Data-Driven App Modernization

1


2

Modernization initiatives were accelerating even before the COVID-19 crisis landed. A March 2020 IDG report found that more than two thirds of IT leaders saw IT modernization as a prerequisite for business transformation. Yet the pandemic has accelerated this imperative. IDG’s CIO Pandemic Business Impact Survey, released last year, found that leading digital business initiatives remained the top priority for CIOs, and that increasing operational efficiency was now their number one objective. As enterprises enter recovery mode, they’re looking for technology that enhances their agility and resilience, and that enables new digital operating models. What’s more, they’re looking to end their reliance on legacy technology that inhibits their ability to adapt.

The Complexity Challenge

Of course, modernization isn’t without its challenges. Apps at various stages of modernization still need to work with each other, and apps built using cloud-native technologies can appear harder to monitor, secure, and troubleshoot.

Crucially, the need to monitor and secure a wider range of services and infrastructures, along with the new potential points of failure, all create additional complexity. Therefore application modernization requires observability - the need to have total visibility and insight into the state of distributed systems.

For example, take an application that handles deposits at a bank. Once, this might have involved one multi-tiered monolithic application managing the whole transaction and running from the bank’s data center. A modernized version may be broken down into dozens of microservices, running on public cloud infrastructure, but still communicating with the bank’s legacy technology. This makes the application more efficient and resilient, and enables faster innovation. However, it also creates more complexity and makes end-to-end observability a challenge, in turn making it difficult to pinpoint the cause of any problems. Without the ability to have full-stack, end-to-end observability and security, app modernization efforts can become headaches.

Creating a Data-Driven Application Modernization Strategy

Organizations can manage this complexity through a data-driven application modernization strategy. An effective strategy starts with a full inventory of legacy applications and the performance requirements and dependencies of each, focused on business needs as much as technical demands. Such a strategy helps leaders select the right modernization approach for each application, taking into account how it fits within the business’s over-arching plans. Most of all, though, organizations need observability across their environments, from modern, cloud-native applications to the monoliths still sitting in the on-premises data center. And they need that observability to extend across the entire stack.

Capabilities to Drive Effective Application Modernization

Organizations that are effectively modernizing their applications are relying on a key set of capabilities. The foundation is the ability to take in all data from every kind of application, from modern cloud-native applications built using microservices and serverless functions, all the way to three-tiered monoliths sitting in on-premises data centers. Organizations need to prioritize capturing data from front-end to back-end, and across the whole stack, including real user interactions with an application’s interface, traces of transactions flowing through an application, network performance data, infrastructure and database performance, and more. This comprehensive data set then must be made available to the teams responsible for securing, operating, and innovating.

Yet simply collecting the data isn’t enough. Due to the velocity and volume of data generated by cloud-native technologies, leaders need to provide their teams with the ability to process this data in as close to real time as possible. With near real-time mirror images of their applications, DevOps teams can rapidly see performance bottlenecks and troubleshoot issues when they occur. Real-time capabilities not only prevent devastating customer switching behavior, but also enable well-paid and innovative engineers to spend their time building and optimizing rather than solving performance mysteries. To truly adopt a world class data strategy for application modernization efforts, organizations also expect the assistance of practical and actionable AI/ML that helps prioritize alerts and direct troubleshooting.

When investing in a data strategy for application modernization, a common pitfall is relying on too many tools for specific use cases. Point solutions can tackle parts of the challenge effectively, but they quickly lead to silos and data overload, without a comprehensive plan for managing increasing data feeds and alerts. Organizations need a unified solution to provide end-to-end visibility and the ability to monitor and troubleshoot across their environment. DevOps teams should also demand the ability to instrument their applications leveraging Open Telemetry so they are not locked into specific vendor solutions.

Using Data to Drive Ongoing Optimization

Data is a key driver for success post-modernization as well. For example, data can provide a level of visibility into cost, usage and capacity requirements to help minimize cloud expenses. Contextual end-to-end visibility also helps simplify and accelerate remediation of any security incidents. And by continuing to monitor and optimize their workloads, organizations create a positive feedback loop, continually enhancing performance to deliver a better experience for customers as well as the best tools and services for employees. This information can then help organizations identify the next applications to modernize and which approach and platform would be the best fit, which continues the modernization cycle.

Using data in this way can have a significant impact. Harvard Business Review’s State of Cloud-Driven Transformation report found that 66% of the IT leaders surveyed said that leveraging real-time data analytics (enabled by artificial intelligence and machine learning) was 'very' or 'extremely' important to monitoring and gaining insights across cloud services, applications and infrastructure. All organizations need the ability to leverage any data in real-time, supported by AI/ML and in one unified solution to ensure their app modernization efforts are successful.