Sign up ×
Programmers Stack Exchange is a question and answer site for professional programmers interested in conceptual questions about software development. It's 100% free.

"Premature optimization is root of all evil" is something almost all of us have heard/read. What I am curious what kind of optimization not premature, i.e. at every stage of software development (high level design, detailed design, high level implementation, detailed implementation etc) what is extent of optimization we can consider without it crossing over to dark side.

share|improve this question
See also: Is micro-optimisation important when coding? – gnat May 7 '13 at 8:50

12 Answers 12

up vote 77 down vote accepted

When you're basing it off of experience? Not evil. "Every time we've done X, we've suffered a brutal performance hit. Let's plan on either optimizing or avoiding X entirely this time."

When it's relatively painless? Not evil. "Implementing this as either Foo or Bar will take just as much work, but in theory, Bar should be a lot more efficient. Let's Bar it."

When you're avoiding crappy algorithms that will scale terribly? Not evil. "Our tech lead says our proposed path selection algorithm runs in factorial time; I'm not sure what that means, but she suggests we commit seppuku for even considering it. Let's consider something else."

The evil comes from spending a whole lot of time and energy solving problems that you don't know actually exist. When the problems definitely exist, or when the phantom psudo-problems may be solved cheaply, the evil goes away.

Steve314 and Matthieu M. raise points in the comments that ought be considered. Basically, some varieties of "painless" optimizations simply aren't worth it either because the trivial performance upgrade they offer isn't worth the code obfuscation, they're duplicating enhancements the compiler is already performing, or both. See the comments for some nice examples of too-clever-by-half non-improvements.

share|improve this answer
Occasionally, solving a phantom problem easily is still mildly evil, as it can result in harder to read, harder to maintain code. Not much harder (or it wouldn't be an easy solution), but perhaps occasionally still relevant. An example might be using a clever bitwise trick that some people won't recognise, and which the compiler will probably apply anyway if it's useful. – Steve314 Jan 1 '11 at 12:03
I agree with Steve here, sometimes the "optimization" is simply not worth it, especially because compilers are so damn good. Example ? if i is unsigned, i / 2 can be replaced by i >> 1. It is faster. But it is also more cryptic (not everyone will see the effect, even those who do may lose time). But the worst of it is that the compiler will do it anyway, so why obfuscate the source code ;) ? – Matthieu M. Jan 1 '11 at 13:53
@Larry : I didn't, so I guess it's a good example. – Joris Meys Jan 1 '11 at 15:01
In my view, optimisations, even simple ones, should also be regarded evil if they impact readabiliy/maintainabiliy of the code and are not based on actual performance measurements. – Bart van Ingen Schenau Jan 1 '11 at 17:17
@Matthew: Teach them what? Dirty and unnecessary tricks? Why? If profiling shows that a i/2 is indeed a hot spot and that (unbelievable, but let's assume) i>>1 makes it faster, so do it, and put a comment to it that this profiling showed that this is faster. If this is indeed needed anywhere (which I doubt, since, as Matthieu said, compilers should be smart enough to do this themselves), novices will learn something, if it isn't (which is likely), why do you want to plug their heads with unneeded folklore? – sbi Jan 2 '11 at 20:23

Application code should only be as good as necessary, but library code should be as good as possible, since you never know how your library is going to be used. So when you're writing library code, it needs too be good in all aspects, be it performance, robustness, or any other category.

Also, you need to think about performance when you design your application and when you pick algorithms. If it isn't designed to be performant, no degree of hackery can make it performant afterwards and no micro-optimizations will outweigh a superior algorithm.

share|improve this answer
Library code should document whether it's trying to be "as good as possible", or what its objective is. Code need not be absolutely optimal in order to be useful, provided that consumers only use it when appropriate. – supercat May 13 '14 at 1:56
Sorry, but "be good in all aspects" sounds suspiciously like overengineering. Plus, it's probably not realistic - life is always about tradeoffs. – sleske Jun 17 at 10:23

what kind of optimization [is] not premature

The kind that come as a result of known issues.

share|improve this answer

You should always choose a "good enough" solution in all cases based on your experiences.

The optimization saying refers to writing "more complex code than 'good enough' to make it faster" before actually knowing that it is necessary, hence making the code more complex than necessary. Complexity is what makes things hard, so that isn't a good thing.

