Moving off Data Domain Primary Storage

[chrisBrindley]

I have decided after 6 months to redesign our backup infrastructure, the Data domain we have which is the DD860 is a great storage device for long term retention, this is what it needs to be sold as EMC will drown you with Dedupe numbers, and they are right, we have two 20TB shelfs holding almost 360TB of Info, but we cannot use it our environment, we are trying to offer our customer self restores and the rehydration of data sometimes takes 1 - 2 hours to complete, its just painfully slow.

My question is.
We have several other storage units that are not so great on dedupe, bust very fast on rehydration, is there away to setup a job to have a 7 day retention, but keep 2 days on the faster storage and then keep 5 days on the Data domain, backup copies only copy the data from the original backup so its not true retention just copied backup data.

[foggy]

Chris, there's no way to set a single job to write to two different repositories, so using backup copy jobs seems to be a proper alternative to that.

[chrisBrindley]

Foggy, i am putting this on paper to see if i can make sense of the scenario.
7 Day retention is needed, can i setup a primary job for 1 day retention on main storage and then a backup copy with 7 day retention on secondary storage and obtain what i need as long as the timelines of the backup copy complete i should be good

[foggy]

What backup method do you use (forward or reverse incremental)? With forward incremental you will most likely have more than 1 restore point, depending on the frequency of the full backup (just due to how forward incremental mode works). Moreover, single restore point retention is not generally recommended as it could potentially result in situations, where you find yourself with no valid restore point available.

[chrisBrindley]

This is all in test right now using beta 8 so just playing trying to determine the correct settings, going to try reverse incremental, never used reverse before, always standard, with reverse do we still need an active full to reduce the risk of data corruption and to remove the incremental from the disk?
Does a backup copy ever check the consistency of the backup data it is copying, for instance the reason for an active full is to make sure you dont end up with a massive incremental chain that may have corrupt data?

[foggy]

This is all in test right now using beta 8 so just playing trying to determine the correct settings, going to try reverse incremental, never used reverse before, always standard, with reverse do we still need an active full to reduce the risk of data corruption and to remove the incremental from the disk?

Please review these recommendations on the active full frequency for different backup methods. With reverse incremental you always have the number of restore points on disk equal to the one specified in the job settings.

Does a backup copy ever check the consistency of the backup data it is copying, for instance the reason for an active full is to make sure you dont end up with a massive incremental chain that may have corrupt data?

Yes, backup copy job has built-in integrity check that is performed according to the specified settings.