Optimal Backup

Full Backup (Sun 1am) -->Translog hourly -->Diff Backup(Mon 1am). Then Diff
backup every day thru Sat at 1am with hourly trans log backup in between.
Then from the sat diff backup at 1am -->trans log backup-->Sun full backup
at 1am.
Is this optimal?Hi
Being optimal would depend on what you wish to optimize! It is not optimal
in terms of recovery time. If you wanted to optimize that, you would do more
full backups. A full backup each night would remove the need to restore the
differential backup on top of the full backup. Differential backup once or
twice during the day should reduce the need to restore log files since the
full backup. If you want to minimize the risk of loosing data you would
increase the frequency of the log backups.
You should also make sure that you move the backups quickly onto a safe
media, this could be onto a different machine and/or tape or SAN replication
.
There is no points doing backups to a disc that is the same disc for databas
e
data files (which would not perform very well) and then have that disc fail.
John
"mmc" wrote:

> Full Backup (Sun 1am) -->Translog hourly -->Diff Backup(Mon 1am). Then Di
ff
> backup every day thru Sat at 1am with hourly trans log backup in between.
> Then from the sat diff backup at 1am -->trans log backup-->Sun full backu
p
> at 1am.
> Is this optimal?
>|||Thanks
"John Bell" wrote:
[vbcol=seagreen]
> Hi
> Being optimal would depend on what you wish to optimize! It is not optimal
> in terms of recovery time. If you wanted to optimize that, you would do mo
re
> full backups. A full backup each night would remove the need to restore th
e
> differential backup on top of the full backup. Differential backup once or
> twice during the day should reduce the need to restore log files since the
> full backup. If you want to minimize the risk of loosing data you would
> increase the frequency of the log backups.
> You should also make sure that you move the backups quickly onto a safe
> media, this could be onto a different machine and/or tape or SAN replicati
on.
> There is no points doing backups to a disc that is the same disc for datab
ase
> data files (which would not perform very well) and then have that disc fai
l.
> John
> "mmc" wrote:
>|||Hi
http://vyaskn.tripod.com/ sql_serve...r />
.htm#Step1
--administaiting best practices
"morphius" <morphius@.discussions.microsoft.com> wrote in message
news:E58E31C3-0800-4DB7-ABA1-80AEDF05C222@.microsoft.com...[vbcol=seagreen]
> Thanks
> "John Bell" wrote:
>