Backup Exec New Implementation
We are looking to implement a new backup solution and I was wondering someone might be able to help me..
We have 2 sites.
- Site 1 is the Data Center
- Site 2 is where end users are
Data to be backed up will be
- MS SQL (that will be the bulk of the data)
- Exchange 2003
- Netapp cifs
The requirements are
- All data to be backed up will be in site 1
- Backup data in site 1 using deduplication to iscsi (20TB storage)
- Replicate all backups to site 2 (deduplication iscsi – 20TB storage)
- Should either site go down in a DR situation, we need to be ready to go by having all data ready to be restored without having to mess around rebuilding a BE server.
From what I have gathered, looking at various bits of documentation, it would appear that the following set up would be the best solution for us.
- Enterprise Server Option with 2 servers (one at each site) with CASO
- Each server will have local iscsi storage
- We’ll run backups from site1 (dedupe)
- Replicate backups to site 2 (dedupe)
- Would the above give us what we want to achieve at the best cost?
- I understand that if we use NDMP for nettap, we can only restore to a nettap device – Is this correct. If it is, backing up using NDMP is rather inconvenient since we have no guarantee that we’ll have a netapp to restore data to in a DR situation.
- To maximize the benefit of deduplication, we should use MS SQL itself to backup database, as opposed to 3rd party we have been using (lightspeed) or is it best to let BE manage database and tlog backups?
- 32 TB is the max storage we can have at each site per BE server?
- During verification, data gets rehydrated – Would this be for replication as well or is it just backups at site 1 that would get rehydrated
- The policy will probably be:
- First backup will be full
- Daily incrementals (Monday to Thursday)
- Weekly full
- Monthly full
Since we are looking at deduplication, am I right in assuming that we don’t really need to worry about expiring data and we can have infinite retention without impacting storage utilisation?
- Can servers/individual backups/files be selectively deleted from within BE if the data is no longer required to free up storage?
- Resuming after DR, we would like to be able to restore the backup environment (rebuild the missing backup exec server and data at which ever site. Is this easily achievable?
- What is the best location for the CASO server and why?
- I understand that dedupe volume cannot be easily extended. Does that mean that I would have to purchase additional license and add 1 server on each site to give me another 32 TB ... and so on for each 32TB of storage per site?