Related
Suppose we have a function foo that does something
to all the elements between *firsta and *lastb:
foo(RandomAccessIterator1 firsta,RandomAccessIterator1 lasta){
for (RandomAccessIterator1 it=firsta;it!=lasta+1;it++){
//here stuff happens...
}
}
question a): is there a way to skip an index firsta<i<lastb by only
modifying the inputs to foo --e.g. the random iterators,
in other words without changing foo itself, just its input?
--Unfortunately the index I want to skip are not in the edges
(they are often deep between firsta and lasta) and foo
is a complicated divide&conquer algorithm that's not amenable
to being called on subsets of the original array the iterators
are pointing to.
question b): if doing a) is possible, what's the cost of doing that?
constant or does it depend on (lasta-firsta)?
The best way to do this would be to use an iterator that knows how to skip that element. A more generalized idea though, is an iterator that simply iterates over two separate ranges under the hood. I don't know of anything in boost that does this, so, here's one I just whipped up: http://coliru.stacked-crooked.com/a/588afa2a353942fc
Unfortunately, the code to detect which element to skip adds a teeny tiny amount of overhead to each and every iterator increment, so the overhead is technically proportional to lasta-firsta. Realistically, using this wrapper around a vector::iterator or a char* should bring it roughly to the same performance level as std::deque::iterator, so it's not like this should be a major slowdown.
The answer might be a bit picky, but you could call foo(firsta,i-1) and foo(i+1,lastb) or something similar to have the desired effect.
I'm trying to work out the best method to search a vector of type "Tracklet" (a class I have built myself) to find the first and last occurrence of a given value for one of its variables. For example, I have the following classes (simplified for this example):
class Tracklet {
TimePoint *start;
TimePoint *end;
int angle;
public:
Tracklet(CvPoint*, CvPoint*, int, int);
}
class TimePoint {
int x, y, t;
public:
TimePoint(int, int, int);
TimePoint(CvPoint*, int);
// Relevant getters and setters exist here
};
I have a vector "vector<Tracklet> tracklets" and I need to search for any tracklets with a given value of "t" for the end timepoint. The vector is ordered in terms of end time (i.e. tracklet.end->t).
I'm happy to code up a search algorithm, but am unsure of which route to take with it. I'm not sure binary search would be suitable, as I seem to remember it won't necessarily find the first. I was thinking of a method where I use binary search to find an index of an element with the correct time, then iterate back to find the first and forward to find the last. I'm sure there's a better way than that, since it wastes binary searches O(log n) by iterating.
Hopefully that makes sense: I struggled to explain it a bit!
Cheers!
If the vector is sorted and contains the value, std::lower_bound will give you an iterator to the first element with a given value and std::upper_bound will give you an iterator to one element past the last one containing the value. Compare the value with the returned element to see if it existed in the vector. Both these functions use binary search, so time is O(logN).
To compare on tracklet.end->t, use:
bool compareTracklets(const Tracklet &tr1, const Tracklet &tr2) {
return (tr1.end->t < tr2.end->t);
}
and pass compareTracklets as the fourth argument to lower_bound or upper_bound
I'd just use find and find_end, and then do something more complicated only if testing showed it to be too slow.
If you're really concerned about lookup performance, you might consider a different data structure, like a map with timestamp as the key and a vector or list of elements as the value.
A binary search seems like your best option here, as long as your vector remains sorted. It's essentially identical, performance-wise, to performing a lookup in a binary tree-structure.
dirkgently referred to a sweet optimization comparative. But I would in fact not use a std::vector for this.
Usually, when deciding to use a STL container, I don't really consider the performance aspect, but I do consider its interface regarding the type of operation I wish to use.
std::set<T>::find
std::set<T>::lower_bound
std::set<T>::upper_bound
std::set<T>::equal_range
Really, if you want an ordered sequence, outside of a key/value setup, std::set is just easier to use than any other.
You don't have to worry about inserting at a 'bad' position
You don't have problems of iterators invalidation when adding / removing an element
You have built-in methods for searching
Of course, you also want your Comparison Predicate to really shine (hopes the compiler inlines the operator() implementation), in every case.
But really, if you are not convinced, try a build with a std::vector and manual insertion / searching (using the <algorithm> header) and try another build using std::set.
Compare the size of the implementations (number of lines of code), compare the number of bugs, compare the speed, and then decide.
Most often, the 'optimization' you aim for is actually a pessimization, and in those rares times it's not, it's just so complicated that it's not worth it.
Optimization:
Don't
Expert only: Don't, we mean it
The vector is ordered in terms of time
The start time or the end time?
What is wrong with a naive O(n) search? Remember you are only searching and not sorting. You could use a sorted container as well (if that doesn't go against the basic design).
