Case Study: SWIFT® Infrastructure and Application Upgrade for a leading bank in Michigan
The client for this project is a leading regional bank headquartered in Dallas, Texas, strategically aligned into three major business segments: the Business Bank, the Retail Bank, and Wealth Management. It has retail banking operations in Texas, Michigan, Arizona, California and Florida, with select business operations in several other U.S. states.
The bank wanted to upgrade the SWIFT application which was hosted in a 30 year old UNIX server infrastructure and ensure the application does not go out of support. The SWIFT application has to be upgraded and cutover to the new system within a minimal downtime and cannot afford to be down for a longer time, which could lead to loss of reputation and also compliance breaches.
The bank was using SWIFT application for all their customer transactions involving international payments. The application had not been updated with vendor patches for the past few years since the hardware did not support the latest updates, which left the application outdated and vulnerable to external threats. The vendor had given an ultimatum for the end of support for the application version being used and the internal teams were unable to upgrade the software. The client had a little more than a year to upgrade the system so they do not run out of support.
The infrastructure was more than 30 years old and the application was running on UNIX servers. There was no documentation on several scripts running on the system and most of the operations were being managed by a support team member, who was also managing the infrastructure for several other applications, who had limited availability for the project team.
The SWIFT application was at the heart of the payments department and had approximately 25 upstream and downstreams applications that were impacted whenever a change was made. All the other applications were owned by different departments within the bank and had a big list of stakeholders that had varying interests in different areas of the bank.
There were also other big projects running within the bank that were using shared environments with SWIFT application, which made it more difficult to come up with a fixed plan especially for testing and deployment, since they were inflight projects scheduled to use the existing legacy swift environment.
The project had several workstreams, as in upgrading the servers, the SWIFT hardware upgrade (HSM) and application development. Although they were different workstreams, they had several inter-dependencies in terms of requirements, resources and progress.
Our assigned Program Manager immediately identified and met with all the key internal stakeholders at the bank and started to work with the solution architect to come up with a high level project plan and estimates.
A high level schedule was prepared based on the high level requirements available while a Business Analyst was working in parallel on putting together the detailed requirements. The high level plan was shared with the steering committee and communicated to the other projects running within the same portfolio that could be impacted or impact the SWIFT Project.
Resources were gathered and the team was built based on the high level schedule, without wasting any time, and the project schedule and estimates were done again once the detailed requirements were available. The program was split into three different workstreams and managed in conjunction.
The key for the success of the project was to gain the support of the senior management across various departments within the bank and the project management team worked on ensuring that the support was available by educating the various teams on the importance of the project, the timelines and negotiation to sort out risks of overlaps with other projects and programs from the beginning of the project.
System testing was conducted with all the approximately 25 upstream and downstream applications that interacted with SWIFT.
The RHEL test Environment was built at the very beginning and all other impacted projects that were going live after a particular deadline were asked to test in the RHEL test environment, so they were not testing against the existing legacy system that was going to be decommissioned.
The project was delivered on time and within budget as per the client’s quality standards.
The legacy UNIX servers were upgraded to a RHEL platform which allowed the SWIFT application to be upgraded to the latest version. The HSM modules were upgraded as a sub project within the same program.
A full round of DR Testing was conducted successfully to ensure that the Active-Active configuration worked as per expectation with the participation of all the teams involved in the event of a DR situation.
A solid implementation plan was created with the participation of all the different teams involved and the team was well prepared for the Go-Live weekend. The cut-over was planned to be conducted during non-business hours and completed successfully with no interruption to business operations.
The project team transitioned the required information for the customer’s support team with the required documentation and training. The transition was a process that began during the execution phase methodically to ensure that the sup[port team had all the information and knowledge to run the new systems smoothly.
The project team was available on call and were prepared to be on-site, if required, during the warranty period to make sure the client had all the support with their upgraded new SWIFT system.