Downtime incidents are serious business. Natural disasters, power failures, employee mistakes, and malicious attacks can pound your bottom line. A serious downtime event can immediately lose money when customers cannot place orders. In a regulated industry, losing data can result in large fines. And downtime can hurt the company’s reputation and invite customer lawsuits.
Traditionally, the best protection against application downtime is a remote hot site that can quickly take over application processing. But mirrored hot sites are expensive to buy, maintain, and staff. Only the enterprise has the budget and resources to maintain fully mirrored hot sites and to replicate data to multiple locations for disaster recovery.
This expensive proposition works for them. Most large enterprise already owns multiple data centers, and downtime can cost them millions of dollars in lost data and reputation hits. But its expense and complexity leaves most smaller businesses out in the cold.
Failover for the Rest of Us
Now these companies can use remote failover thanks to cloud-based disaster recovery services.
In this scenario, providers store their customers’ virtual server images and replicated data in the cloud. Once the initial upload is completed, the service continually replicates and verifies application and data changes. When a disaster occurs, application processing fails over to the cloud where the provider spins up the virtual server images and replicated data.
Users typically access the applications via VPNs or Remote Desktop Protocol (RDP). Simultaneously IT begins to recover their physical site, or to set up a secondary site if the primary is inaccessible. As IT brings the on-site physical and virtual servers online, the service fails back to the primary site and continues to restore data as needed.
What You Need to Know to Choose the Right Cloud Based Disaster Recovery Provider
- Understand what you need for disaster recovery.The difference between cloud backup and recovery and cloud failover is that with failover, you replicate server images as well as data to the cloud. Upon failure, the cloud site takes over application processing. If the only thing that you need to do is to store backup to the cloud and restore it in a reasonable timeframe, then your only real concern will be the speed and integrity of data backup and the ease of restore. But if you want failover in the cloud then you’ll need to look for more.
- Pick your best provider. The ideal failover environment should support your operating systems including Windows, Linux, and Mac. They should also support major hypervisors. Data transport speeds are extremely important considerations with cloud-based DR. Look for a provider that optimizes data movement performance to and from the cloud. At a minimum the service should optimize WANs. Some services are also built to quickly move large data sets. The service should also include automated DR verification services so you can be confident of data integrity.
- Prioritize your applications. Every application should have an RTO, but not every application has to failover to the cloud. Assign RTO to each application and negotiate with your provider over service level parameters. Identify your critical applications with RTOs of minutes to a few hours and add these to the failover application list. Business-critical applications may not require failover, but should have service levels attached for fast restore. Less critical applications follow in priority order until the entire primary data center is restored.
- Know how much this will cost you. Failover to the cloud is far more economically feasible than traditional hot sites, but will be more expensive than simple replication to the cloud. Look closely at provider pricing models. Although they differ from provider to provider, most of them bill monthly based on the amount of bandwidth used, the amount of storage within the cloud, service level agreements, and added services including failover. Understand what you are paying for and how that budget could grow as you store more data in the cloud.
A major downtime event does not have to threaten your business if you are prepared for it. Don’t fall into the trap of thinking that you can save money by skipping failover services. Cloud disaster recovery services are a type of insurance, and we all hope that we never have to use it. But if an event downs a critical application — and your company hardly even notices the difference – then investing in a cloud based disaster recovery service will be one of the best decisions you will ever make.