Translate Fortran to C++ with AI and RAG
lanl.gov
lanl.gov
I've actually had to do this with a couple of different Fortran projects when I was in college, I translated them to C for various reasons.
Maybe it's because it was specifically code written by scientists (i.e somewhat brute force and very straightforward) but there really wasn't very many features that I can recall that didn't have a direct C counterpart, other than column major ordering and arrays staring at 1.
Was I just blissfully unaware?
In theory AI could do a more idiomatic translation, but I think I would still prefer the janky but correct translation over the looks nice but probably subtly buggy AI one.
I tried to find reference to how they did it, does anyone know?
It sounds like this approach of translating old code could help speed up teams that are looking at rewrites. I also have some old code that's in Kotlin that I'd like to move to something else. I had a bad NullPointerException take down my application after missing a breaking change in a Kotlin update.
do concurrent (i = 1:n)
y(i) = y(i) + a*x(i)
enddo
std::transform(par, x, x+n, y, y,
[=](float x, float y){ return y + a*x; }
);
D has a module system, where you import a module and it pulls the global declarations out of it. To speed up this process, there's a compiler switch to output just the global declarations to another file, a .di file, that functions much like a .h file in C.
Then there came along ImportC, where a C lexer/parser was welded onto the D compiler logic.
aaaand it wasn't long before the switch was thrown to generate a .di file, and voila, the C code got translated to D!
This resulted in making it dramatically easier to use existing C headers and source files with D.
Using automatic tools - whether AI-based or transpilers - leaves that opportunity unused, and both approaches are likely to create some additional technical debt (errors in translation, odd, non-idiomatic ways of doing things introduced by the automatism etc.).
Fortran also hasn’t been faster than C++ for a very long time. This was demonstrable even back when I worked in HPC, and Fortran can be quite a bit worse for some useful types of performance engineering. The only reason we continued to use it was that a lot of legacy numerics code was written in it. Almost all new code was written in C++ because it was easier to maintain. I actually worked in Fortran before I worked in HPC, it was already dying in HPC by the time I got there. Nothing has changed in the interim. If anything, C++ is a much stronger language today than it was back then.
Also to me seems more likely that people that enjoy Fortran in HPC are more likely to change to Chapel than use C++.
What makes you say so? See musicale's comment above. I have a hard time seeing C++ as easier to maintain, if we are just talking about the language. The ecosystem is a different story.
This was before modern C++ existed, so most of the code was “C with classes” style C++. If you can read C then you can read that code. I don’t consider that to be particularly maintainable by modern standards but neither is Fortran.
Modern C++ dialects, on the other hand, are much more maintainable than either. Better results with a fraction of the code. The article doesn’t say but I would expect at least idiomatic C++11, if not later dialects.
ISO Fortran does not acknowledge the existences of FPM, just like any programming language ISO standard, the ecosystem is not part of the standard.
Thus, availability of a compiler is but a small piece of the puzzle. The real problem is the spider web of dependencies on the mainframe environment, as the enterprises business processes have been intertwined into the mainframe system over decades.
I’ve seen firsthand how much frustration COBOL can cause. Many of my colleagues didn’t enjoy writing it, they stuck with it because it paid well, not because they loved the work. The language itself isn’t the hard part; it’s the decades of accumulated technical debt and the sheer complexity of the environment. Over time, these systems become so intertwined with business processes that untangling them feels impossible. But what if we approached it incrementally?
Imagine taking an existing COBOL codebase, say, for a large insurance system and identifying the core business logic buried within it. These are the rules and conditions that power critical operations, like calculating premiums or processing claims. Now, instead of trying to rewrite everything at once, you build a parallel backend in a modern language like Java or Go. You don’t aim for a literal translation of the COBOL code, you focus on replicating the functionality in a way that makes sense in a modern context. For example, replace hardcoded file operations with database calls, or screen based interactions with REST APIs.
Most mainframe customers already use middleware like MuleSoft or IBM Z/OS Connect to route requests to both systems simultaneously. For every write operation, you update both the mainframe’s DB2 database and a modern relational database like Postgres. For every read operation, you compare the results from both systems. If there’s a discrepancy, you flag it for investigation. Over time, as you handle more and more business scenarios, you’d start covering all the edge cases. This dual system approach lets you validate the new backend without risking critical operations.
Of course, this process isn’t without its struggles. Testing is a huge challenge because mainframe systems often rely on implicit behaviors that aren’t documented anywhere. My colleagues used to joke that the only way to understand some parts of the system was to run it and see what happened. That’s why rigorous testing and monitoring are essential you need to catch discrepancies early before they cause problems. There’s also the cultural side of things. People get attached to their mainframes, especially when they’ve been running reliably for decades. Convincing stakeholders to invest in a multi year migration effort requires strong leadership and a clear case for ROI.
But I think the effort is worth it. Moving off the mainframe isn’t just about saving money though that’s a big part of it. It’s about future proofing your organization. Mainframes are great at what they do, but they’re also a bottleneck when it comes to innovation. Want to integrate with a third party service? Good luck. Need to hire new developers? Most of them have never touched COBOL. By transitioning to a modern platform, you open up opportunities to innovate faster, integrate with other systems more easily, and attract talent who can actually work on your codebase.
In the end, this isn’t a quick fix it’s a long term strategy. But I believe it’s achievable if you take it step by step. Start small, validate constantly, and gradually build up to a full replacement. What do others think? Are there better ways to tackle this problem, or am I missing something obvious?
COBOL varies greatly, the dialect depends on the mainframe. Chatbots will get quite confused about this. AI training data doesn't have much true COBOL, the internet is polluted with GnuCOBOL which is a mismash of a bunch of different dialects, minus all the things that make a mainframe a mainframe. So it will assume the COBOL code is more modern than it is. In terms of generating COBOL (e.g. for adding some debugging code to an existing system to analyze its behavior) it won't be able to stay within the 80 column limit due to tokenization, it will just be riddled with syntax errors.
Data matters, and mainframes have a rather specific way they store and retrieve data. Just operating the mainframe to get the data out of an old system and into a new database in a workable & well-architected format will be its own chore.
Finally, the reason these systems haven't been ported is because requirements for how the system needs to work are tight. The COBOL holdouts are exclusively financial, government, and healthcare -- no one else is stuck on old mainframes for any other reason. The new system to replace it needs to exactly match the behavior of the old system, the developer has to know how to figure out the exact confines of the laws and regulations or they are not qualified to do the task of porting it. All an LLM will do is hallucinate a new set of requirements and ignore the old ones. And aside from just knowing the requirements on paper, you'd need to spend a good chunk of time just checking what the existing system is even doing, because there will be plenty of surprises in such an old system.
There are also modern compilers to IBM mainframes, including Go, C++, Java, PHP,..
Also outside DevOps and CNCF application space very few people bother with Go, specially not the kind of customers that buy IBM mainframes.
You can run cobol on x86, there are at least two compilers.
yeah but how much did it cost?
At a large financial news company I worked for, the AWS opex was £4m, All to put text on a page. You know, a solved problem. We spent years fucking about with fleet to make something "resilient". (then early, shitty k8s)
The opex for FAANG is astronomical. facebook spends something like 40 billion a year on infra. Thats without staffing costs. (manifold, the S3 clone, is a bastard to use.)
My point is, which I should make more obvious: if you buy the right mainframe, you can literally blow one of them up, and not loose any uptime. Yes, its expensive.
But, hiring a team to fuck about with k8s on AWS is going to cost more, and never be finished, because true redundant systems are hard to design and build, even by distributed systems experts.
f2c ? But yeah, 1 level of abstraction sucks. We need around 10 to be satisfied.