By Oliver Presland, Vice President, Global Product Management, at Ensono
Microservices – an IT model that breaks down slow-moving, monolithic IT systems into multiple small independent services that are highly decoupled and self-contained – has grown to become a vital asset for financial services companies on their digital transformation journeys. And it’s easy to understand why. With benefits ranging from increased scalability, security and reliability, such IT infrastructure allows financial organisations to move fast, stay proactive and constantly innovate.
So, what are ‘microservices’? Initially devised to solve the inherent problem of monolithic design, microservices breaks up applications into smaller, loosely coupled and independently deployable services. These suites of small services are each designed around different and specific business capabilities.
The model is set to become even more popular in the coming years, with the microservices market anticipated to reach $32 billion by 2023. Even in 2020, many years after the term microservices entered the industry’s lexicon, interest continues to grow according to Google Trends; in February of this year, the search term reached an all-time high with users in the UK.
An example of how microservices have already been successfully utilised can be found in streaming-giant Netflix. Until 2008, Netflix relied on a monolithic model when it suffered an IT outage due to major database corruption. Following this, the decision was made to break apart the application, moving from monolithic to multiple independent services in order to increase scalability, reliability and availability.
This continued growth in interest is understandable, given the success of businesses such as Amazon, PayPal, Capital One, and Monzo. Organisations like these have demonstrated the ability to rapidly scale and deliver value because of their microservices architecture – essential qualities in the financial industry.
Microservices challenges
Nevertheless, for organisations considering the move, it’s important to be aware that the evolution from monolithic infrastructure to a microservices architecture isn’t always simple. Not all organisations get it right and it is not appropriate for every application.
A transition to microservices can be particularly painful for banks: these companies and certain other financial institutions typically rely on batch-based processing, and architecture is largely monolithic – ancient by technology standards. The challenge with these monolithic platforms is that customer-facing applications need to have fast development and release cycles to keep up with customer demands. However, there are complex data dependencies, especially when mainframes are thrown into the mix.
Even if a mobile banking application uses an agile methodology to go to market, core banking architectures can cause delays in the pipeline. As Ensono’s own data shows, 29 per cent of businesses struggle with delays in their deployment cycles. 16 per cent are failing to meet delivery deadlines. Amongst banks, those figures are likely to be much, much higher.
Benefits within reach
If banks and other financial institutions do get microservices right, then there are plenty of advantages to look forward to. Layered on top of core banking, microservices can:
- Enable developers to change and re-deploy software without fear of compromising the core application.
- Decouple release cycles and increase agility, meaning app updates can be developed and deployed in hours.
- More effectively scale for growth in users and transaction volumes, since the infrastructure can be easily replicated.
- Reduce the infrastructure’s failure footprint, since code is much less interdependent.
- Make apps easier to refactor or replace going forwards.
- Increase scalability, since microservices can be easily replicated.
- Offer greater coding language freedom.
- Improve security, since services can be isolated and a security breach does not necessarily threaten an entire application.
These are all considerable advantages for sure, but banks need to ensure that their IT strategy is robust enough to achieve this transition. To run a microservices infrastructure, the whole organisation must align its operations, and other IT units and leaders will need to be pulled into the transition.
Team sizes may need to shrink to accommodate for more agile working practices; DevOps practices and methodologies will likely need to be adopted so that the larger number of services can benefit from fully automated deployments; organizational communication will need to be optimized; and businesses will need to have the tools and processes in place to trace, alert, and recover any faulty services.
Even before all of this, companies need to ask several critical questions: do they really need it, do they have the know-how in house, and is there agreement from all departments. If there are positive answers to all three, then the path ahead may well be one that leads to a microservices model.
