Disaster Recovery as a Service (DRaaS) is becoming increasingly popular as a way to ensure business continuity in the event of a natural or manmade disaster. In this post, we’ll cover some of the most common questions we field from customers looking to ensure the availability of their systems.
Q: What is Disaster Recovery as a Service?
A: Disaster Recovery as a Service or DRaaS refers to the replication and hosting of physical or virtual servers to the cloud (including the data and applications running on them) by a commercial hosting provider to enable failover of your primary systems – within a defined time period – in the event of man-made or natural disaster.
To put this in everyday terms, if your primary systems were to go down, your users would be connected to a “failover” site that contains a replica of your applications and data.
Q: How long does failover take, and will ALL of my data be there once we’re back online?
A: There are several ways to approach disaster recovery. To choose the best approach, the business needs to determine its Recovery Time Objectives (RTO) and Recovery Point Objectives (RPO). As you’ll see from the examples below, these can be different for different workloads.
RTO refers to how quickly you need to regain access to your systems and data. For example, a manufacturer might want to minimize downtime for the systems that control production to a half hour or less but decide that their marketing systems can afford to be down for several hours. A rarely used archival database for old inventory parts could have an RTO measured in weeks.
RPO refers to how much data loss you could accept in the event of a disaster. It is easy to confuse RPO with RTO as both are expressed as a measure of time. For example, if you were running a high-traffic e-commerce website, you might set an RPO of zero because even a minute’s worth of data loss could mean the loss of millions of dollars’ in sales. On the other hand, an accounts payable application might be willing to accept a 1-hour RPO as the data lost could more easily be reconstructed.
Setting RTO and RPO appropriately is crucial as they are key factors when deciding which replication solutions will best suit the business, and in turn, how much will need to be budgeted for disaster recovery. Generally, the lower the RTO and the RPO, the more expensive the solution.
Q: What else factors into my decision on which solution I use?
A: Network connectivity is the first piece that has to be in place to have a usable DRaaS solution, i.e., a way for your users to connect to your DRaaS solution in the event your primary site(s) are down for any reason. There are many options including redundant telecommunications circuits to your DRaaS provider, secure Internet Virtual Private Networks (VPNs), Domain Name System (DNS) Failover, and Software Defined Wide Area Networks (SD-WAN) solutions among many others.
Q: Does my failover site have to have the same configuration as my primary site? Meaning, am I essentially paying for two data centers?
A: Not necessarily. As a rule of thumb, companies will often have production, quality assurance, and development systems within their primary environment. Not all of these systems may need to be replicated in the disaster recovery location, and organizations may opt for just those systems which they deem critical for business functionality.
In terms of hardware, generally speaking, there is no requirement to have like-for-like hardware in two data centers. As long as the disaster recovery environment has sufficient compute and storage capacities for those systems deemed mission critical. Often, different vendors and/or architectures can be used with little or no impact to the disaster recovery environment.
Q: What if my primary environment isn’t virtualized?
A: If you still have an environment that can’t be virtualized for some reason, Connectria has DRaaS options for non-virtualized environments too. For example:
Windows – Whether it’s a SQL Server database or an application that needs dedicated hardware, we can utilize several different options including DoubleTake Availability for Windows, SAN-to-SAN replication, Exchange Asynchronous Replication, or SQL Server Always On Availability Group to replicate critical data to another location.
Linux – We have DRaaS options for non-virtualized Linux environments, too, that utilize DoubleTake Availability for Linux, rSync, and SAN-to-SAN replication among other options.
Solaris – We offer DRaaS options for Solaris environments using a number of different options including Oracle Solaris Cluster software and rSync.
For more on DRaaS, click here.