Technology Disaster Recovery Plan

 

Disaster Recovery-As-A-Service

Companies are starting to rely more on technology for their critical operations, data protection, and the availability of their systems. As a result companies or providers are offering new solutions and services in order to make IT disaster recovery cheaper and stress-free. With so many options available for Disaster-Recovery-as-a-Service it can be difficult to determine what your business actually needs.

The main purpose of a disaster recovery plan is to have the ability to recover any of your hardware, software, and apps after disaster strikes.  According to Dustin Bolander, CIO at Technology Pointe, “With so many businesses being IT driven, often times business continuity cannot happen without IT having a solid DR plan. The biggest gap we run across is getting the business’ workforce access again. Many IT departments are prepared to recover servers and networks, but have not considered how to communicate and re-enable the staff to resume work” (2017).  DRaaS was developed due to the increased demand in technology and instability of disaster recovery.

According to Doug Hazelman, “Without a robust DR infrastructure, a fire, a flood, an earthquake or even a really bad storm could not only take a business offline, it could take it out completely” (2017).  When executed correctly, DRaaS is more than a safety net for your information and technology because your users never experience the disaster to begin with.  DRaaS is the replication and hosting of physical and virtual servers to a second location, either to a second appliance or the cloud, which is usually located in a distant second site.  In the event of a man-made or natural catastrophe, those replicated systems and data can be booted and accessed.  According to Rohde and Haskett (1990), disaster recovery encompasses four distint phases: prevention, prerecovery, immediate recovery, and return to normal operations (p. 652).  At some point servers, operating systems, and networks will fail and that is when DRaaS will be vital in continuing business operations.

When it comes to Disaster-Recovery-as-a-Service there are three types to choose from: Self Service DRaaS, Assisted DRaaS, and Managed DRaaS.  The Self Service DRaaS model gives you the tools to establish a disaster recovery plan yourself.  This means you are solely responsible for monitoring the recovery environments as well as deciding how often to test to ensure the solutions are working appropriately.  For example, when a disaster happens your IT team is exclusively responsible for the recovery.  This model comes with the lowest investment option, but comes with the tradeoff of time and resources to manage them.  In addition, this model is best suited for a business that has internal disaster recovery expertise and enough bandwidth to manage the recovery environment.

A second option to the Self Service DRaaS is the Assisted DRaaS.  In this model a provider will advise you as you implement, test and manage your solutions.  You are still responsible for your disaster recovery plan like in the Self Service model, but a DRaaS provider is standing by if their assistance is needed.  Although you are still responsible for your disaster recovery plan a DRaaS provider may step in if IT team members are unavailable to help with failover in case of an emergency or disaster.  This option is more expensive than the Self Service model, but it is still cheaper than the Managed model.  This model is best for organizations that have IT assets, but are looking for addition help to provide direction and support with a DRaaS solution.  Lastly, few providers offer this solution and will not offer a recovery SLA, meaning you are ultimately responsible for the recovery during an emergency or disaster.

The third and final DRaaS solution is a Managed model.  In this model the vendor or provider is responsible for the development, testing, and management of your disaster recovery plan.  They will manage nearly every aspect of your recovery plan in order for the team to focus on the companies’ priorities.  When disaster strikes the Managed DRaaS provider provides and supports the recovery process.  This model provides a larger investment than a Self Service or Assisted model, but it provides the highest level of support and surety that the disaster recovery plan will meet the companies’ needs.  Organizations with high demand for IT teams involved in the tactical operations of their business should use this model.  Furthermore, this model will come with a SLA from the provider to ensure they will be responsible for the recovery in case of an event or disaster.

In order to best understand the replication and backups of DRaaS it is best to picture it on a hot to cold map.  This meaning that the “hotter” the solution the faster the recovery will have.  For example, it a solution is hot then the system could be restored in a matter of minutes versus days on the other end of the spectrum.  Another thing to consider when looking at a hot solution is the investment.  The hotter the solution typically equates to a larger investment.  When choosing a solution it is best to find one that balances the needs of your applications to the cost and timeliness of your recovery.

