Comment by drob518

Comment by drob518 12 hours ago

17 replies

SAT, SMT, and constraint solvers are criminally underutilized in the software industry. We need more education about what they are, how they work, and what sorts of problems they can solve.

LegionMammal978 11 hours ago

At least personally, I've been very underwhelmed by their performance when I've tried using them. Usually past a few dozen variables or so is when I start hitting unacceptable exponential runtimes, especially for problem instances that are unsatisfiable or barely-satisfiable. Maybe their optimizations are well-suited for knapsack problems and other classic OR stuff, but if your problem doesn't fit the mold, then it's very hit-or-miss.

  • cerved 6 hours ago

    I've worked on a model with thousands of variables and hundreds of thousands of parameters with a hundred constraints. There are pitfalls you need to avoid, like reification, but it's definitely doable.

    Of course, NP hard problems become complex at an exponential rate but that doesn't change if you use another exact solving technique.

    Using local-search are very useful for scaling but at the cost of proven optimality

  • j2kun 10 hours ago

    I think this hits the nail on the head: performance is the obstacle, and you can't get good performance without some modeling expertise, which most people don't have.

  • js8 10 hours ago

    I wish I knew better how to use them for these coding problems, because I agree with GP they're underutilized.

    But I think if you have constraint problem, that has an efficient algorithm, but chokes a general constraint solver, that should be treated as a bug in the solver. It means that the solver uses bad heuristics, somewhere.

    • LegionMammal978 10 hours ago

      I'm pretty sure that due to Rice's theorem, etc., any finite set of heuristics will always miss some constraint problems that have an efficient solution. There's very rarely a silver bullet when it comes to generic algorithms.

      • sigbottle 9 hours ago

        I think they're saying that the types of counter-examples are so pathological in most cases that if you're doing any kind of auto-generation of constraints - for example, a DSL backed by a solver - should have good enough heuristics.

        Like it might even be the case that certain types of pretty powerful DSLs just never generate "bad structures". I don't know, I've not done research on circuits, but this kind of analysis shows up all the time in other adjacent fields.

  • drob518 11 hours ago

    Well, they aren’t magic. You have to use them correctly and apply them to problems that match how they work. Proving something is unsat is worst case NP. These solvers don’t change that.

    • LegionMammal978 10 hours ago

      Of course they aren't magic, but people keep talking about them as if they're perfectly robust and ready-to-use for any problem within their domain. In reality, unless you have lots of experience in how to "use them correctly" (which is not something I think can be taught by rote), you'd be better off restricting their use to precisely the OR/verification problems they're already popular for.

      • drob518 10 hours ago

        Hence my statement about education. All tools must be used correctly in their proper domain, that is true. Don’t try to drive screws with a hammer. But I'm curious what problems you tried them on and found them wanting and what your alternative was? I actually find that custom solutions work better for simple problems and that solvers do a lot better when the problem complexity grows. You’re better off solving the Zebra puzzle and its ilk with brute force code, not a solver, for instance.

loeg 11 hours ago

In what way? They're useful for toy problems like this but they're very slow on larger problems.

  • drob518 10 hours ago

    SAT solvers are used daily to generate solutions for problems that have literally millions of variables. So, what you said is just wrong on the face. Yes, some talented people can write custom code that solves specific problems faster than a general purpose solver, particularly for easy special cases of the general problem, but most of the time that results in the programmer recreating the guts of a solver customized to a specific problem. There’s sort of a corollary of Greenspun’s Tenth Rule that every sufficiently complicated program also contains an ad hoc, informally-specified, bug-ridden, slow-implementation of half of a SAT or SMT solver.

    • sigbottle 10 hours ago

      I mean right tool for the right job. Plenty of formulations and problems (our job has plenty of arbitrarily hard graph algorithms) that have 90% of the problem just being a very clever reduction with nice structure.

      Then the final 10% is either NP hard, or we want to add some DSL flexibility which introduces halting problem issues. Once you lower it enough, then comes the SMT solvers.

  • cerved 6 hours ago

    Define large. We've written model which solves real business issues in 8K lines of MiniZinc and it wasn't slow.

    The conventional wisdom is the larger you make an NP hard problem, the slower is going to get. Irregardless of algorithm.