Hydra 1303

All News Items

App Refactoring and Considerations

from the Hydra High Council Jul 10th 2018

When we say "app refactoring," we are referring to the process of rewriting parts, or the entirety of, a software application without changing the app's functionality. This may occur for many reasons, such as scalability, deprecated stacks, different resource consumption requirements or limitations reached due to technical debt. Sometimes to take advantage of highly scalable and quickly deployable on-prem cloud environments, old applications also need to be refactored.

Migrating applications from traditional on-prem environments to the public or private (on-prem) cloud without refactoring is often called "lift and shift." While you would get some benefits from the cloud features, you'd also miss out on many. For example, refactoring a monolithic app into microservice architecture would enable you to take further advantages of the flexibility and scalability the cloud offers. Refactoring to leverage cloud services such as managed databases, messaging systems, and PaaS/FaaS systems can make life easier for ops teams and add resiliency.

What exactly needs to be taking into consideration before starting such efforts, you may ask? Below, we cover a few of our concerns and common questions.

  • App Profile -What is the current state and stack? Are you fully aware of all the components of your systems? Do you know all the dependencies? It is always a good idea to sit with your team and make sure everyone is aware of such details.

  • Infrastructure Profile -In order to assess its resource allocation and possible optimization strategies, you need to understand what payloads your applications generate, the physical resources it needs, the level of infrastructure-provided security it requires and how can you optimize your new application to benefit from new architectural paradigms

  • Security and Criticality -Have you done a deep analysis of your security requirements? What about redundancy? Rollback and easy deployment with CD/CI? While this may sound superficial or irrelevant, lack of doing so may actually result in security vulnerabilities or masked points of failure, that may only show when it is too late

  • Customer -What about the user base? Have you considered the usage and the necessary resources to serve your community? There are many differences between refactoring an application for 1000 users versus for 1 million users. Do you and your team have a migration path in place to ensure the refactoring efforts actually support the users and their requirements?

  • Tools -Have you assessed what stacks will be used for the newer application? Have you ensured that all the requirements are covered and there won't be any unknowns/limitations being discovered only later during development? Things like the limitations of your chosen protocols, hardware dependencies and long-term support are a few considerations.

While this list is by no means exhaustive, it surely gives you an idea of the effort you and your team have ahead and also gives you some insight on your current step of the journey. Need some help? Contact us and we can help you through your journey into app refactoring and on-prem cloud solutions.