Which is better (or faster), a C++ for loop or the foreach operator provided by Qt? For example, the following condition
QList<QString> listofstrings;
Which is better?
foreach(QString str, listofstrings)
{
//code
}
or
int count = listofstrings.count();
QString str = QString();
for(int i=0;i<count;i++)
{
str = listofstrings.at(i);
//Code
}
It really doesn't matter in most cases.
The large number of questions on StackOverflow regarding whether this method or that method is faster, belie the fact that, in the vast majority of cases, code spends most of its time sitting around waiting for users to do something.
If you are really concerned, profile it for yourself and act on what you find.
But I think you'll most likely find that only in the most intense data-processing-heavy work does this question matter. The difference may well be only a couple of seconds and even then, only when processing huge numbers of elements.
Get your code working first. Then get it working fast (and only if you find an actual performance issue).
Time spent optimising before you've finished the functionality and can properly profile, is mostly wasted time.
First off, I'd just like to say I agree with Pax, and that the speed probably doesn't enter into it. foreach wins hands down based on readability, and that's enough in 98% of cases.
But of course the Qt guys have looked into it and actually done some profiling:
http://blog.qt.io/blog/2009/01/23/iterating-efficiently/
The main lesson to take away from that is: use const references in read only loops as it avoids the creation of temporary instances. It also make the purpose of the loop more explicit, regardless of the looping method you use.
It really doesn't matter. Odds are if your program is slow, this isn't the problem. However, it should be noted that you aren't make a completely equal comparison. Qt's foreach is more similar to this (this example will use QList<QString>):
for(QList<QString>::iterator it = Con.begin(); it != Con.end(); ++it) {
QString &str = *it;
// your code here
}
The macro is able to do this by using some compiler extensions (like GCC's __typeof__) to get the type of the container passed. Also imagine that boost's BOOST_FOREACH is very similar in concept.
The reason why your example isn't fair is that your non-Qt version is adding extra work.
You are indexing instead of really iterating. If you are using a type with non-contiguous allocation (I suspect this might be the case with QList<>), then indexing will be more expensive since the code has to calculate "where" the n-th item is.
That being said. It still doesn't matter. The timing difference between those two pieces of code will be negligible if existent at all. Don't waste your time worrying about it. Write whichever you find more clear and understandable.
EDIT: As a bonus, currently I strongly favor the C++11 version of container iteration, it is clean, concise and simple:
for(QString &s : Con) {
// you code here
}
Since Qt 5.7 the foreach macro is deprecated, Qt encourages you to use the C++11 for instead.
http://doc.qt.io/qt-5/qtglobal.html#foreach
(more details about the difference here : https://www.kdab.com/goodbye-q_foreach/)
I don't want to answer the question which is faster, but I do want to say which is better.
The biggest problem with Qt's foreach is the fact that it takes a copy of your container before iterating over it. You could say 'this doesn't matter because Qt classes are refcounted' but because a copy is used you don't actually change your original container at all.
In summary, Qt's foreach can only be used for read-only loops and thus should be avoided. Qt will happily let you write a foreach loop which you think will update/modify your container but in the end all changes are thrown away.
First, I completely agree with the answer that "it doesn't matter". Pick the cleanest solution, and optimize if it becomes a problem.
But another way to look at it is that often, the fastest solution is the one that describes your intent most accurately. In this case, QT's foreach says that you'd like to apply some action for each element in the container.
A plain for loop say that you'd like a counter i. You want to repeatedly add one to this value i, and as long as it is less than the number of elements in the container, you would like to perform some action.
In other words, the plain for loop overspecifies the problem. It adds a lot of requirements that aren't actually part of what you're trying to do. You don't care about the loop counter. But as soon as you write a for loop, it has to be there.
On the other hand, the QT people have made no additional promises that may affect performance. They simply guarantee to iterate through the container and apply an action to each.
In other words, often the cleanest and most elegant solution is also the fastest.
The foreach from Qt has a clearer syntax for the for loop IMHO, so it's better in that sense. Performance wise I doubt there's anything in it.
You could consider using the BOOST_FOREACH instead, as it is a well thought out fancy for loop, and it's portable (and presumably will make it's way into C++ some day and is future proof too).
A benchmark, and its results, on this can be found at http://richelbilderbeek.nl/CppExerciseAddOneAnswer.htm
IMHO (and many others here) it (that is speed) does not matter.
But feel free to draw your own conclusions.
For small collections, it should matter and foreach tends to be clearer.
However, for larger collections, for will begin to beat foreach at some point. (assuming that the 'at()' operator is efficient.
