Creating a High-Availability Strategy for a Hybrid World

In my last article on network resilience and disaster recovery, I wrote about the importance of hoping for the best, but preparing for the worst. Enterprises need to have the same mentality when it comes to high availability and redundancy planning. It seems like every time you turn on the news there’s a new threat emerging that could take enterprise IT infrastructure offline. Given the shift to remote work we’ve seen over the past few years—with employees needing uninterrupted access to enterprise systems and collaboration/productivity apps—it’s safe to say the costs of falling victim to an outage have never been higher.

Disaster recovery and high availability are frequently lumped together, and it’s true they both play an important role in the business continuity equation. However, it’s important to understand how they’re different, and how an effective business continuity strategy needs to include both. As the name suggests, disaster recovery is about getting systems back online as quickly as possible in the aftermath of an outage; high availability is about protecting against threats while systems are still running, with the goal of preventing outages in the first place.

The good thing is that if we plan appropriately, we don’t have to worry too much about the things we hear on the news. An enterprise can’t predict the specific nature of the threats their digital infrastructure will face, but it’s a good bet that something will come up sooner or later. That’s why high-availability planning is all about being ready for anything.

Understanding the true impact of downtime

We’ve all seen the statistics about the financial impact network and system outages can create. I won’t try to put a specific dollar amount on how much downtime costs an enterprise, but suffice to say, it’s significant. Beyond that, there’s also the more difficult-to-quantify cost enterprises pay in terms of damage to their brand and reputation: if your outage shows up on the front page of the newspaper, it won’t inspire much confidence in your customers or partners.

Even with the business imperative being as clear as it is, businesses still have significant room for improvement when it comes to avoiding outages. This means that businesses that do prioritize high-availability planning today could create significant competitive advantage over their peers in the future. According to IDC:

“By 2025, G2000 organizations are still experiencing two to three systemic service provider network outages per year, showcasing the importance of added investments in connectivity redundancy and service resiliency.”[1]

Hybrid infrastructure necessitates a hybrid approach to high availability

Enterprise IT architectures are growing more complex all the time, with cloud, colocation, XaaS offerings and various virtual network functions either replacing or complementing traditional on-premises data centers and physical network infrastructure. Businesses are deploying hybrid infrastructures that incorporate these new technologies because hybrid helps give them the best combination of results for flexibility, efficiency, performance, reliability and security across their various workloads. Despite the clear benefits it can offer, hybrid infrastructure also creates new challenges when it comes to high-availability planning.

Because hybrid infrastructures are inherently more complex, they’re also more difficult to protect. Many organizations are blindsided by outages because they didn’t fully understand the vulnerabilities facing each individual element of their hybrid system. Unfortunately, many of these organizations find out the hard way that a threat to any one element of a hybrid infrastructure could indeed be a threat to the whole. So, it stands to reason that a hybrid approach to IT infrastructure also requires a hybrid approach to high-availability planning, pulling together both physical and virtual redundancy.

Geo-redundancy at the edge unlocks a modern approach to high availability

In the past, when centralized IT architectures were the norm, geographic redundancy of core physical infrastructure was really the only high-availability strategy you needed. You had your primary on-premises data center, and you had your secondary redundant site. If the primary site got knocked offline, all operations shifted seamlessly to the secondary site, and applications and data remained available to end users.

Even with the shift from centralized to distributed IT, physical redundancy is still extremely important. What’s changed is where organizations need to deploy physical redundancy. Many enterprises are shifting infrastructure from the core to the edge to get closer to end users and enable advanced applications that require extremely low latency. As a result, they must also start making edge locations part of their high-availability strategy. Maintaining redundant infrastructure in different geographic locations is important both to keep on-premises systems online, and to ensure uninterrupted access to key cloud services.

Geo-redundancy for high availability is a two-step process: getting the right infrastructure in the right places, and ensuring you can quickly access that infrastructure when the need arises. Your redundant sites should ideally be far enough away from primary sites to ensure they don’t fall victim to the same disasters, but they also need to be interconnected to one another in a way that doesn’t create high latency, delays and poor user experience.

Equinix supports the hybrid approach to high availability

With Platform Equinix®, enterprises can gain access to a global network of colocation facilities, making it easy for them to deploy redundant sites where they’re needed most. Equinix International Business Exchange™ (IBX®) data centers are available in more than 65 metros across the world. In addition, software-defined interconnection from Equinix Fabric™ makes it quick and easy to build connections between redundant sites in different regions on demand, as well as connections to the cloud and network service providers that play an integral role in ensuring high availability.

To learn more about how Platform Equinix can support a distributed approach to digital infrastructure while also unlocking modern high-availability planning, read the vision paper today.

[1] IDC, “IDC FutureScape: Worldwide Future of Connectedness 2022 Predictions”. Paul Hughes, Doc #US47438921, October 2021.