We are entering an era in which legacy systems are legitimately seen as somehow “sacred” under the influence of a number of inter-related drivers:
The explosion in data volume: It is in this area that we clearly see that not all database management systems are equal. The “old” DBMS such as DB2 or Oracle have proved their worth, yet the more recent ones such as Maria DB or Mongo DB still have everything left to prove. Once again, senior executives who operate legacy application estates are increasingly facing a classic risk management situation. Minimizing exposure to risk is the main reason why companies keep their legacy systems. This is particularly the case on IBM i, but holds true even in the mainframe world, despite the mainframe’s exorbitant operating costs.
Code volume: The billions of lines of code developed over the last thirty or even fifty years are becoming almost impossible to replace. So, what is the real issue? That the mass of legacy code should be converted over time to reduce technical debt and at the same time some peripheral functionality be replaced with more modern applications? It is much easier to consider this kind of “step by step” approach than a radical big bang. Once again, risk management is the driving force behind this strategy.
The criticality of applications: Migrating from one system to another represents such a high risk that it becomes almost prohibitive. The banking sector is probably the best example of this. Most companies carry at least one legacy application in their information system, which may be old, but is still vital for the ongoing support of the company’s business. The sheer “mission criticality” of such systems means that a specific risk management approach becomes the top priority. Obsolescence of technologies and shortage of competent resources really are the key issues.
Competitive advantage: A large number of companies have developed their applications over many years according to their own business model and needs. These heritage applications constitute a true competitive advantage. The specific functionality they contain has been created, nurtured and maintained over decades and ERP software providers are unable to replicate the value contained within them, without massive bespoke configuration and development, almost impossible to re-create in most cases. The implementation of any ERP system can force an organisation towards accepting a significant and dangerous loss of their unique competitive advantage.
1. Ensure the sustainability of a Legacy system
Even if a technology or platform appears “exotic” at first glance in an increasingly standardized infrastructure, it is certainly not from a global standardization perspective that we should be approaching the issue. This is the easiest way to push an organisation down a “one-way street” from which it is difficult to reverse but also dangerous and expensive to continue. Assessing the long-term sustainability of any information system is a challenging exercise as it involves taking into account a significant number of conflicting parameters: the sustainability of the platform itself, the sustainability of the IT ecosystem, and the ability to blend into the overall strategic direction.
2. Durability of the IBM i platform
Classified in this category of “legacy system”, the IBM i platform has been under constant attack for more than 20 years now. IBM itself started the process when it released its RS6000 platforms on Unix. At the time, it was even the same sales people who were going to praise the added value of the Unix world as a future “de facto standard”. Then came Microsoft, which tried to create a consortium of suppliers each offering migration solutions.
The result: a customer erosion rate of 2-3% per year!
Another interesting example: the biggest ERP vendor on the IBM i platform is JD Edwards, now part of Oracle! At the beginning of the 2000s, JDE completely redeveloped this ERP with Oracle/Linux/Java technology, a solution called One World. Today, Oracle finds itself having to support two co-existing products “living their lives separately” and an IBM i customer base of 4,000 customers who refuse the switch to the new system.
As mentioned above, being able to leave the platform is the main reason why some organizations opt for a “complete rewrite”. So let us look at the counter-arguments on the other side of the coin – the overriding reasons to remain:
The lowest TCO on the market: what the IBM i platform offers which is absolutely unique in the IT industry is the tight integration between the O/S and the database manager.
The result? Extremely low operating costs. There is no need for expensive database administrators in the IBM i world since it is the O/S itself which manages the database automatically.
The other arguments are just as important: unrivalled reliability, proven security: no viruses detected for decades, scalability (an extremely important point given the exponential growth in the volume of data in the world) and a product roadmap running through to 2029. Some customers mention the risk that IBM will eventually abandon this system. You don’t kill a cash cow! The hardware is strictly identical to Unix and Linux platforms, at no extra cost for IBM. The OS is very stable (and very intuitive). The database is where IBM has concentrated all its resources and efforts. It has become a true relational database. All IBM’s R&D work on the database ensure we can exploit the real benefits of the IBM i platform.