The Cloud is Where Financial Services Firms Want to Go, LaunchDarkly Accelerates The Journey

Traditional financial services firms (banks, credit unions, insurance houses, etc.) have known for years that harnessing the speed and agility of the cloud would be necessary to better compete against digital-native fintechs. But outside of progress made on the consumer-facing side, much of the industry’s core operations remains woefully antiquated. Leading financial services firms have only moved 8% of their overall workloads into the cloud according to a recent Accenture global survey.

So what’s been holding the industry back?

A common reason is that legacy applications and infrastructure are too complex for traditional “lift-and-shift” cloud migration strategies. These monoliths weren’t just built on outdated languages, they are also calcified with decades of business logic written directly into the code. And given how dependent current operations are on these systems, the need for zero downtime, zero data loss, and zero performance degradation further complicates the process.

With so much at stake, financial industry IT leaders need a solution that’s powerful enough to neutralize the enormous risks that come with cloud migration projects. And this is where LaunchDarkly steps in. Having proven the ability to consistently facilitate successful, zero downtime migrations, LaunchDarkly’s feature management platform has emerged as an indispensable tool for cloud migration teams.

And we do this by enabling three powerful, risk-mitigating capabilities:

1. Gradual migrations through granular traffic control + performance visibility

Taking a step back, Feature flags are a software risk mitigation tool that can instantaneously turn a select feature or application 100% on or off at runtime, without the need for an application reset or deployment of new code (think — a light switch for code in any environment).

Feature flags are not a new concept, but Launchdarkly’s pioneering feature management platform meaningfully expanded their functionality beyond pure on/off, by enabling granular control over what, where, when, and to whom software or application traffic is exposed.

This newfound control has transformed a variety of application modernization strategies; the evolution of the Blue-Green deployment method is a perfect example. Historically, these binary deployments were considered an improvement over previous migration methods because traffic could be returned to the original application if something went wrong. Despite this advancement, Blue-Green deployments remained fraught with opacity. When failures occurred, the all-or-nothing, “push-and-pray” nature made it incredibly difficult to determine whether deployment process issues, release issues, or some combination, was the cause.

Today’s Blue-Green deployments powered by LaunchDarkly deliver a controllable process in which precise amounts of user traffic, requests, and data, can be gradually migrated between specific cloud and legacy endpoints, all while performance monitoring occurs. This gives migration teams command over when, and how workloads are migrated, while also being able to accurately measure the impact of those workloads on response times and error rates across multiple applications.

And if something goes wrong, systemic risk is instantly neutralized by:

2. Automated Kill Switches + “failing” back into the old environment

In the old world, migration errors or unexpected performance degradation could easily put a business at risk by shutting down a new application, or worse, taking an entire system offline. Feature management puts those harrowing experiences in the rearview mirror by activating feature flags as kill switches. If something goes wrong, switches can automatically, instantaneously revert some, or all traffic and data back to the original, stable environment.

Combining granular traffic control with automated monitoring and kill switches delivers safety and reliability to migration teams in ways that previously did not exist. In less time than it takes to blink an eye, changes can be automatically made that avoid any service disruption or data loss. And as soon as issues have been resolved, the same gradual release process can be reactivated using the same feature flags and automated monitoring capabilities, until a fully stable migration has been achieved.

3. Continuous Modernization

But this isn’t the end of the story. The process of digitally transforming legacy systems throughout the financial services industry will take many shapes. There will be scenarios where individual components of a technology stack will be modernized, but must continue to coexist and interact with legacy systems. There will be scenarios where a hybrid cloud or multi-cloud approach is strategically optimal. There will be application refactoring scenarios en route to microservices and containerization. And there will be scenarios where one cloud application or service must be replaced by another (read about how we use our own platform to flawlessly execute our own cloud migrations).

As needs evolve, another important way that the LaunchDarkly platform delivers value is in the enablement of Continuous Modernization. Each of these scenarios (as well as many others) will benefit from the same risk-mitigation capabilities offered by the LaunchDarkly platform to ensure consistently successful outcomes.

This is the promise that feature management delivers, and the reason why financial services IT leaders should leverage LaunchDarkly to accelerate their cloud journeys today, as well as their application modernization projects of tomorrow.