Availability for the Always-On Enterprise
Post Reply
jr991
Lurker
Posts: 1
Liked: never
Joined: Dec 04, 2017 3:42 pm
Full Name: Jeff
Contact:

Archiving Backups

Post by jr991 » Dec 04, 2017 4:02 pm

Just wanted to get some thoughts on best practice for archiving backups for long term retention. I currently have a physical Veeam backup server which I backup to daily with 14 restore points stored on that server. I then replicate each hour to an off-site ESX host at our recovery site. I currently don't archive backups and I want to start doing this... just not sure what's best. We are a small environment, 1 production ESX host with 8 VMs totaling 1.5TB in size. I've read that best practice is to have a high performance backup server as the primary backup target and a second backup target which would be for the archive. I'm just wondering if for an environment my size would it be okay to have 1 Backup server act as the Veeam server/proxy and repository for both short term and long term backup retention. I'm needing to replace the current Veeam server and not sure whether to get 1 server with lots of storage or 2 servers to serve the different repositories. Then I'm wondering about whether it would be best to have the archive repository at the recovery site... or one at each site. So many options I'm spinning my wheels. Would really appreciate some thoughts and suggesstions from the community. Thank you.

csinetops
Expert
Posts: 113
Liked: 15 times
Joined: Jun 06, 2014 2:45 pm
Full Name: csinetops
Contact:

Re: Archiving Backups

Post by csinetops » Dec 04, 2017 6:43 pm

I would use cloud connect to push your backup copies off site to a cloud connect partner. Its pretty cheap and No hardware etc to worry about. We use SingleHop for a provider to store around 1TB of data for our remote site. It has been working well for a few years now.

Mike Resseler
Veeam Software
Posts: 4825
Liked: 509 times
Joined: Feb 08, 2013 3:08 pm
Full Name: Mike Resseler
Location: Belgium
Contact:

Re: Archiving Backups

Post by Mike Resseler » Dec 05, 2017 7:26 am

Hi Jeff,

First: Welcome to the forums

Second: The solution of csinetops is certainly a good one. Another one could be tapes or rotating devices. And finally there is indeed the possibility to do a backup copy job to a repository off-site.

A few things to keep in mind:
1. Cloud Connect provider: Easy to use. No additional hardware/ maintenance and so on. Plus, those backups are more or less "air-gapped" which means that you normally can recover in case of Ransomware. While indeed mostly not expensive, it will be a bit more expensive as (for example) tape but peace of mind and less work that you have to do yourself is also something that comes with less cost :-)
2. Tapes: Still very popular (No, tape is not dead :-D) but that means you need to handle the tapes yourself (They are air-gapped so that is great) but you need to do the handling yourself and get them offsite also. Most probably (in most cases anyway) still the cheapest solution
3. Rotating devices: Good solution, but again comes with a cost of "handling" the devices. Ideally those would be located offsite, but you are probably not there every day
4. Backup copy job to offsite location. You don't need a secondary backup server. Do a backup copy job to a repository server offsite. In worst case scenario (meaning you lost also the VBR server), you go to the offsite location, install a clean VBR server (which doesn't take that long), do a scan of the repository and you can start restoring whatever is needed.

My advice would be to investigate first:
1. Will I be at the offsite location or is there someone who can do some work
2. What is the money I want to spend on it and do I also have the time (or FTE) to do that additional work
3. What is my RTO in case of a restore from those archives.

Based on these responses, you can investigate the different options and choose what is best for your organization
Makes sense?

M/

Post Reply

Who is online

Users browsing this forum: Bing [Bot] and 38 guests