So far in this paper, we have covered 5 types of backup technology: File-level backups, block-level backups, complete database backups, granular database backups, and virtual machine image backups.We have covered a range of common disaster scenarios that each highlight the strengths and weaknesses of the different types of backup in order to understand how they would provide different recovery outcomes for each scenario, and we have discussed the reasons why each type of restore in each scenario provides those different outcomes.
In the same way, different types of backup storage have different strengths and weaknesses. Those strengths and weaknesses need to be considered carefully before incorporating them into your disaster planning. If backup data can be described as a safety net, then backup storage is the underlying hardware that allows that safety net to work.
Backup storage is what protects, stores, and restores your backup data. In many cases, recovering from a disaster is a time in which mistakes and delays are painfully costly. Just like in production, poor habits and decisions relating to your backup storage can make your recovery processes more time consuming and more complicated.
An effective and prepared disaster recovery plan calls for backup storage that is ready for any disaster at a moment’s notice and also provides performance that is acceptable to your recovery time requirements.
There are many different disasters that can have varying ranges of effect on the systems in your organization. Some disasters will be limited in effect and bring a more limited disruption to production, while other disasters that could entirely halt production can have effects ranging from equipment failure up to and including destruction of your building.
In other words, backup data that can provide both high performance and high survivability needs to have backup storage in at least 2 locations.
Your local backup storage is the storage on which your backup data will be created and it will be used most often in your recoveries. Because it is local to your equipment, it should provide the best possible performance. However, that locality also means it will have the lowest survivability. A disaster that can affect your production premises could also destroy your local backup storage, which is one scenario your disaster planning must cover.
Depending on the size and complexity of your backup data, local backup storage can range from anything as simple as a single USB hard drive to a NAS, or other networked desktop or server. It only needs to meet your capacity and performance requirements, which can only be assured by regular restore testing. Most external hard drives should be replaced annually because they are the least reliable of all hard drives and are not rated for most backup workloads. Additionally, local backup storage should be secured against unauthorized access to avoid interference from ransomware or other viruses running in production.
Good storage organization dictates that backup data should be dedicated. It should never be mixed with production data or hosted on production storage. If you are using production storage to create backup data, including database dumps or internal backups created by specific products, they will cause unnecessarily large incrementals in your image backup, and that will increase the time it takes to recover from an image backup.
It is best to configure backups to occur directly on your local backup storage to prevent mixing the different data on production storage. If there is a failure with your production storage, your backup data could be inaccessible if it was left on production storage. It is important to always host backup data on physically separate equipment from production.
Offsite storage has requirements and characteristics much different from local storage. Your offsite storage must provide the highest possible survivability so that it can be used in any disaster at any time. While you may prefer to use local storage for most recoveries because of the higher performance, your offsite backup storage must be able to fulfill any recovery requirement in the event that your local storage cannot.
In the case of local backup storage failure, or a total loss complete from a natural disaster, offsite backup storage is the last preventive measure you can have against data loss.
The high survivability aspect of offsite backup storage brings additional requirements for redundant, enterprise-grade equipment, and that the offsite storage location always is available, highly secure, and geographically very distant from the production site in order to avoid being affected simultaneously by a natural disaster.
Because of these requirements, industry-accredited data centers are the preferred choice for offsite backup storage. However, the costs of building and internally managing infrastructure in a data center that could fulfill the requirements of offsite backup storage would be prohibitive for many organizations.
Not only that, but as we have discussed, backup and restore operations are not as simple as they seem.
For these reasons, starting a relationship with a company specializing in offsite backup will give you lower cost of entry into a good facility and also give you access to individuals with many years of experience and knowledge that can be leveraged in the formulation of your disaster planning as well as testing, validating, and assisting your backup and restore procedures.
Proper data protection and data loss prevention are great responsibilities. It could be argued that monitoring and maintaining backups is too great a responsibility for a small IT department that is already busy with operational matters, and it is especially true in the case of outsourced IT service providers that are in a constant state of near chaos keeping up with the demands of many different clients who do not share concern specifically for your data.
Because of the this and the inherently high stakes regarding offsite backup storage as your last line of defense, it is best for the provider of offsite backup to be a dedicated relationship, separate from all other IT services in use by the organization, and that the offsite backup provider be a company that specializes in, and does nothing other than, provide backup and disaster recovery services.
In the event of a total disaster with unreasonably long or undetermined downtime, the shortest path back to production could be to restore your backup data onto servers hosted by the backup company.
