Copy Job of a Copy Job? (GFS Points to different storage)

[david.buchanan]

Hi All,

Sounds like a weird question but basically I want to have my GFS points (Weekly, Monthly and Yearly) go to different storage than where the Copy Job VBK goes.

Our client has Backups locally which are then copied off-site over a 10MB link to our DC. These copy job files are stored on fast disks in case we need to use them for DR.

However over the long term we will be keeping Weekly, Monthly and Yearly images. This adds up to a lot of data. I would like to push these into our de-dup appliance while keeping the base image on fast disks.

I know I could run two Copy jobs but this means I would need to send the data across the link twice (we have Wan Accelerators that will help with this), but I still feel like there is a better way to approach this.

Any suggestions?

Thanks,
David.

[veremin]

Not sure whether it helps, but if you create a SOBR in performance mode, add two repositories as extents to it, set performance policy for it, make a dedupe repository host only full backup and the other repository only incremental ones. Finally, enable active full backups for the backup copy job.

This way, full backups will be stored on one repository and increments on the other. Even performance should not decreased severely in this case.

Thanks.

[david.buchanan]

Hi Ermin, I thought I read somewhere that you should not use a Scale-out repo for copy jobs? If that's not correct then your idea is an interesting one. I'll have to look into it.


Sent from my iPhone using Tapatalk

[veremin]

Hi Ermin, I thought I read somewhere that you should not use a Scale-out repo for copy jobs?

Nope, this certainly does not sound right at all. So, feel free to check the suggested approach. Thanks.