Move-VBOEntityData

[aheath]

Following on from my previous posts I need to split 2 very large repositories in to smaller more manageable repos, in the past I have tried using Move-VBOEntityData, but was extremely slow and in the end gave up and resorted to cloning the repo and using Remove-VBOEntityData to remove the data we didn't want in each which was much faster

However due to the size and what I am trying to achieve I am evaluating the options again

1. Migrate all of the data in to smaller repos using Move-VBOEntityData, then delete then delete the original parent repo
2. Clone the original repo, remove the unwanted data in each using Remove-VBOEntityData, shrink the adb files using esentutl (unsupported), then move everything to smaller disk volumes
3. Any other suggestions welcome

Also, has there been any performance improvements with Move-VBOEntityData in later versions? Appreciate its an item level move so will never be as quick as cloning

Any input or guidance much appreciated

[Polina]

Hi Ashley,

The proven solution is to use the Move-VBOEntity data (so #1 in your list).

What product version are you currently running?
Are there any concurrent backup/restore tasks along with the data move operation?

Also, check out this KB https://www.veeam.com/kb3067, there're some tips that may help you optimize your script.

Thanks!

[aheath]

Hi Polina, we are running 5.0.2.42, which I believe is the latest version

I was looking to move the data in 4 batches to new repositories, approx 15-20TB in each batch. Depending on how long the operation is expected to take we can pause all backup jobs and restore tasks, but if it is likely to take several days or worst case run in to weeks then we will need to keep backups to the other repositories running somehow

Our Veeam B&R backups on very similar servers/storage can hit 650+ MB/s, but I haven't tried Move-VBOEntityData for about 2 years and don't want to start down that road if performance hasn't improved significantly since then, it was really slow the last time I used it using a very similar script. Do you have any stats that you could share that may help before I resort to testing it in our live environment ?

Thanks