If this is really important (and I'm assuming it is since you are asking) then the best thing to do is measure it. A profiler should do the trick, or you could build a test version with some instrumentation.
You might look at the STL's for_each function. I don't know whether it will be faster than the two options you present, but it is more standardized than the Qt foreach and avoids some of the problems that you may run into with a regular for loop (namely out of bounds indexing and difficulties with translating the loop to a different data structure).
I would expect foreach to work nominally faster in some cases, and the about same in others, except in cases where the items are an actual array in which case the performace difference is negligible.
If it is implemented on top of an enumerator, it may be more efficient than a straight indexing, depending on implementation. It's unlikely to be less efficient. For example, if someone exposed a balanced tree as both indexable and enumerable, then foreach will be decently faster. This is because each index will have to independently find the referenced item, while an enumerator has the context of the current node to more efficiently navigate to the next ont.
If you have an actual array, then it depends on the implementation of the language and class whether foreach will be faster for the same as for.
If indexing is a literal memory offset(such as C++), then for should be marginally faster since you're avoiding a function call. If indexing is an indirection just like a call, then it should be the same.
All that being said... I find it hard to find a case for generalization here. This is the last sort of optimization you should be looking for, even if there is a performance problem in your application. If you have a performance problem that can be solved by changing how you iterate, you don't really have a performance problem. You have a BUG, because someone wrote either a really crappy iterator, or a really crappy indexer.
Why does nobody seem to use tuples in C++, either the Boost Tuple Library or the standard library for TR1? I have read a lot of C++ code, and very rarely do I see the use of tuples, but I often see lots of places where tuples would solve many problems (usually returning multiple values from functions).
Tuples allow you to do all kinds of cool things like this:
tie(a,b) = make_tuple(b,a); //swap a and b
That is certainly better than this:
temp=a;
a=b;
b=temp;
Of course you could always do this:
swap(a,b);
But what if you want to rotate three values? You can do this with tuples:
tie(a,b,c) = make_tuple(b,c,a);
Tuples also make it much easier to return multiple variable from a function, which is probably a much more common case than swapping values. Using references to return values is certainly not very elegant.
Are there any big drawbacks to tuples that I'm not thinking of? If not, why are they rarely used? Are they slower? Or is it just that people are not used to them? Is it a good idea to use tuples?
A cynical answer is that many people program in C++, but do not understand and/or use the higher level functionality. Sometimes it is because they are not allowed, but many simply do not try (or even understand).
As a non-boost example: how many folks use functionality found in <algorithm>?
In other words, many C++ programmers are simply C programmers using C++ compilers, and perhaps std::vector and std::list. That is one reason why the use of boost::tuple is not more common.
Because it's not yet standard. Anything non-standard has a much higher hurdle. Pieces of Boost have become popular because programmers were clamoring for them. (hash_map leaps to mind). But while tuple is handy, it's not such an overwhelming and clear win that people bother with it.
The C++ tuple syntax can be quite a bit more verbose than most people would like.
Consider:
typedef boost::tuple<MyClass1,MyClass2,MyClass3> MyTuple;
So if you want to make extensive use of tuples you either get tuple typedefs everywhere or you get annoyingly long type names everywhere. I like tuples. I use them when necessary. But it's usually limited to a couple of situations, like an N-element index or when using multimaps to tie the range iterator pairs. And it's usually in a very limited scope.
It's all very ugly and hacky looking when compared to something like Haskell or Python. When C++0x gets here and we get the 'auto' keyword tuples will begin to look a lot more attractive.
The usefulness of tuples is inversely proportional to the number of keystrokes required to declare, pack, and unpack them.
For me, it's habit, hands down: Tuples don't solve any new problems for me, just a few I can already handle just fine. Swapping values still feels easier the old fashioned way -- and, more importantly, I don't really think about how to swap "better." It's good enough as-is.
Personally, I don't think tuples are a great solution to returning multiple values -- sounds like a job for structs.
But what if you want to rotate three values?
swap(a,b);
swap(b,c); // I knew those permutation theory lectures would come in handy.
OK, so with 4 etc values, eventually the n-tuple becomes less code than n-1 swaps. And with default swap this does 6 assignments instead of the 4 you'd have if you implemented a three-cycle template yourself, although I'd hope the compiler would solve that for simple types.
You can come up with scenarios where swaps are unwieldy or inappropriate, for example:
tie(a,b,c) = make_tuple(b*c,a*c,a*b);
is a bit awkward to unpack.
