04-05-2016 07:53 AM
I have setup a google cloud bucket for purpose of backing up monthly full backups. I find however that I can't afford to do a full backup each month, as it takes too long with my currenty internet connection.
How (if possible) can I do a full backup one time, but then have only the changes updated to the cloud, but that change be like a snapshot in the cloud, so that I have a restore option of a full backup (snapshot).
Is this possible using backup exec and google cloud s3 buckets? (or any cloud options with backup exec) Or does backup exec not operate in this fashion?
04-06-2016 03:05 AM
You are kind of asking about synthetic backups - unfortunately this won't help you with cloud - to explain:
The way a synthetic works is
1) Full backup written to disk storage device
2) One (or more) Incremental backup(s) written to the same disk storage device
3) Synthetic job runs to combine 1 and 2 and this is still written to the same disk storage device (and becomes the most recent full)
4) 2 and 3 then repeat but now combining 3 and 2 to make each new synthetic (most recent) full
The above process is often combined with duplication of the full to tape after step 3
If you then consider S3, if the disk storage device was S3 bucket, then step 3 would read everything back from the cloud to the backup server combine it and then write it back to the cloud (so loads of bandwidth used and read operation costs accrued)
If instead you thought OK I will duplicate to S3 bucket after stage 3 and use local disks for the backups, then you are in effect writing a full backup up to the cloud, the only difference is the read operation would be from the backup server and not from the production servers that owned the data (or from the S3 cloud)