an option for a backup set. Each run the backup gets executed, the latest
revision of an item is stored, if rollback option was enabled, the new run is saved
separated from the original backup. This gives the option for keeping track of the
revisions. Here you have modifications for each run, and you can choose to restore
a any run ignoring any modifications that took place after this run. If this option
is disabled, the history for revisions is not saved, and only the items in the last
run is available for restore. Meaning latest revision of an item replaces older
ones.
|
|
|
|
Data Integrity Check
|
|
Backups are usually stored in compressed format. Usually called 'Archive' or the
'Container'. These archives holds the compressed data, and the information required
to decompress the data. Since archives are usually large there can be sometimes
a problem due to network packet loss or due to media corruption. The Integrity check
in term means to make sure the data is accessible and can be correctly decompressed.
Or in the case if no compression was selected. The exact term is applied, making
sure that data is accessible. Sometimes the data is placed and
connected over multiple mediums, the data integrity check makes sure the entire
data for the selected run is accessible over these connected mediums.
|
|
|
|
The Archive Flag
|
|
Ever noticed the 'Archive' flag in the files' properties dialog? This flag is meant
to show if changes to this specific file was made since the last backup. When a
backup software backs up an item, it lowers the backup flag. And windows or the
'OS' in general, is responsible for raising that flag whenever the file gets modified.
|
|
|
|
Self Executable Backup Archive
|
|
A self executable backup job is an archive that can restore itself. It is usually
a small version of a software that is responsible for only restoring your files.
Meaning if the target machine doesn't have a software that understands the format
of an archive. Then using the self executable backup as the output target of the
backup run will make it easier to restore the files on that target machine without
the need for installing any software for managing the restore.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|