Dramatic events like the Covid-19 pandemic and the widespread power outages in Texas in February 2021 highlight what many IT professionals already know; many businesses’ disaster recovery and business continuity plans are sorely lacking. Businesses need disaster recovery plans that will ensure they remain operational even in extreme situations when an entire region loses power or physical access to offices and data centers is cut off. Creating this kind of resilience usually means incorporating the cloud into disaster recovery.
Disaster Recovery as a Service (DRaaS), is one of the most common use cases for organizations starting to explore the cloud. It allows an organization to back up and recover data and IT infrastructure in a third-party cloud computing environment, rather than to a backup server of their own. Like other “as-a-service” models, the cloud infrastructure provider handles management of the resources used for the backups and charges the customer based on usage. Cloud-based DRaaS can offer several benefits over on-premise DR, but there are several considerations that organizations should keep in mind when setting up their DRaaS program. Let’s go through these in detail.
- Cost savings – Like most software-as-a-service models, DRaaS reduces costs significantly for most customers. Since users pay per hour of use, they can reduce resources to only the level necessary to replicate data to the backup server under normal conditions. Then if disaster strikes and they need to switch key workloads to the backup server, they have the computing power necessary to do that. This is sometimes called the “pilot light” model, because usage will be low most of the time, and only “turn on” when it’s needed. Compared to purchasing and maintaining one or more backup servers, this model typically saves an extraordinary amount of money. This includes the salaries of any new hires needed to manage and maintain the backup servers, as well as the cost of the hardware itself.
- Geographical distribution and resilience – Co-located backup servers are also vulnerable to the same natural disasters as the primary server. A cyberattack or an application issue may only affect the primary server, but an earthquake, tsunami or widespread power outage could knock out the backup as well. Power outages or lockdowns like those put in place during the early days of Covid-19 can make it difficult or impossible for companies to access their servers. These issues are especially pressing for SMBs that house their servers at headquarters rather than in a data center. For these companies, hyperscale cloud providers will likely offer better protection against power outages, cyberattacks and natural disasters, so moving backups to the cloud ensures they’re accessible even if physical access is cut off.
- DRaaS allows organizations to place their backup servers in different regions than their primary service. Geographic distribution makes them more resilient to disasters that affect large areas. This is often required by security compliance standards and is a significant concern for many companies considering DRaaS.
- Ease of use – Because DRaaS providers manage most of the DR process on behalf of their customers, they take on the responsibility of finding the proper hardware, locating an appropriate data center, setting up and managing the backups, etc. This can be a significant issue for some companies, especially if they don’t have employees with the right skillsets or time for these tasks.
Different companies have different needs for their disaster recovery, so it’s important for each organization to customize their arrangements as much as possible to help ensure business continuity. Here are some considerations for organizations that are evaluating a DRaaS program. Ultimately, the goal is to determine if the cloud is the best fit for their needs and what parameters and details should be included in their contract with the service provider.
- How much data do we have and how often is data changing? These answers help frame the network and capacity needed, along with the tool that will be utilized to the best effect.
- Do you need geographic distribution for your backups to make them more resistant to natural disasters or for regulation/compliance reasons? On the other hand, is your organization subject to data sovereignty laws (like GDPR) that require customer data to stay in the same country? These considerations affect which data center to place your backups in (and possibly your choice of vendor depending on where their data centers are located).
- Do you have the required expertise on staff, or should you consider working with a partner or managed service provider? If so, do they specialize in DRaaS?
- What is the targeted recovery time window and how much data might be lost in a disaster event per second, per minute or per hour? Some organizations (like online sales platforms) need their operation to be back online with minimal downtime after an incident. Others need to make sure all of their data is captured but can afford to wait several hours or more before their backup system comes online.
- What clouds are your traditional applications compatible with? Many traditional applications built long before the cloud (often in operating systems for IBM Power, such as AIX and IBM i) can’t simply be moved to the cloud without rewriting or rearchitecting. This effort can entail a lot of extra work and cost if you don’t work with a cloud provider that has support for moving these applications to their cloud, as-is.
With all these considerations in mind, DRaaS makes sense for many if not all customers. It’s typically a lack of understanding, confidence or the right skillset that holds companies back from trying the cloud. In many cases, organizations are understandably reluctant to do anything that could negatively affect long standing traditional applications that their business depends on! In this scenario, DRaaS can be a “toe in the water” for companies to experiment with the cloud without significant risk. By taking such a step, organizations will get a better understanding of how the cloud works, how these per-use cost frameworks are beneficial and build up a level of comfort before considering migrating other workloads that could benefit from the cloud.
George Stamos, Director of Strategic Alliances and Solutions Architecture, Skytap