I’ve been seeing FUD (Fear! Uncertainty! Doubt!) articles like this one for years now and while it’s pretty easy to see through them, this time I decided that I just have to say something. The gist of the article, as the title suggests, is that legacy systems are old and clunky, and are holding the banks back and preventing them from benefiting from the modern features that they want: personalization, predicting customer needs, delivering on improved customer satisfaction, etc. (Legacy systems stand in the way of customisation for banks)
To be fair, this article could be aimed at smaller banks that have (very) limited IT budgets, and have mainframe systems that possibly don’t serve them well. It’s true that a mainframe system must have a certain utilization to be cost effective, and in the past few years this has been a big deal. However, that has all changed now since IBM introduced the zEnterprise 114, and other entry-level systems, and changed some of their pricing scenarios. Despite that, some smaller banks may have been best served by getting off the mainframe, and you could argue the case for the smaller banks to do so even now.
Nevertheless, to imply that all banks need to dump their mainframe systems to get into the modern world is definitely incorrect – more than that, it’s potentially misleading, and borders on dishonest. Further, anyone blaming their legacy systems for slow uptake in the technology needed to interpret customer data just isn’t doing their homework. The truth is that both IBM and third-party mainframe ISVs have offerings today that address all of the apparent shortcomings mentioned in this article.
Both “generation selfie” and the internet-enabled population have been doing online banking for years using web-based applications and mobile devices connecting directly to their banks’ mainframe systems. This capability has been handled by IBM solutions and third-party solutions like those provided by Compuware, CA, and niche vendors like those of the Planet Mainframe Alliance.
Today, several large banks can generate single customer views using existing COBOL applications, leveraging mainframe-based in-memory technology and multi-platform data replication techniques; others do it using newer Java, JSON and .NET applications that seamlessly access their mainframe-based databases using advanced integration solutions.
Many solutions exist that allow legacy systems to connect to data provided by connected devices, whether they are mobile, IoT, or anything else. IBM has offered this for years with several products, and third-party vendors also have very competitive solutions. In fact, IoT, the most non-mainframe-seeming application yet, interacts directly with the mainframe. IoT sensors can detect changes to an established trend, and a mainframe system can apply transactional knowledge to determine what happens next. IBM has had those types of solutions for a while now, with the mainframe and their IoT cloud-based offerings. The solutions are out there to be used, one just has to look for them.
When I say that the solutions are out there, the response might be “Yes, but at what cost?” which is a fair question on the surface. But if you start looking at the real costs involved in forklift infrastructure replacement projects, and the ongoing costs of “newer” infrastructures afterwards, it should shock you into looking a lot closer at leveraging what you have now. And I’m not at all saying “if it ain’t broke…” The mainframe-based systems being used today by the biggest banks are actually at the cutting edge of both transaction throughput performance and new technology adaptability.
As I said, the message of the aforementioned article is fine for small banks that cannot justify their mainframe systems or some of the required upgrade activity. However, if third-party mainframe migration specialists want to attract business from larger banks, they really should avoid using such transparent FUD, and at least look for FUD that is more convincing to the CIOs and IT management at those banks. Conversely, if an IT professional at a bank can’t find ways to leverage the current IT assets, and believes that the only road to progress is to first burn everything to the ground, then maybe they’re not qualified to make these decisions.
Regular Planet Mainframe Blog Contributor
Allan Zander is the CEO of DataKinetics – the global leader in Data Performance and Optimization. As a “Friend of the Mainframe”, Allan’s experience addressing both the technical and business needs of Global Fortune 500 customers has provided him with great insight into the industry’s opportunities and challenges – making him a sought-after writer and speaker on the topic of databases and mainframes.