This means that you should not choose a super complex "can sort 100 Gb files by transparently swapping to disk" sorting routine when a simple sort will do, but you should also make a good choice for the simple sort in the first place. Blindly choosing Bubble Sort or "pick all entries randomly and see if they are in order. Repeat." is rarely good.

share|improve this answer

The full quote defines when optimization is not premature:

A good programmer will not be lulled into complacency by such reasoning, he will be wise to look carefully at the critical code; but only after that code has been identified. [emphasis mine]

You can identify critical code in many ways: critical data structures or algorithms (e.g. used heavily or the "core" of the project) can give major optimizations, many minor optimizations are identified through profilers, and so on.

share|improve this answer
Yeah... It's all well and good to shave 90% off the time a random function call takes, but maybe you'd get a bigger impact by looking at the code where your app actually spends 80% of its time and shaving off a few percent there. – fennec Jan 14 '11 at 3:51

It's not optimisation when picking things that are hard to change eg: hardware platform.

Picking data structures is a good example - critical to meeting both functional and non-functional (performance) requirements. Not easily changed and yet it will drive everything else in your app. Your data structures change what algorithms are available etc.

share|improve this answer

My general rule of thumb: if you're not sure you'll need the optimization, assume you don't. But keep it in mind for when you do need to optimize. There are some issues that you can know about up front though. This usually involves choosing good algorithms and data structures. For instance, if you need to check membership in a collection, you can be pretty sure you will need some type of set data structure.

share|improve this answer

In my experience, at the detailed implementation phase the answer lies in profiling the code. Its important to know what needs to be faster and what is acceptably fast.

It is also important to know where exactly the performance bottleneck is - optimizing a part of the code which takes only 5% of the total time to run wont do any good.

Steps 2 and 3 describe non-premature optimization:

  1. Make it work
  2. Test. Not fast enough? Profile it.
  3. Using the data from step 2, optimize the slowest sections of the code.
share|improve this answer
You forgot step 0, which is: architect the application properly so that you can expect reasonable performance from the start. – Robert Harvey Jan 6 '11 at 23:53
I was only speaking about the detailed implementation phase. – Gorgi Kosev Jan 7 '11 at 1:21
I question step #3--very often the best answer is to figure out a different approach so you're not doing the slow bit of code in the first place. – Loren Pechtel Jan 14 '11 at 5:25
Pick the right data structures. – jasonk Mar 15 '12 at 13:25

When you have tests to make sure you don't break anything.

share|improve this answer

When the requirements or the market specifically asks for it.

For example performance is a requirement in most financial applications because low latency is crucial. Depending on the nature of the traded instrument, optimization can go from using non-locking algorithms in a high-level language to using a low-level language and the even the extreme - implementing the order matching algorithms in hardware itself (using FPGA for example).

Other example would be some types of embedded devices. Take for example the ABS brake; firstly there is the safety, when you hit the break the car should slow down. But there is also performance, you would not want any delays when you hit the break.

share|improve this answer

I only know of one way to answer this question, and that is to get experience in performance tuning. That means - write programs, and after they are written, find speedups in them, and do it iteratively. Here's one example.

Here's the mistake most people make: They try to optimize the program before actually running it. If they have taken a course in programming (from a professor who doesn't actually have much practical experience) they will have big-O colored glasses, and they will think that's what it's all about. It's all the same problem, prior optimization.**

Somebody said: First make it right, Then make it fast. They were right.

But now for the kicker: If you have done this a few times, you recognize the silly things you earlier did that cause speed problems, so you instinctively avoid them. (Things like making your class structure too heavy, getting swamped with notifications, confusing size of function calls with their time cost, the list goes on and on ...) You instinctively avoid these, but guess what it looks like to the less-experienced: premature optimization!

So these silly debates go on and on :)

** Another thing they say is you don't have to worry about it any more, because compilers are so good, and machines are so fast nowadays. (KIWI - Kill It With Iron.) There are no exponential hardware or system speedups (done by very smart hard-working engineers) that can possibly compensate for exponential software slowdowns (done by programmers who think this way).

share|improve this answer

Most people would call optimization premature, if you're optimizing something that isn't resulting in a "soft failure" (it works but it's still useless) of the system due to performance.

Real world examples.

  • If my bubble sort takes 20ms to run, optimizing it to 1ms quicksort is not going to enhance the overall utility in any meaningful way despite being a 2000% performance increase.

  • If a web page takes 20s to load and we decrease it to 1s, this can increase the utility of the website from 0 to near infinity. Basically something that was broken because it was too slow, is now useful.

share|improve this answer

protected by MichaelT Jun 15 at 20:07

Thank you for your interest in this question. Because it has attracted low-quality answers, posting an answer now requires 10 reputation on this site.

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged or ask your own question.