Deploying Offsite Disaster Recovery Infrastructure at the Digital Edge

How customers simplified and strengthened strategies for bringing workloads back online quickly, avoiding downtime

Ian Botbyl
Deploying Offsite Disaster Recovery Infrastructure at the Digital Edge

As organizations migrate applications and workloads to public clouds, they need to deliver those applications in a secure and compliant fashion across a resilient distributed digital architecture. Public cloud services can fail, which makes disaster recovery critical to maintain the 24/7 operations of a business.

To build an offsite disaster recovery (DR) strategy, organizations must determine how much they’re willing to spend, in terms of both time and money, to achieve their desired DR outcome. Because a disaster event can potentially take down workloads and/or applications, DR objectives should include seamlessly recovering applications/workloads or avoiding downtime altogether.

While backup and recovery options may seem straightforward, the reality is that preparing for a disaster recovery event is daunting. However, you can create a solid plan in a few short steps if you invest some time up front. In this article, we share how two Equinix customers created DR strategies to deliver the best outcomes for their businesses.

AdTech company chooses a network-centric approach for its DR solution

The IT team at this publicly-traded, advertising technology company started their strategy discussion because they recognized the need for a DR solution in the event of a full-scale data center outage. In addition, they were running outdated hardware in an Equinix International Business Exchange™ (IBX®) data center. DR on the East coast was a priority. Initially the team wanted to go all-in with a public cloud, but quickly learned it was too expensive to deliver the promised benefits.

Another inhibitor to a full-scale public cloud deployment was the inconvenience of having to constantly scale services up and down. When the architecture team and their technical champion started to evaluate a leading cloud service provider (CSP) for their deployment, they were offered the following four options, starting with the least costly and complex option, and becoming increasingly more expensive and complex.

  • Simple backup and restore: minimizes data loss, but restoring data may take hours or days after the DR event occurs due to cold storage retrieval.
  • Pilot light: replicates data from one geographic region to another and provisions a copy of the underlying application infrastructure. Resources such as servers are only turned on for testing or failover. While there is some downtime, workloads are back online quickly—in minutes to hours—depending on how much data needs to be replicated.
  • Warm standby: maintains a scaled-down version of the production environment in another location. The downtime is minimal—typically a minute—because the workloads remain functional in the second region.
  • Multi-site active/active: runs workloads in multiple locations simultaneously, ensuring little or no service interruption. While it’s the most complex and expensive option, it can bring recovery times down to near zero.

Redirecting DR solution option to Bare Metal as a Service and virtual interconnection

After considering the cloud DR options, the team shifted their focus to using Equinix Metal™ automated Bare Metal as a Service instead of a major CSP, and Equinix Fabric™ software-defined interconnection to achieve a multi-site active/active deployment.

Further, the team recognized that hardware capacity, traffic needs, and application dependencies were expanding and changing unexpectedly. These factors not only affected the architecture of the deployment; there was the potential for budget overruns from data egress and storage costs and increased latency.

What started out as a DR discussion with their Equinix account team expanded to a broader goal of increasing compute power and lowering latency for their primary ad exchange application. Led by their technical champion they adopted a proactive network-centric approach, and the customer became more comfortable with running their own hardware.

The customer spun up their own Equinix Fabric ports and adopted the proposed architecture without hesitation. By running the ad exchange app with CentOS on Equinix Metal in colocated infrastructure over Equinix Fabric, the customer achieved their business goals.

The customer found these key capabilities from Equinix to be crucial to their business moving forward:

  • Using Equinix Metal as an extension of their colocation environment
  • Starting the move toward more of a hybrid IT deployment
  • Improving the engineer experience with networking and hardware automation
  • Avoiding “virtualization as a penalty” with high performance
  • Making it possible to run multiple applications on Equinix Metal in the future
  • Having the option to change their deployment and continue improving the design

Automation with global software-defined interconnection enabled the digital transformation of their offsite DR strategy. This organization can now store its data at the digital edge, with storage located privately “next to” rather than “in” public clouds. The proximity of workloads to public cloud services combined with dedicated connections provide low latency, improved performance, and a high level of security. Other enterprises can benefit from following this model for the elasticity of cloud services while maintaining complete control of their most important asset—data.

Regional healthcare organization deploys flexible, cloud-like DR solution

A regional US-based healthcare organization was challenged with finding a third site within its existing environment that could handle a Tier 1 workload, plus general files for disaster recovery. Originally, the organization’s IT team searched for an on-premises compute and storage solution for DR. However, they didn’t want to sacrifice performance and security that would occur by shifting to a hands-off, public cloud model, which was the only other available option. Since Dell Technologies was a key supplier, the customer requested that Equinix and Dell collaborate on developing a solution to solve their needs.

Together, Dell and Equinix created a scaled-out storage solution adjacent to Equinix Metal to leverage its cloud-like services. Through this partnership the customer moved its workloads on to Equinix Metal and deployed Dell Isilon in the same Equinix IBX data center. This solution provided a cloud-like capability and scale using the functionality of products already used by the customer.

Equinix Metal’s flexible, pay-per-use consumption model along with a long-term usage commitment met their budget and use case requirements. The customer was able to maintain ownership and control of its data without sacrificing performance. Deploying its Dell Isilon storage within an Equinix IBX data center while privately connecting via Equinix Fabric within the same facility as the Equinix Metal service, ensured super-fast recovery of their workload and general files.

The customer found these benefits worked in their favor for maintaining a sustainable business model for DR.

  • Moved to an OPEX model instead of CAPEX with pay as you go pricing
  • Paid for actual storage used versus investing upfront in hardware
  • Eliminated hardware management
  • Minimized storage costs
  • Scaled flexibly when necessary

Implement a DR strategy today to future-proof your business

Every business requires a faster and easier offsite disaster recovery strategy to maintain business continuity. Nonetheless, a DR plan doesn’t have to be built around a single approach. Applying the same DR strategy to all workloads would likely be cost prohibitive and restrictive. Some workloads need a higher level of protection against downtime than others.

As an organization, choose DR strategies that reflect your priorities and cost preferences and adjust them over time. The first step is understanding the business requirements for your workload so you can choose an appropriate DR strategy. Then, design architecture that runs on Platform Equinix®, to achieve the recovery time and recovery point objectives that meet your business needs while mitigating or reducing risk.

Deploying disaster recovery infrastructure at the digital edge helps enterprises provide consistent application protection for on-premises and cloud-based applications/workloads. This digital edge infrastructure is woven into Equinix Fabric, which privately, securely and dynamically connects distributed infrastructure and digital ecosystems globally on Platform Equinix. This virtual connectivity allows enterprises to orchestrate control of traffic and security across all IT platforms and meet the needs of a distributed workforce.

To learn more about how Equinix helps customers deploy digital infrastructure that supports your offsite disaster recovery strategy and architecture, read the Platform Equinix Vision Paper.

You may also be interested in reading:

Equinix Metal Disaster Recovery Planning



Subscribe to the Equinix Blog