The cheapest recovery option on the scale would be backups.  According to Wu and Li (2014), it is a process that copies all or part of data from one hard disk to another storage medium (p. 1207).  They are often disconnected from running technology and can be retrieved in order to restore system applications.  Since these data backups are housed in a physical form like a tape or disk, there will be a large increase in downtime to identify and restore the system.  This system typically does not offer the most up to date recovery solution, but they do offer a great long term solution for data preservation for a low cost.  If a companies needing a faster recovery timeline then replication would be the best solution.

While backups store information for long periods of time, replication will duplicate data as changes in the environment occur.  According to Margaret Rouse, “The result is a distributed database in which users can access data relevant to their tasks without interfering with the work of others. The implementation of database replication for the purpose of eliminating data ambiguity or inconsistency among users is known as normalization” (2012).  This option will allow you to quickly access recent changes to your data during a disaster, but it is not meant for long term retention.

Real-Time Replication vs. Backup-Based Replication
Provides recovery in seconds or minutes	Records full environment
Tracks Changes as they occur in the environment	Reports changes on regular basis (typically once a day)
Enables failback to earlier version if needed

Table 1

You should not have to scramble in the event that disaster strikes.  For every minute your business is down it is losing revenue and reputation.  That is why it is best to approach disaster recovery from the viewpoint of disaster avoidance.  With this in mind the term IT Availability has emerged.  IT Availability differs in several ways from Disaster Recovery. IT Availability recognizes how IT and businesses depend on each other whereas Disaster Recovery views technology in a boxed perspective.

Disaster Recovery Focus vs. IT Availability Focus
Invest in an “insurance policy”	Invest in ability to serve clients
React to downtime and events	Be proactive to minimize risk
Rely on backups to store data	Failover and failback to ensure service
Recover anywhere from hours-days	Recover in minutes to hours
Emphasis on technical infrastructure	Emphasis on serving end users
Develop minimal process and reporting	Process-driven and documented
Emphasis on avoiding “catastrophes”	Emphasis on continuous improvement

Table 2

With the increased need of Disaster Recovery and the customer demand for uptime, many companies struggle implementing an effective solution before a disaster impacts them directly.  According to 2016 Cost of Data Center Outage: Ponemon Institute, it is indicated that just one minute of downtime could cost a company nearly $8,000, but with a proactive approach to disaster recovery could equate to a substantial cost savings. DRaaS can be an effective solution to combat downtime and achieve IT availability. If approached in the right way, it can provide comprehensive proof of recovery to stakeholders and most importantly, increase your availability during an event.

Quite simply, the success of your business depends heavily on both DR and BC plans working in conjunction. Most of today’s network outages don’t occur because of weather-related incidents, but because of human error. It’s crucial that companies plan for the full gamut of events that could likely, and unlikely, impact their operations. In conclusion, DRaaS can increase you uptime and cost savings when disaster unexpectedly strikes.  It is essentially insurance for your business.

 

References

Bolander, D. (2017). Disaster Recovery vs. Business Continuity. Retrieved March 14, 2017, from https://www.bluelock.com/practical-guide-to-draas/disaster-recovery-vs-business-continuity/

Guster, D., & Lee, O. (2012). Outsourcing and replication considerations in disaster recovery planning. Disaster Prevention & Management, 21(2), pp. 172-183.

Hazelman, D. (2017). What is Disaster Recovery as a Service (DRaaS). Retrieved March 14, 2017, from https://www.bluelock.com/practical-guide-to-draas/what-is-draas/

Rhode, R., & Haskett, J. (1990). Disaster recovery planning for academic computing centers. Communications of the ACM, 652-657(33), pp. 1207-1210.

Rouse, M. (2012, April). What is database replication? – Definition from WhatIs.com. Retrieved March 14, 2017, from http://searchsqlserver.techtarget.com/definition/database-replication

Wu, Z., & Li, H. (2014). Analysis of data backup and recovery system. Applied Mechanics & Materials, 631-632, pp. 1207-1210.