We have been discussing network security for the last couple of weeks. How to create your digital fortress and protect your data.
Photo courtesy of Grant Wickes used with permission under CC license
Edge protection, anti-virus, firewalls and other security appliances minimize the chances of data loss from a digital attack but they aren’t fool proof. There are times when the only option is to restore a known good copy of your data. Furthermore, digital protections won’t do a thing to help you recover from a force majeur. For both of these instances you need to have a good backup and disaster recover paln
Backup and Disaster Recover
Historically, companies would periodically perform a full backup with incremental backups in between. Backups would go on some sort of external media such as a tape drive, CD/DVD drive or external hard drive. This backup schema has the added benefit of providing version control, i.e. the ability to recover an older version of a file even if the current production file has been changed.
In order to prevent the loss of data from a catastrophic event such as a fire, best practice was to remove the media every day thereby ensuring a copy of the backup existed at another physical location other than the office. Assuming that there was always compliance with the removal of the media, the problem with this scenario is that the best case scenario had you recovering data that was 24 hours old. This assumes that you a server to restore to. If your server has a catastrophic failure it may be several days before you can restore your data.
As internet bandwidth grew, the necessity to “sneaker net” your data offsite diminished. Today, incremental cloud storage backups are the norm. With the advent of cloud storage services like Google for Work, Office 365, etc., backing up your data is not even always required if your data is stored in the cloud.
However, if regulatory requirements or business considerations require data to be stored on-site we are still faced with the requirement to backup data both on and off site.
The answer lies in virtualization. Today’s technology allows you to install a backup appliance that periodically images your server(s), not just the data. If there is a failure a virtual server can be activated that acts exactly like the lost, production server. The process of spinning up the virtual server takes 10 to 20 minutes. Once spun up, everything functions exactly like they did when the production server was operational. Now if it takes a couple of days to repair the production server, business services are not impacted. This allows companies to restore business services quickly while the production server is replaced.
These appliances have the additional advantage that they automatically send incremental backups to the cloud. In the event of a catastrophic disaster, a new production server can be established at a different location or users can access a virtual server in the cloud. This is the best way to ensure business continuity.
To summarize the best backup and disaster recovery schemes have the following:
- Local backups
- Cloud based backups
- Bare Metal backup
- multiple version control
- the ability to quickly restore data
- the ability to quickly replicate failed servers