I’m not sure who first started using the word ‘legacy’ to describe older, usually mainframe, systems. But I think it’s misleading. For many companies, this technology is still very much part of their day-to-day processes. It’s their business engine packed with mission-critical data.
Spending cuts are one reason so many of these systems are still in use - but only one of many. Perhaps the most important reason is that they are usually highly reliable and play a pivotal role in the enterprise. Often, they have been tailored to fit the business and represent major investment over the years.
These systems are often unfairly depicted as lumbering giants. True, they may have been written in old programming languages and rarely have entry points or APIs, but they are still highly valuable to many operations. For most companies, legacy systems represent both a blessing and a curse.
So the question is - does a company stay with a system which represents a major investment over the years and a backbone for current business processes, but lacks the agility and openness of more modern software? It’s a difficult decision, especially in highly competitive industries such as financial services where younger companies are able to set up without the ‘baggage’ of established systems and processes and so get their new products and services to market faster. The challenge for IT managers is to find a quick and easy way to make established enterprise applications meet these demands.
For many businesses the situation has come to a head with the mobile revolution and the need for lighter and more intuitive, web-based applications. But while as consumers we love new mobile apps that enable us to do things we never believed possible, in business we need them to enable us to do what we always do, but while on the move or away from the office. In other words, although we all, ideally, want to be able to use our own chosen tablets and smartphones, we want to use them to securely access existing business applications.
Unfortunately, this is not always easy as critical mainframe or legacy host-based applications are notoriously adverse to integration. But there are solutions that can either:
- Generate new user interfaces for any HTML 5-compatible device. This means employees can use their tablets or smartphones to see exactly what they are used to seeing on their desktops - with no need to develop any new applications.
- Create customised mobile front ends. This approach provides the control to create a whole new application UI and even modify the associated workflow, or
- Build service-orientated architecture (SOA) style services directly out of host-based applications. This means you can create entirely new mobile applications from current enterprise assets without the need for modification. This is quick and low risk, while giving full control over every aspect of the mobile application
Although mobile apps are the first thing to spring to mind there are other just as important reasons for modernisation; for example, the need to integrate customer account information to ensure smooth customer service. Too often, the required customer data is hard or impossible to access because it’s held on legacy systems.
There are also three options here. Old applications can be re-written so that they can talk to newer technologies but this is expensive, time-consuming and error-prone. Older applications can be replaced entirely - but once again this can prove an unwieldy and risky effort. Not least because it’s easy to underestimate how much a business relies on certain legacy functions.
The third, service enablement, is an advanced approach that is gaining in popularity. This addresses the underlying issues tied to legacy systems, enabling businesses to reuse applications and databases already there while leaving functional legacy logic in place. As a result, they avoid the risks of recreating critical functionality and mitigate concerns about Quality of Service loss.
With this approach, application components are wrapped and exposed for re-use. Current technologies can then plug into and leverage legacy assets. This enables the extension of previously locked up host applications, making them agile and accessible across the enterprise.
The key to the success of this approach lies in making each project small and working at an application level rather than a systems level. This way, a business can pursue a series of projects that will each start to reap return on investment upon completion. Taken as a whole, it’s a strategic process known as rapid service enablement.
In essence it means small, risk-free steps to achieve a larger goal. This is where rapid service enablement has great value; it’s not an all or nothing effort and the work done now can be reused at any time to complement other solutions. The incremental approach means applications can be successfully modernised without disrupting an entire IT infrastructure, helping diffuse the problem of high, up-front costs. Decision makers are more likely to respond favourably if solutions are presented in smaller staged terms and legacy assets are left undisturbed.
Modernising doesn’t necessarily mean starting all over again and new doesn’t always mean better. Putting a modern interface on existing systems is quicker and far more economic - and offers the best of both worlds.
Posted by Barry Davis, Attachmate
Find your next job with computerworld UK jobs