Leveraging AI for Application Modernization
As enterprises continue to move towards containerization, their DevOps teams are knee deep in analyzing their monolithic applications running in their environment. The task of breaking down the current million-line code, which were potentially written in last 7 to 10 years is undoubtedly a daunting task.
In some cases enterprises are fortunate to have good documentation on the application code. However, in many cases no one has a clue why a piece of code exists in some of the complex applications in their estate. Many CIOs believe it is easier to write new code instead of decomposing the current application into microservices. That being said, in most cases, neither of these approaches are found to be more economically viable nor quicker to deliver. This is further compounded by the challenge of finding the right skills and experienced resources to deliver application modernization irrespective of the market maturity. If we talk about the Middle East and Africa, it is apparent that many enterprises are indeed struggling with the availability and access to skilled resources.
IBM has taken a unique approach in order to address these challenges. IBM researchers believe that a good start is the use of AI to analyze the code and merge relevant groupings of classes with data dependencies to generate natural seams-based grouping. IBM’s approach involves the use of a semi-automated transformation utility that assists architects in properly transforming monoliths to full (or partial) microservices, without rewriting the entire application.
Simply put it takes away the need for a highly-skilled resource with the knowledge of both application runtime and micro-service architectures, to bring value to the application modernization task.
I hence would like to introduce Mono2Micro, an AI-driven utility for semi-automatically transforming Java monoliths to microservices. A simple non-disruptive and economical approach. Having said that, while the Mono2Micro tool addresses the challenge of analyzing the app and advice on the possible breakdown of service, there is still a need to transform or rewrite the current code to run efficiently on a container. The current monolith application code is designed to run on VM or bare metal server. However, with a light weight container and stateless nature, some of the code will need to be change. IBM had addressed this challenge as well, and complements the Mono2Micro tool with another critical tool which is IBM Transformation Advisor.
Transformation Advisor, once implemented against your on-premises applications, produces a detailed analysis that includes several reports. These reports help you understand potential modernization challenges and point to where code changes may be required. It also creates migration artifacts for you to get you started with containerizing and moving your application to a cloud native environment. Transformation Advisor also provides a cost estimation and suggests a target environment
IBM tools like Mono2Micro and Transformation Adviser, when leverage will highly accelerate your journey to cloud and App modernization. Most CIOs are challenged by the line of business to roll out changes overnight. Faster time to market is key to success. By modernization of apps and introducing Agile development practices, CIOs can achieve this goal. These tools are part of IBM’s cloud app platform solution IBM Cloud Pak for Applications along with a variety of other modernization tools.
I highly encourage you to dig deeper into these 2 technologies in more details.
· Sign up for the Mono2Micro beta release.
· Read more about Transformation Advisor in IBM Knowledge Center.