Video Screencast Help

SSR to backup small files and replicate....

Created: 13 Apr 2013 • Updated: 25 Apr 2013 | 5 comments
KeirL's picture
This issue has been solved. See solution.

Hi

I understand that Symantec System Recovery is a good option when having to backup small (1TB with >million files) on a single client as I'm seeing very poor performance with Backup Exec - please correct me if I'm wrong. Is there any restriction with folder structure as I have about 10 folder levels down on this volume.

Also I'd like to understand where\when the processing takes place to allow for individual file restore? Is this done on the before the backup runs (or after) and does it add much time\processing to the server itself?

I'm then wanting to replicate this using the built in copy option to a DR site but I guess I would need to 'seed' the destination as the WAN link is quite slow.... what is the best way to achieve this? Can I copy the .v2i file to USB and ship to the DR site and somehow 'import' it? I'm not sure how this bit really works.

Finally - is there any integration with Backup Exec 2012 - eg can the destination be a Backup Exec dedupe folder?

kind regards

Keir

Operating Systems:

Comments 5 CommentsJump to latest comment

Markus Koestler's picture

Since SSR does a block level backup of the drive you backup everything on that drive and no individual files are transfered.

afaik there is no way to seed the data to the offsite location.

Afaik there is no integration, but try to post this as a separate question in the Backup Exec Forum, they should know.

*** Please mark thread as solved if you consider this to have answered your question(s) ***

KeirL's picture

Thanks Markus

So can I assume from your first sentence that it shouldn't make any difference if the files are large or small - therefore no performance decrease either way.....?

Can it determine whitespace? I've done some testng on a 32GB volume with only 1GB of data - it completed in just a few minutes and indicated it had backed up 1GB so I assume it didn't transfer the full 32GB worth of blocks?

And I assume it must have some kind of database somewhere where it determines which blocks form which  file (as I can do single file restores) - Is this correct?  Does it take long to complete this part of the process?

many thanks in advance

Keir

Markus Koestler's picture

Out of my experience only the size of the data matters, but this is my experience. It determines whitspace in form of unused blocks, yes. SSR uses the VSS volume shadow copy provider to generate a snapshot of the volume and then compare this to latest recovery point. Normally this process doesn't take that long, but again it depends on the amount of data to be backed up.

*** Please mark thread as solved if you consider this to have answered your question(s) ***

SOLUTION
Markus Koestler's picture

Anything still unclear ?

*** Please mark thread as solved if you consider this to have answered your question(s) ***

KeirL's picture

Hi Markus

No  - many thanks

Apologies for not marking this as a solution earlier.

kind regards