Good performance is not just big O - Julio Merino (jmmv.dev)
jmmv.dev
Having a fast and responsive app is orthogonal to “knowing your big Os”. Unfortunately, most tech companies over-emphasize algorithms in interviews and downplay systems knowledge, and I believe that’s one reason behind sluggish apps and bloated systems. I’ve seen this play out repeatedly. Interviewers ask a LeetCode-style coding question, which is then followed by the ritual of discussing time and memory complexity. Candidates ace the answers. But then… their “real” code suffers from subtle yet impactful performance problems. Focusing on big O complexity rarely matters in most apps. Sure, it’s important to think about your algorithmic choices, but there are so many more details to worry about that have a direct impact on app performance and responsiveness. Let’s look at a bunch of them!

Having a fast and responsive app is orthogonal to “knowing your big Os”. Unfortunately, most tech companies over-emphasize algorithms in interviews and downplay systems knowledge, and I believe that’s one reason behind sluggish apps and bloated systems.

I’ve seen this play out repeatedly. Interviewers ask a LeetCode-style coding question, which is then followed by the ritual of discussing time and memory complexity. Candidates ace the answers. But then… their “real” code suffers from subtle yet impactful performance problems.

  • chunkystylesEnglish
    14·
    10 months ago

    Big O is a useful tool, but it doesn’t directly translate to performance. Understanding how systems work is a lot more useful and important if you really care about optimization and performance.

    This is something I learned pretty early on as a professional developer. I got a computer science degree and was taught data structures, algorithms, and big O. In my first job, I came across a small piece of Java code that was being run a lot that had a small list being searched each time. I figured converting to a HashSet would be faster, but in testing it was actually slower. I forget the exact details, but I learned to test my assumptions about performance and not to just blindly change things.

    My degree, for the most part, did not prepare me for working with large, interconnected systems. The only course that came close was an elective called “concurrent programming” which was really just “algorithms, but parallel”.