Expensive to replace, hard to service, and impossible to implement large-scale changes without having to replace most of the infrastructure. This could describe a well-loved classic car or a legacy monolith. In either case, technology is evolving at such a rapid pace that it’s very challenging to try and maintain the same product for more than a few years. With both cars and software, there are always newer, better options that give you more speed, agility and value for money.
Why are legacy monoliths still out there?
One of the main obstacles when it comes to replacing legacy monoliths is that it can be a very expensive process. As many of these legacy monoliths still offer decent value for business, it can be hard to make the case in terms of return on investment.
The issue is that, generally speaking, legacy monoliths cannot offer the quality and performance that you would expect from a modern system. Granted, many are still functional, but that doesn’t make them suitable for a fast-moving, modern business, and that obviously becomes a problem when you want to expand and explore new ideas and ways of working to make productivity gains to get a competitive edge over your rivals.
Most of the time, software monoliths tend to be a collection of apps used in a business environment. They are old, they don’t really offer a lot of value based on modern standards, but they still get the job done. In a situation like this, you really want all the help that you can get, so it’s tempting to stick with them until major problems prompt a rethink.
Common characteristics of legacy monoliths
Most legacy monoliths feature either old technology or programming languages – or both. These need to be replaced if you want a fast, dependable and functional system that really brings you the quality and value you need. When you use older hardware or software in the business world, you may find colleagues’ familiarity with its limitations has resulted in ingenious workarounds that have mitigated the difficulties. But the fact that legacy monoliths will be slow and lack many of the modern features that you need can still act as a drag on doing business.
That’s why you need to be very careful when it comes to the tools you use and what features you want added. Every small detail matters when it comes to legacy monoliths, and that’s why you want everything upgraded to fit modern standards in a consistent way. This has clear advantages for your business as it will enable it to scale up and expand, which after all is what matters the most.
What causes the legacy monoliths to not be as appealing or helpful when compared to modern standards? First, the lack of vendor support is a huge problem. When you have a business, you want to make sure that everything works in a predictable way. Where legacy monoliths do not receive any support, and unless you know how to repair the hardware or software yourself, you will not be able to use it at the highest capacity. Obviously, that means lower productivity and many other problems related to the limitations you are imposing on your business.
Then there’s also the fact that you will suffer from outdated architecture and design principles. In an ideal world, you’d want to make sure that everything is updated and that it works the best way that it can. With modern architecture you don’t have to worry about any problems, since it’s most likely to be adaptable, scalable and also fully customizable. However, legacy monoliths don’t deliver in this department, and in the end it can generate serious problems. This is useful information to include in a business case about why you don’t want to stick with a legacy monolith: replacing them can save more money in the future than inertia, and it will offer a more professional way to eliminate problems and focus on the best results.
Another thing to notice here aside from the lack of support is that the overall legacy monolith environment can become very costly in time. Sure, you might eventually find someone that provides support, but that can be very expensive. The reason for that is the overall support experience is a lot more complex. Using modern tools to offer support for legacy monolith is not that simple, so you have to acquire legacy software and hardware for professional support. Achieving all that is not exactly straightforward, which is why your typical legacy monolith will tend to have little to no support.
When your entire company relies on legacy monoliths and they barely have any support or assistance, problems are inevitable. This is why you need to commit to the change and gain as much senior buy-in as possible. While wrangling with finance might not be easy, it’s certainly worth the effort in the end.
When is it time to upgrade legacy monolith systems?
As soon as vendor support ends, you most likely have to go and upgrade the legacy monolith. It makes a lot of sense to do that, because you want hardware and software that has great support in order to respond and provide extra value to your business and your customers. You should upgrade as fast as possible, as the longer you leave it, the more likely problems are likely to pile up.
Of course, you can also upgrade legacy monolith systems when you want to correct errors. Most of the time, upgrades will help you get past any issues, and the results can be pretty impressive. To make things even better, you won’t have to worry about double-checking for any issues, as those checkups tend to not require a lot of assistance or attention to begin with.
You might also need to upgrade legacy monoliths when you want to fulfill new performance requirements or develop functionality. When that happens, you really need to rely on the best and most reliable tools, and with legacy monoliths you can’t do that. Upgrading to more modern systems can really make a huge difference, and people throughout the business will appreciate the difference. Again, a caveat: success or failure in securing an upgrade comes down to a commitment to the modernization process, because if this is lacking at the top of the business, it will be very difficult to get the right sort of investment.
As mentioned earlier, lack of support should be a great incentive for upgrades, too. One way to make the case is to indicate how expensive support for legacy monoliths can get, to the point where it will be cheaper to acquire modern systems anyway. With the right comparison, the leadership team will be impressed with how affordable it really is to have support for modern systems when compared with legacy monoliths. It’s all about pointing out the advantages as well as the risks of doing nothing.
When is there still time to formulate a modernization/transition strategy?
If you think it’s time to upgrading, it’s worth setting business as usual to one side and taking a good hard look at your system. You’ll probably notice a few things from the following list:
1. There are errors in the system that you cannot correct and their number is growing;
2. Your outdated system is the main blocker to developing an important piece of functionality or fulfilling performance requirements;
3. It’s hard to find people to support the system. The market for people who know technology is shrinking rapidly – and new entrants only want to work with modern technology;
4. The high cost of support means it would be cheaper to upgrade – although ROI will not be instantaneous;
5. No one at your business knows how the system actually works – the developers left ages ago, the documentation is terrible and the system itself is a complete enigma.
If points 1-3 sound like your system, don’t delay! Modernization is urgent. A system suffering from points 4-5 will limp on for some time, but the time is right to develop a modernization/transition strategy.
Forming a strategy
Your strategy will need to take a balanced view. The benefits of microservices include efficiency, team independence, better control over your code base and easy testing. But there are drawbacks, too. You’ll need to consider system resilience in the event of service failure, and logging, coordinating features for release and rolling back affected services during release.
Unlike a monolith, microservices can be located on many machines, written in different languages and with different methods and control capabilities. Again, this seeming advantage – flexibility – can also complicate matters, so don’t underestimate the time you need to take over this. Anyone who has wrestled with the dashboard of a new car will appreciate it can take a bit of getting used to.
With microservices, you will find processes that dragged in the past become a lot simpler, and you can finally focus on the results you want. It’s the best way to replace a long-serving legacy monolith and finally focus on the future and the features that you and your business needs. Microservices, because they offer a better level of support, often prove to offer a return on investment far in advance of ad hoc fixes and tweaks.
New microservice architecture, implemented sensibly, will help boost business agility and maintainability, while also bringing in a lot of quality, value and support. Microservices offer more control, upgradability and transparency. These are all things that you need to consider when thinking about how to move on from your monolith. But it’s worth bearing in mind that your business, just like your drive to work, will have its own unique challenges. With a new car or software, your strategy is a good place to spell these out.
Can we help you with legacy systems? Then don’t hesitate to contact us.
