In our previous post, we talked about the top IT pain points of the organizations. One of the most prominent ones is the reliance and lack of progress on innovation because of legacy systems.
Legacy systems, which form the backbone of many enterprises are holding organizations back from leveraging new digital technologies and creating new experiences for their customers. Many organizations find it difficult to create an IT strategy that supports the quick adoption of newer technologies such as cloud, big data, IoT, and mobile with reliance on the legacy systems.
Existing legacy interfaces, developed in a world of daily batch calls, are not fit for purpose for today’s digital channels that require real-time data.
There are two approaches to work with Legacy systems
- Re-platform Legacy systems – to newer modern architecture and ecosystem. While this approach might reap several benefits, it is expensive and it is difficult to keep the business running on the old platforms while rebuilding a parallel modern product. Usually, the way proposed for this kind of transformation is to start converting portions of the legacy system and make it work with modern architecture.
- Enable point to point integrations between collaborating systems – With this approach, if sales force needs to talk to the payments system then we create a point to point link between the two. This approach works well when the number of legacy systems is less. However, for most enterprises, they have scores of systems which need to collaborate and this approach breaks down very quickly.
Another approach which is viable and has led to success with various organizations is an API led approach. This approach abstracts the legacy systems by exposing either the database or business logic of these systems as an API. If you have a project that requires the use of the Mainframe, for example, a reusable API that exposes the information contained in that database means that it is accessible to anyone who needs to use it, without creating a custom-coded point-to-point connection.
With this approach, the old systems can be wrapped with API endpoints and new systems can be plugged to work with the legacy without a lot of changes to the underlying systems. This makes retiring the underlying legacy systems easier as they have been abstracted out. Further, innovating on top of the abstraction is much easier.
There are 3 levels of APIs which are generally helpful for exposing the legacy ecosystem functionality
- Systems API – This is the first level of API wrap. This is the direct wrapper over the legacy systems without any transformations. Some legacy systems like SAP, SalesForce, etc might be exposing their own APIs, others like databases, mainframes might need a wrapper to built with Functional Microservices.
- Process API – This is an important layer of the API ecosystem. In this layer, we would take care of
- Orchestration – combining results from several underlying system APIs into a composite result.
- Data and transport transformation – converting from legacy format to something that modern digital systems understand. Like EDI to JSON. Similarly converting from FTP to internet-based formats etc
- Business delegate – Acting as a business delegate for exposing business specific errors and other information to the consumer of the API
- Experience API – These are the APIs which are built on top of the process API for a very specific product which the organization now wants to innovate on or bring out to the market. The experience API would take into account a composition of process APIs. They might have an orchestration role to play at this level as well.
Knoldus case study
Knoldus has been driving several legacy modernization engagements with our Reactive and Fast Data ecosystem.
At one of our major healthcare clients, where 4+ million senior patients are taken care of annually and are a part of the aging journey. Whenever a health journey professional saw a patient, they had to look through an average of 7 applications to understand their aging record. Additionally, with the hardware integration of their systems with external systems, there were further integrations needed to get the validated results back into the mix. External partners had no visibility into the internal systems and hence could not innovate their product offerings.
Knoldus laid down the legacy modernization road map with the help of the business units. We wrapped the legacy systems with reactive micro-services based endpoints. These endpoints were exposed both to the process API and were needed to other departments of the health care company as well as the external vendors. The process API was further released in the same fashion along with strong ACLs and rate limiting and auditing in place. This required breaking the silos between both legacy and modern medical and non-medical data, applications, systems, and devices.
Today, with the modernization, health care professionals are able to create a care circle for the patient. This allows them to innovate on the information which is available through various silos and build a unified lens for the senior. Further, since a big chunk of senior data is now available, the client is able to run various machine learning models on the data to extract maximum benefit for the senior.
Another similar case study is a huge transformation exercise that we have conducted for a major cruising client where the guest satisfaction on the ships has gone up by a factor 3x.
If your organization is on a modernization journey and would like to create a roadmap to solve the challenges with legacy silos, we would like to talk to you and walk you through the successes that we have delivered in your domain.