Re: diff between incremental and archive backups
- From: Tim Gorman <tim@xxxxxxxxx>
- To: oracle-l@xxxxxxxxxxxxx
- Date: Sat, 14 Nov 2015 23:10:50 -0700
Mladen,
Narrowing the discussion to eliminate topics where we've agreed...
The problem with incremental backups is that they increase the time to
restore. If you need to do backup, and that is not always the case,
then you always need to restore a full backup. So full backup cannot
be avoided. Restoring incremental backups can be avoided. And if the
backups are proxy backups, taken from some kind of a clone (storage
snapshot or replica, standby database), resources are not wasted.
Resources are dedicated for taking backups.
The question is not one of feasibility, but motivation. Of course full
backups can be minimized or completely avoided (i.e. ZDLRA
<
https://www.oracle.com/engineered-systems/zero-data-loss-recovery-appliance/index.html>).
The decision to implement incremental backup strategies is a conscious
tradeoff between the elapsed time and resources expended on backup
versus restore.
Since backup is often the second tier of data protection for strict RTO
requirements, making such a tradeoff is not irrational. And if RTO does
not have strict requirements, so that backups are the primary form of
data protection, then where is the problem with increased time to restore?
Cumulative incremental backups meet halfway on the tradeoff between
backups and restores that you've mentioned, and might well satisfy
RTO requirements while minimizing resources consumed. Maybe so? Maybe
not?
Why take incremental backups at all? You do need to restore them in
addition to the full backup. The only reason why taking all full
backups was so prohibitively expensive was storage consumption. That
is resolved by deduplication.
Deduplication indeed solves the problem of volume of backup media. But,
it does not resolve the problem of resources expended in getting the
data to the backup media. Not everyone wants to take their disk drives
for a brisk gallop on a frequent basis. Not doing so on a frequent
basis eliminates the advantage in MTTR that full backups were supposed
to resolve.
And, even with no incrementals, restore of a multi-TB database is not
a happy occasion, ever. That is why it is the last resort.
Tim, backups are also used for copying databases to development and
test databases. And yes, backups are the solution of the last resort,
I definitely agree with you.
Now you've stepped squarely into *my* area of interest as an employee of
a vendor <
http://www.delphix.com>.
Employing backups to copy production databases and application software
to non-production environments like development and testing is a
holdover from the 20th century, and is fully obsolete. Those continuing
to provision non-production from restored backups or from SAN copies are
placing themselves at a disadvantage to those who have adopted data
virtualization
<
http://www.delphix.com/products/how-delphix-data-virtualization-works/>, plain
and simple.
A year and a half ago, when Delphix offered me a job, my initial
response was, as an independent consultant for 16 years, "/No thanks,
I've already got my dream job/". But when I realized what percentage of
IT departments worldwide needed data virtualization (i.e. 100%), I
changed my mind.
Thanks!
-Tim
Other related posts:
