Backup is not Disaster Recovery, DR is not Backup

Backup does not solve everything.

Backup is the most basic form of a lifeline, it’s like insurance. Nobody cares about it until the time when a disaster really happens. Backups are important as they are needed to restore the original data if an incident happens.

There are cases when backups do not work in emergency situations due to incorrect procedures, data corruption or tape deterioration. You will need backup when data is accidentally deleted, overwritten, system updates fail etc. Recovery Time Objective (RTO) and Recovery Point Objective (RPO) are two of the most important measurements of a disaster recovery plan.

RTOs

The recovery time objective (RTO) is the targeted duration of time and service level within which a business process must be restored after a disaster (or disruption) in order to avoid unacceptable consequences associated with a break in business continuity. For example, business will be in serious trouble if the system is down for more than RTO. Every second of down time means business operations are affected. Let’s say that you did your backups, have you considered how fast your backups will take before they can be restored? Will it take one minute, one hour, or even days? This makes having a Service Level Agreement (SLA) with your customer extremely important. Ask yourself, how much faith will a customer put in your service if they had to wait one day or one week for you to recover their systems? It is highly likely that your customers will jump to your competitors in the event of an outage. And also, how much productivity will your employees lose? How much impact will there be if suppliers and vendors are not able to reach you? What will be the potential loss for businesses? All these factors need to be taken into consideration when you determine your RTOs.

RPOs

RPO determines the maximum allowable data lost in the event of a disaster. Even if your backups are working, the typical time required to restore a system for backup is typically hours or days. How would you decide which backup to restore to? The one completed 24 hours ago, last week, or the day before? And did you even do a backup in the first place? If you had done a backup 24 hours ago, the Recovery Point Objective (RPO) will be 24 hours. This means that you can recover your data back to 24 hours, provided that your backups work. This also means that from the time you did the backup (which was 24 hours ago), whatever information you had will be lost.

The reality is, having backups is good, but we need to answer these questions first when planning a disaster recovery scenario:

How fast can we restore the data that has been backed up?
Which point can your data be restored?
Do you need to plan for a standby site (usually termed as Disaster Recovery (DR) Site)?

D is for Disaster. R is for Recovery.

So, it’s great if your organisation has a DR Site. Having a DR Site means that you can get your site up quickly in the case of an outage. But this does not equate to having a backup with different versions that can be restored. I do agree that some backups do have some versioning to roll back to but there is usually a limit to the versioning. What happens if due to an audit or some reason, there was a need to obtain information from a previous month’s backup. Oh oh, did I hear someone say “No backups needed when you have a DR site?”

Solutions from rhipe

Some of our customers are using Veeam Backup and Replication as it enables you to do both backup and DR within the same console. With the replication portion, you can customise a retention period for replication and keep up to 28 restore points for every replica. What this means is that you can rollback to any of the previous 28 restore points.

If you are an Azure user, do note that Azure offers Site Recovery to orchestrate your disaster recovery plan. Site Recovery provides recovery plans that can include scripts. In the recent Veeam Availability Suite 9.5 launch, Veeam integrated the feature enabling Direct Restore to Azure. What this means is that you will be able to use Veeam Backup & Replication to select and recover any restore point directly to Microsoft Azure.

On the other hand, if you need a dedicated DR site with full control over the hypervisor; SoftLayer will be the answer. It will be like having your own site as you can even restart the hypervisor if needed. Not forgetting, if you want to back up a wider range of devices like servers, Virtual Machines (VM), laptops and even mobile devices, Acronis will be your answer for both android and iOS devices.

In summary, backups do serve their purpose and the same goes for DR; but they are used for different purposes. It is always good to have Backup AND DR if budgeting allows. When a disaster really happens, you will be glad that you have both a Backup and DR to keep your job!

So repeat after me “Backup is not DR and DR is not backup, backup is not DR and DR is not backup”.

By Nick Tan, Partner Enablement Specialist, rhipe

e: nick.tan@rhipe.com