There are four things you should know about legacy systems: they no longer receive service support, they are based on outdated technology, they are no longer available for purchase and they pose security risks. Imagine sticking with Windows XP while most computers run on Windows 10 with Windows 11 coming soon to replace it. Actually, the US navy did just that, paying millions of dollars to keep using Windows XP until 2018 — four years after the official support ended.
Failure to modernize will most likely lead to additional hidden costs in special licensing, staff training, in-house tech support and system management. As a result, sticking with a legacy system will cost you more in the long run than switching to a newer one today.
Read on to learn more about the hidden costs of maintaining legacy systems. The knowledge will help ensure you don’t currently overspend as well as realize when an upgrade is needed.
Some legacy systems still used today were built 20 or 30 years ago and are no longer supported by vendors. They run on old programming languages, use outdated network infrastructures and hardware and are barely compatible with new technology, such as cloud and IoT. As a result, managing and supporting such systems becomes financially cumbersome.
Why are legacy system still used then? Well, the problem is not just technical. Many old systems were built as custom solutions and require additional resources for training management and support staff. In addition, it might be hard to motivate IT professionals to work with a legacy system. This kind of work precludes employees from developing new skill sets to stay competitive in the market. As a result, keeping such professionals in the company as well as hiring outside consultants may become too costly and equally problematic.
At the onset of the Covid-19 crisis, an illustrative case of how legacy systems may preclude upgrading and maintaining databases presented itself. Between March 15th and April 4th of 2020, the US federal labor department reported 16.8 million unemployment claims filed.
Okay, so what exactly happened? Nothing special, really — a pretty banal issue that you can expect from outdated software: the users were unable to submit their unemployment claims via an online filling form because the system ran on a 50-year-old COBOL programming language (what could have gone wrong?) was not able to deal with updating so many entries in the database.
This half-forgotten programming language was initially written for mainframes and giant computer systems which offer significantly lower data processing capacity than modern cloud infrastructures. As a result, the US digital unemployment system was not able to process the increased amount of data, because it was simply outdated.
This was pretty painful for the federal government, but imagine what could happen to a small company if something like that occurs — the financial and reputational losses could be intractable.
Moreover, many COBOL programmers reached near-retirement age and did not properly document what they did with the code over the years. All this creates additional difficulty for new employees in maintaining and upgrading the databases because they lack the COBOL knowledge and will have to go over large amounts of already written undocumented code. And this bumps back into additional difficulties and financial losses associated with keeping and attracting developers and support specialists to work with your company.
And here’s a curious, relevant fact about finances: 43% of banking systems and 95% of ATM swipes in the U.S. still rely on COBOL — our money is safe, yay!
Logically enough, the previous case points to another problem — legacy systems are prone to increased security risks. Outdated software architecture, use of old mainframes and “code smells” that got accumulated from numerous patches and attempts to integrate with newer technology may result in system vulnerabilities and security breaches. Eventually, private user data and sensitive information may leak, creating negative consequences for your business.
A notable example is the BlueKeep system vulnerability that was discovered in some versions of Microsoft Windows a few years ago. Essentially, the attacks attempted to install crypto-mining software on Remote Desktop Protocol (RDP) servers that had exposed port 3389 to the Internet. As a result, many systems were crushed, left with the dreaded “blue screen,” and Microsoft had to issue patches to support both current and older versions of the operating system.
The attack failed, because the crashing computer problem allowed Microsoft to spot it in a timely manner, as the exploit code appeared to be incompatible with the previous patch issued to fight another attack called Meltdown. Bottom line: the older a system is, the more patches it requires and the more security vulnerabilities it accumulates, which means increasing the risks of the next BlueKeep-like attempt becoming fatal.
Legacy systems were not designed to work with such modern technology as cloud computing, artificial intelligence (AI) and IoT. Thus, integrating aging systems with modern environments requires a multilayered architecture of middleware and connectors that are built on top of the core technology.
Eventually, if not maintained properly, the system becomes slower and too complex to operate smoothly, which costs you more in every subsequent update and integration endeavor. This is not to say that maintaining such juggernauts automatically means bleeding cash unceasingly.
For example, the US government allocated more than $100 billion in 2021 compared to about $90 billion in 2019 on IT and cyber-related investments. More than 80% of this amount went on system operation and maintenance, including legacy software. According to the US Government Accountability Office (GAO), more funds are being spent, because more IT systems are getting obsolete.
Moreover, maintaining a legacy infrastructure bears additional environmental costs as you consume more power to keep long rows of old-generation servers on your business premises. Migrating to the cloud or switching to new generation hardware could be a way to reduce this extra spending.
There is a point in the lifespan of your company when new software integrations no longer suffice, because your aging system simply stops performing well enough to keep up with the pace of your business. It not only creates losses in efficiency but becomes a barrier to your further business growth. While your competitors migrate to the cloud, getting the most of microservices, containers and reusable code that saves them both time and money, you keep running circles with patching and patching and patching.
Aging systems also limit your opportunities to enjoy new technologies, such as machine learning, AI, IoT, augmented reality (AR) and other emerging trends that can help you with data analytics, user engagement, payment processing, customer service automation and more. They just can’t scale sufficiently and aren’t agile enough to avoid negatively impacting your business. System modernization can help you avoid these negative effects.
Deloitte suggests that modern-day technological transformations are driven by two interconnected processes: ongoing platform modernization and cloud platform migration. Cloud migration is seen by Deloitte experts as both a means of and a logical outcome of data modernization. This suggests that modernization should be seen as a strategic and multistep process, but not a one-time endeavor.
Not surprisingly, 98% of organizations with a minimum turnover of $1 billion surveyed for the 2020 Mainframe Modernization Business Barometer were planning a migration to the cloud. Eventually, these organizations can save up to $31 million if they modernize the most crucial aspects of their legacy systems, according to Business Wire.
However, in certain business environments, mainframes still provide better security than cloud platforms, and fully migrating to the cloud might not be the best fit. When this is the case, hybrid computing that involves gradual modernization combining mainframe reliability and security with cloud agility and flexibility might be a solution.
Having a legacy system does not mean it is no longer viable. What is more important is not how old a system is but whether its performance efficiency, maintenance costs, security, agility and reliability are comparable to the newer solutions on the market. Once your system becomes inferior in any of these parameters, it is all but certain it’s time to upgrade.
At IntexSoft, we can help you deal with hidden costs associated with legacy systems. We perform full-scale legacy system modernization services, including complete or partial cloud migration.
Thinking of modernizing your system? Drop us a line!