By Yusuf Ali, Solution Architect & Delivery Lead – Fujitsu
By Rob Evans, Application Transformation Product Manager – Fujitsu
By Eric Payeur, Director – Head of Application Modernization, Mainframe Migration – Fujitsu
By Serge Moro and Addy Dubhashi – AWS
In the last two decades, the financial services market has been undergoing significant transformation driven by the competition of new market entrants, customer demand, and regulation. This transformation is powered by new technology, including cloud services.
Furthermore, COVID-19 has brought severe disruption to traditional operations and raised the need of accelerated digital engagement and remote working. This reveals even more how cloud technology can support new operating models and services.
With most of their core services running in mainframe, heritage firms need to overcome the technical and commercial constraints they perceive with mainframe application modernization.
In this post, we will detail what those constraints are and describe how Fujitsu mainframe modernization solution allows making mainframe application modernization possible and sensible.
Fujitsu offers a commercial and technical solution for mainframe modernization on Amazon Web Services (AWS), detailed in their whitepaper Releasing the Shackles of the Mainframe Legacy for Financial Services.
Fujitsu is an AWS Advanced Consulting Partner with the AWS Migration Competency. They are also a member of the AWS Managed Service Provider (MSP) and AWS Well-Architected Partner Programs.
In an increasingly strained global economy, financial institutions are realizing the changes COVID-19 has brought to customers’ engagement and business operations.
As leaders re-align their goals for this new normal, 88 percent of senior leaders within financial institutions say technology investment is a top strategic priority for the next 12 months, according to a recent Lloyds Bank survey.
While some will focus on improving their fintech offering through either acquisitions or partnerships, most firms (62 percent) plan to increase their investments in technology and core systems.
The companies surveyed said their technology investment plans were primarily aimed at modernizing systems that improve productivity (71 percent) and customer experience (70 percent). About half (47 percent) invest to increase their resilience to threats and strengthen cybersecurity (46 percent).
As part of systemic modernization of platforms, mainframe migration projects to the cloud are accelerating. As financial services institutions evolve to operate in this new normal, several concerns shape future technology investment:
- Can we re-prioritize technology resources towards the platforms that will evolve our operations for this new normal?
- Do we have the online channels, features, and tools to offer the personalized digital experience our customers expect of us today?
- As customer engagement operates digitally and internal teams collaborate remotely, are we able to automate processes securely and effectively?
Core business logic is still critically dependent on mainframe applications and capabilities that comes with many limitations (in scalability and elasticity, limitations in data ingestion, and processing).
There is also a lack of legacy skills (COBOL developers and mainframe operators, for instance) to maintain and evolve current mainframe-based applications.
Key Challenges of Mainframe Migration
The first challenges are technical, as full rewrites of mainframe applications are complex and time consuming. A mainframe can hold millions of lines of code that could have a decade of patches applied, and for which little documentation exists.
Customers may also have insufficient skills and awareness of modern technology and migration/transformation solutions. This may lead them to believe moving the business-critical application hosted on the mainframe to modern technology and platforms is risky.
There are commercial challenges as well, such as the significant cost of migration due to complexity (lengthy processes, resource intensive). This can make it difficult to forecast the exact investments required.
Lastly, there is a common misperception that mainframes are more reliable, operationally stable, and secure. While mainframe outclasses commodity servers, AWS provides high-quality services with proven security, availability, scalability, and strong system management that make it the right platform to host business critical mainframe applications.
In reality, mainframe-to-cloud migration programs are increasingly common with some of the largest global banks now embarking on this transformation.
Dependent upon the customer’s long-term strategic and short-term financial and operation constraints, a range of well-tested approaches are available for mainframe migrations. The AWS blog post Yes, You Should Modernize Your Mainframe with AWS Using Patterns and Best Practices describes the various approaches.
Fujitsu Modernization Services
Fujitsu and AWS have worked together to understand the core needs of customers and map the future needs of the financial services industry.
Fujitsu modernization services rely on an automated suite of migration tools (Fujitsu PROGRESSION) that allow customers to move off core legacy applications developed on a mainframe/midrange.
