Despite the hype that leads us to look at cloud computing as the new platform for critical applications, most enterprises are not buying public cloud computing systems as their primary platform for business-critical applications. For now, perhaps they are right to wait. That is, until business continuity or disaster recovery (DR) becomes the next killer app for cloud computing.
The idea is simple. A business continuity app will leverage cloud computing platforms in any combination of IaaS, PaaS, and SaaS as systems that can support the business during times of outages or outright disasters. This means you continue to leverage on-premise systems for day-to-day processing, but your continuity of business strategy will leverage cloud resources, or in simpler terms, your cloud resources are your “hot stand-by.”
Backup systems exist today and have been around for a long time. When I built banking systems, there was always a data center somewhere that had an exact duplicate of the primary data center, which could be turned on at any time if the primary center went down or was taken out by a natural disaster. While it cost millions to stand up a back-up data center, the cost of not having the business continuing to function was in the millions of dollars per hour.
Leveraging cloud computing resources as back-up systems is nothing new, but now we have a greater number of innovative cloud-delivered systems to choose from, providing everything from storage to security, application processes and testing, enterprise software, and complete platforms, all on-demand. It’s now possible to build or port complete systems to the cloud, but not for the purposes of elasticity or cost reduction, but to provide a pay-as-you-need-it platform that provides all core information processing services that can be turned on at any time.
The DR Value of Cloud Computing
First, the cost of leveraging cloud computing as part of your business continuity strategy. No data center investment is required, or hardware and software costs incurred. What’s more, you can turn it on when needed, and they only bill you for the resources you actually leverage. This opens opportunities for businesses that typically could not afford a back-up center. The dollar estimate is that the cost is about a fourth that of traditional backup sites, mostly from ongoing operational savings.
There are three basic types of cloud computing providers: software-as-a-service (SaaS), platform-as-a-service (PaaS), and infrastructure-as-a-service (IaaS). While SaaS and PaaS are important to the world of cloud computing, IaaS cloud computing providers should be the focus when considering DR.
IaaS providers typically offer up entire platforms such as Linux, Windows, or other operating environments within virtual spaces that are allocated to the users of the IaaS provider. You leverage the infrastructure as if the infrastructure were in your own data center, or under your desk. You’re just doing so as a service, meaning you only pay for the increments of use.
Keep in mind that cloud computing providers charge for different types of usage. Some charge for storage used, others charge per hour, and still others charge per-bandwidth leveraged. Understand the usage charges and service level agreements (SLAs), upon entering into the agreement, and make certain the SLA defines your performance expectations to the cloud computing provider.
The second is the speed in standing up a site, typically a month or two to port and test the software. Again, no need to purchase and test hardware and software systems or negotiate data center space that can’t be collocated with your production systems.
Over and above the cost advantages, this is the true strategic advantage of leveraging cloud computing providers for DR. You can become a cloud computing customer today with a credit card, or they typically provide a free trial period or a certain level of usage for free.
This “self provisioning” feature of cloud computing is the real strategic advantage when approaching DR. IT can allocate and configure back-up sites as needed without having to go through waves of hardware and software procurements, or finding data center space which is geographically correct.
Finally, there is the fact that most cloud-computing systems are ubiquitous. Thus, if you have access to a browser and the Internet, then you can access your core business systems and continue your business anywhere in the world. This is handy considering most disasters mean employees won’t have access to their offices or workplaces.
This conceptually means that if your facilities become inaccessible, as long as key employees have an Internet connection they would have access to key applications and data required to run the business. Considering that cloud computing systems are already optimized for Internet delivery, including leveraging redundant sites to optimize performance and for their own DR requirements, moving to cloud computing-based applications should be relatively low risk, depending on the types of applications you’re running.
Mapping a Path
So, how do you move toward a cloud computing-oriented DR strategy? You start, as with any DR strategies, by understanding the requirements, including the relative cost of risk. This provides the foundation of the cost benefit, and thus the amount of investment that should be made.
From there you can assess the fit of a cloud computing provider within the DR strategy. In many instances, cloud computing is not a fit due to specialized requirements of the applications, such as core proprietary enterprise software that does not yet support cloud computing. Or, many enterprises deal with data that has special security requirements, such as PII and HIPAA data. Indeed, in many European nations, financial data is not allowed to leave the country. Many cloud computing providers exist outside of borders, or replicate data to data centers out of the country as a part of their operations. In short, you need to understand all compliance issues.
You also need to consider the inherent risks. You need to trust your cloud computing provider to live up to any SLA, as well as consider the fact that many are small VC-backed startups and could be sold at any time, and thus the level of service could change. Your data is being physically mixed together with data from other companies, even though logically they appear as dedicated systems. There are some situations when your data could leave the premises without your permission, such as when servers are seized by law enforcement around evidence related to other tenants on that cloud computing platform, but your data leaves on the servers with the data owned by the suspect company. Again, you need to weigh the risk with the benefits.
As cloud computing matures, there may be more upsides to leveraging providers as a part of your DR strategy, even with all of the downsides considered. Most providers still don’t view DR as a killer app for their platforms because those who leverage cloud computing for DR will not be intensive users. That is, until they are needed to support business continuity.
For now, this is another good option for DR. One that will most likely be attractive to small- to medium-sized businesses, but larger businesses and perhaps the government will join the trend as things move forward.
David Linthicum serves as CTO of Bick Group where he focuses on emerging technology spaces and the industry’s move toward cloud computing.