What jobs can run concurrent: Backup, Backup Copy, Tape...

[matt_778]

I have a Production and DR Site:

Prod
Vmware hosts with SAN
Physical Backup Host attached to SAN
Veeam 8, no WAN accel
7 x Backup Jobs with 37 VMs protected, Forward incremental (no periodic or synthetic fulls)
Backup Copy (BC) jobs which mirror the Backup jobs
Future: Tape Autoloader controlled by Veeam

DR
Physical Windows host (No Veeam installed)
50Mb L3 WAN connected

The Daily Backup/Backup Copy jobs run fine, though there is a problem monster VM in the environment which creates around 7-800GB of change daily. This VM cant get down to DR via daily Backup Copy job due to the size - I think it would need around 60hrs at current state.
Note: We also want to write to tape weekly (in Prod) and I expect this take take most of a weekend to complete (Also want to try Virtual Full to tape)

We have a goal to get this monster VM in DR as well, though the frequency can be less than the other VMs (i.e weekly rather than daily) client is not so willing to increase bandwidth*

I am thinking of possible workarounds - it would be helpful to know if any of the job types - Backup, BC, Tape conflict each other?

My one thought was to run with backup jobs/BC jobs during the week (Mon-Fri) the monster VM BC job will likely need all this time to complete one rollback point (whilst multiple backups run) - Can the BC job continue to run over the weekend, when I need the Tape job running?
Worded another way - can the BC jobs run 24/7 whilst the backup and tape jobs run exclusively?

*Current bandwidth @ 50Mb is throttled. I might be able to adjust throttles during non-peak times. If a Backup Copy job is in progress and we rollover to a non-peak time, is the increase is bandwidth realized immediately? or does the job continue to run with the bandwidth it started with?

[Shestakov]

Hello Matt,
Backupcopy and Tape job can work at the same time with no issues, because they don`t lock the backup files on write, while backup job does and if backup job starts while dependent backup copy is running, the copy job will be interrupted.
For the "Monster VM", you can use backup copy seeding for the first job run. Then with 50Mbps it will take around 1,5 days to transfer incremental runs outside. So the option is either to leverage WAN accelerator or widen the bandwidth or give the backup copy job the whole weekend.
Thanks.

[foggy]

I strongly recommend to consider using WAN accelerators on such a slow link.

*Current bandwidth @ 50Mb is throttled. I might be able to adjust throttles during non-peak times. If a Backup Copy job is in progress and we rollover to a non-peak time, is the increase is bandwidth realized immediately? or does the job continue to run with the bandwidth it started with?

Throttling rules are applied immediately.

[matt_778]

Thanks both for the replies, that give me something to work with. The monster VM is already seeded in DR, the incrementals are are still massive:/

[Shestakov]

You are welcome, Matt.
The way to go for you is to increase the bandwidth or leverage WAN accelerator.
Let us know if it help once you try