Ported workloads are either C# applications running on .NET or .NET core or Java applications running on openJDK, taking advantage of native tools and capabilities of the AWS Cloud. There’s no need of any dedicated middleware environment to host the ported code; this enables full choice of the deployment technology.
The migrated code can be hosted on AWS Elastic Beanstalk as a .NET or .NET core application for C# option, or as a Java SE application. Amazon Elastic Compute Cloud (Amazon EC2) with auto scaling may also be an option in both cases.
With PROGRESSION also allowing to generate Docker containers, Amazon Elastic Kubernetes Service (Amazon EKS) can also be used to deploy and manage those Dockers.
The Fujitsu PROGRESSION tool suite is an automated migration solution that provides highly configurable options using parameters and templates. This allows for implementation of all coding techniques, standards, and packaging.
The use of the automated tool makes sure there’s a high level of efficiency and consistency in the resulting application, while minimizing the risk of introducing human error.
The Fujitsu PROGRESSION modernization process is depicted below and follows several steps.
Figure 1 – Fujitsu PROGRESSION modernization process.
Steps to modernization include:
- Initial source code analysis of the existing mainframe code.
- Automated code conversion of legacy code using PROGRESSION, including:
- Application code to Java or C#
- Display screens to web forms using HTML5, React.js, Spring MVC, and JSON
- Jobs modernization to PowerShell
- General code logic follows recommended coding standards
- Database migration – schemas and data converted and moved to the new database in the cloud. Databases include Amazon RDS for SQL Server, Amazon RDS for PostgreSQL, and Amazon RDS for Oracle.
- Conversion of the scheduler in the cloud.
- Deploy to AWS – fully scalable, and stability through full load balancing support.
The modern three-tier decoupled design of Fujitsu converted components ensures a complete and independent implementation of the presentation, business, and data layer.
The diagram below depicts a typical AWS deployment for a transformed mainframe application.
Figure 2 – Typical AWS deployment.
The data layer is implemented through database objects and a shared data access component, This ensures the possibility of redesigning the new relational database without affecting the data access logic of the current system.
This approach provides flexibility for future maintenance and enhancement of the application. The shared component is automatically generated based on the design of the old and new database.
Batch components such as JCL for an IBM Mainframe z/OS are also migrated and deployed on Amazon EC2 Auto Scaling or AWS Batch. Reports are redesigned using the new database and can be implemented using SQL Server reporting services or other existing reporting toolsets.
Fujitsu’s approach relies on extensive configuration and version management. At every step, a specific version of all components is retained, making it easy to identify all changes that have been applied.
In conjunction with the automated conversion suite, it’s easy to integrate ongoing maintenance releases into the test cycle by using change management tools that allow comparison and the merging of converted components.
The recent COVID-19 crisis has accelerated the transformation of the market and highlighted the need of applications and mainframe modernization for heritage financial services firms.
Modernization is now a practical and cost effective option, made possible by combining Fujitsu’s 17 years development of mainframe refactoring as a service with the innovation, resilience, security, and cost advantages enabled by AWS.
Fujitsu’s end-to-end approach helps financial services address their challenges in a cost-effective manner. This helps organizations realize the required future-proof scalability, agility, and extensibility of their core legacy systems.
To learn more about Fujitsu migration solution, refer to in Fujitsu whitepaper Releasing the shackles of the Mainframe legacy for Financial Services.
Join Fujitsu and AWS for a hands-on Agile Banking Future Workshop where you’ll learn about Fujitsu’s migration technology. You will also learn about Fujitsu’s methodology that includes uncovering existing mainframe workloads, assessing and prioritizing those workloads, and determining the best migration path to create the case for change.
For information about scheduling an Agile Banking Future Workshop, contact Fujitsu.
Fujitsu – AWS Partner Spotlight
Fujitsu is an AWS Migration Competency Partner and MSP that offers a commercial and technical solution for mainframe modernization on AWS.
*Already worked with Fujitsu? Rate the Partner
*To review an AWS Partner, you must be a customer that has worked with them directly on a project.