Cloud Chronicles, Part 3: Growing the Reach – Strengthening the Roots
Let’s explore an idea: a cloud-based DNS service to help app delivery across a globally distributed network. What could that service look like? It’s fairly obvious that users demand fast, secure, persistent access to their apps, and that the teams delivering these apps are dedicated to providing such access. And while this sounds like a simple concept on the surface—especially when considering something as foundational as a DNS solution—it gets very complex, very quickly, when identifying everything that goes into delivering apps in this manner.
The pressure to meet these requests is real for teams who support applications. Users exhibit notoriously short attention spans when it comes to application load times, and are likely to abandon a site or application if it takes longer than three seconds to load. This is why it’s critical to consider how the performance of an authoritative DNS can directly impact user experience. Consider the homepage of a popular website like cnn.com or mlb.com. It takes dozens of DNS requests to properly display everything on those sites. If those requests are 5–10% slower than normal, the user experience will suffer noticeably and may drive users elsewhere. Because of this, the tools needed to deliver these apps must work and work reliably.
But what’s the answer if the resources aren’t there to manage an on-prem, hardware-based DNS solution? Or, what if that app is a crucial revenue driver, but its present traffic demands don’t merit a full-scale data center build-out to support delivery? Tapping into a cloud provider is an option, but there are enough bad experiences with billing surprises from traffic spikes to make anyone on a budget think twice before jumping into a cloud solution without first doing their due diligence.
This presents an interesting challenge: how can a cloud-based DNS solution boost the availability, reliability, security, and performance of an app without breaking anyone’s budget at a time when security risks and user demand have never been higher? It starts by checking a few boxes.
Availability demands DNS scalability
Application availability means being ready and open to receive traffic. It means users don’t have to wonder about if they can reach the application or site that they need, and when they get there, they don’t have to wait an inordinate amount of time for that application or site to respond.
Everything from accessing an application to sending an email or launching a video call starts with a DNS request. When dozens, hundreds, or even thousands of users want to access the same application at the same time, keeping that application online and available can be tricky. To do that, the teams that manage these applications need tools that can scale with demand to prevent waves of traffic from rendering an app unreachable. This isn’t always easy though, when considering factors like hardware limitations, resource usage, and even cost spikes that correspond with traffic spikes. A consistent solution that scales with application traffic should be on every NetOps or DevOps team’s wish list to ensure that there’s never a question about whether an application can handle a traffic rush.
Resiliency and presence
It’s well accepted that beyond just making sure an app is available, teams need to ensure that those apps can stand up to attacks, outages, and other events that could knock a crucial service offline. This means developing an app delivery strategy that avoids single points of failure. When teams leverage a diversified, globally distributed network, one hosting dozens of points of presence, users shouldn’t notice much of a difference at all in their application experience, even if the original instance of the app they requested is knocked offline.
In a single-point-of-failure scenario, any issue in the central server could lead to a complete service outage. An application fronted by a cloud service that lives on multiple points of presence, however, enjoys a higher level of resiliency. If one node goes down, traffic reroutes to the next nearest operational node in the network, maintaining continuous availability of that application service. This capability is especially crucial for disaster recovery and business continuity, ensuring that regional outages do not disrupt an entire service.
There is also a significant security component to consider with respect to this setup. A network built with distributed, decentralized nodes can better absorb and mitigate volumetric attacks by spreading attack traffic, protecting core infrastructure. Each node can also implement region-specific security measures, addressing local threats more effectively and adhering to regional compliance requirements.
Built on solid foundations
The infrastructure underpinning an application network also deserves to be built on the highest-quality, highest-performing, most secure foundation possible. A house is only as strong as its foundation; an application delivery strategy is no different. Building and hosting a critical application but delivering it with an unreliable, low-performance network means that application is already behind the curve before it even goes public.
Teams that support these critical applications deserve an infrastructure that is up to the task of handling all manner of traffic (good, bad, or otherwise). They deserve infrastructure that not only supports their applications but works as a value-add, giving them visibility into critical traffic metrics as well as the ability to scale to meet demand.
The F5 difference
To support its cloud-based DNS offering, F5 offers Distributed Cloud Services customers access to a private, globally distributed network. The distributed nature of this network enables immense scalability by default, with traffic reaching its destination via a backbone of 26 regional edges that can streamline up to 15+ terabytes of traffic and surge mitigation capacity.
These regional edges allow F5’s Distributed Cloud DNS solution to support applications as close to their requesting clients as possible, enabling users to access content with minimal latency. What’s more: security services are deployed across these locations as well, so that any security solution a user deploys will activate across every regional edge location. And with Regional Edge Pinning, teams may select specific regional edge locations to deploy on if needed to comply with privacy or governmental regulations.
Through F5’s Distributed Cloud DNS solution, teams can leverage the core technology that has driven F5 products for years, now on a global platform, without needing to worry about excessive resource usage or unpleasant surprises from traffic spikes. If you want to find out how your team and customers can benefit from a robust, distributed, global app network, contact us today.
Learn more about F5’s Global Network.