This is a post about what to back up and what to do with that data once it is backed up.
A successful backup strategy will consider the following 7 questions:
- What files and configurations are backed up?
- Where the data is stored?
- How often the data is backed up?
- How long the backups are stored for?
- Who gets alerted when backups fail?
- What actions are taken after backups fail?
- How often and to what degree restorations are tested?
1. What to back up?
At a minimum, critical files - such as working documents, databases, spreadsheets, employee and customer information, etc. - should be backed up every day. Going a step further would be backing up entire hard drives and configurations via image snapshots.
If you imagine a hard drive to be a long line of zeros and ones, an image snapshot would be equivalent to copy/pasting that string of digits into a notepad document. Image backups store operating system information, end-user configurations, application data, and all files/folders. The drawback is the sheer size of the backup files.
2. Where is the data stored?
We would recommend storing images onsite and files/folders offsite. Taking it a step further would be throttled replication of the image snapshots to an offsite location.
We emphasize throttled replication, because the image files are very large and could easily bog down entire networks and adversely affect business productivity.
Offsite replication of critical data is essential to overall business continuity because it allows organizations to recover in times of sitewide disaster, such as flood, fire, vandalism, theft, and acts of God.
3. How often to back up?
This will largely depend on the company industry and nature of its business. For example:
- How often is data modified?
- How much data is modified often?
- How many locations is data stored in?
- How much total data is there?
- What are typical file sizes?
At a minimum, modified files and folders ought to be backed up nightly and replicated weekly.
4. What's the retention period?
Again, there is quite a bit of variance here, as various compliance standards require slightly differing retentions. Generally speaking for business not subject to industry compliance guidelines, 3-6 months of archived backups should be sufficient. However, various compliance guidelines (HIPAA, for example) do require some data to be stored for up to 10 years or more.
5. Who gets alerted of failure?
At a minimum, a designated member of the IT support team should be alerted of failed backups. Ideally, this team member will have the experience and authority to resolve the error and restart the backups. We would also recommend alerting a non-technical team member, such as an Operations Manager, Compliance Officer, or Practice Administrator, so s/he can appropriately follow up with IT when necessary.
6. What actions are taken after failure?
Quite simply, the backup administrator should have the authority to investigate, resolve, and restart backup systems after failure. If there are too many check and balances here, or too much bureaucracy around remediation, issues may not get resolved in a timely manner.
7. How often - and to what degree - do we test?
This is the question that separates a truly successful disaster recovery plan from the rest. Restoring files and folders is critical to every day operations, but in times of crisis, we should also consider:
- How will the servers and network configurations be restored?
- Is licensing and software information up-to-date?
- How quickly can the systems be restored?
- If a site goes down, is remote work a possibility?
- How long would it take for the virtual offices to go live?
- How much revenue is lost every hour the systems are down?
Reliable disaster recovery plans are thoroughly tested in a lab environment and then maintained over time.
To learn more about options available through Scalable Business Technologies in Murfreesboro, TN, watch the video below.
"Prepare the umbrella before it rains."
-- Malay Proverb