Point is, though, there are known ways of dealing with the most common situations that tuples are good for, and hence no great urgency to take up tuples. If nothing else, I'm not confident that:
tie(a,b,c) = make_tuple(b,c,a);
doesn't do 6 copies, making it utterly unsuitable for some types (collections being the most obvious). Feel free to persuade me that tuples are a good idea for "large" types, by saying this ain't so :-)
For returning multiple values, tuples are perfect if the values are of incompatible types, but some folks don't like them if it's possible for the caller to get them in the wrong order. Some folks don't like multiple return values at all, and don't want to encourage their use by making them easier. Some folks just prefer named structures for in and out parameters, and probably couldn't be persuaded with a baseball bat to use tuples. No accounting for taste.
As many people pointed out, tuples are just not that useful as other features.
The swapping and rotating gimmicks are just gimmicks. They are utterly confusing to those who have not seen them before, and since it is pretty much everyone, these gimmicks are just poor software engineering practice.
Returning multiple values using tuples is much less self-documenting then the alternatives -- returning named types or using named references. Without this self-documenting, it is easy to confuse the order of the returned values, if they are mutually convertible, and not be any wiser.
Not everyone can use boost, and TR1 isn't widely available yet.
When using C++ on embedded systems, pulling in Boost libraries gets complex. They couple to each other, so library size grows. You return data structures or use parameter passing instead of tuples. When returning tuples in Python the data structure is in the order and type of the returned values its just not explicit.
You rarely see them because well-designed code usually doesn't need them- there are not to many cases in the wild where using an anonymous struct is superior to using a named one.
Since all a tuple really represents is an anonymous struct, most coders in most situations just go with the real thing.
Say we have a function "f" where a tuple return might make sense. As a general rule, such functions are usually complicated enough that they can fail.
If "f" CAN fail, you need a status return- after all, you don't want callers to have to inspect every parameter to detect failure. "f" probably fits into the pattern:
struct ReturnInts ( int y,z; }
bool f(int x, ReturnInts& vals);
int x = 0;
ReturnInts vals;
if(!f(x, vals)) {
..report error..
..error handling/return...
}
That isn't pretty, but look at how ugly the alternative is. Note that I still need a status value, but the code is no more readable and not shorter. It is probably slower too, since I incur the cost of 1 copy with the tuple.
std::tuple<int, int, bool> f(int x);
int x = 0;
std::tuple<int, int, bool> result = f(x); // or "auto result = f(x)"
if(!result.get<2>()) {
... report error, error handling ...
}
Another, significant downside is hidden in here- with "ReturnInts" I can add alter "f"'s return by modifying "ReturnInts" WITHOUT ALTERING "f"'s INTERFACE. The tuple solution does not offer that critical feature, which makes it the inferior answer for any library code.
Certainly tuples can be useful, but as mentioned there's a bit of overhead and a hurdle or two you have to jump through before you can even really use them.
If your program consistently finds places where you need to return multiple values or swap several values, it might be worth it to go the tuple route, but otherwise sometimes it's just easier to do things the classic way.
Generally speaking, not everyone already has Boost installed, and I certainly wouldn't go through the hassle of downloading it and configuring my include directories to work with it just for its tuple facilities. I think you'll find that people already using Boost are more likely to find tuple uses in their programs than non-Boost users, and migrants from other languages (Python comes to mind) are more likely to simply be upset about the lack of tuples in C++ than to explore methods of adding tuple support.
As a data-store std::tuple has the worst characteristics of both a struct and an array; all access is nth position based but one cannot iterate through a tuple using a for loop.
So if the elements in the tuple are conceptually an array, I will use an array and if the elements are not conceptually an array, a struct (which has named elements) is more maintainable. ( a.lastname is more explanatory than std::get<1>(a)).
This leaves the transformation mentioned by the OP as the only viable usecase for tuples.
I have a feeling that many use Boost.Any and Boost.Variant (with some engineering) instead of Boost.Tuple.
I'm using a std::map (VC++ implementation) and it's a little slow for lookups via the map's find method.
The key type is std::string.
Can I increase the performance of this std::map lookup via a custom key compare override for the map? For example, maybe std::string < compare doesn't take into consideration a simple string::size() compare before comparing its data?
Any other ideas to speed up the compare?
In my situation the map will always contain < 15 elements, but it is being queried non stop and performance is critical. Maybe there is a better data structure that I can use that would be faster?
Update: The map contains file paths.
Update2: The map's elements are changing often.
First, turn off all the profiling and DEBUG switches. These can slow down STL immensely.
If that's not it, part of the problem may be that your strings are identical for the first 80-90% of the string. This isn't bad for map, necessarily, but it is for string comparisons. If this is the case, your search can take much longer.
For example, in this code find() will likely result in a couple of string compares, but each will return after comparing the first character until "david", and then the first three characters will be checked. So at most, 5 characters will be checked per call.
map<string,int> names;
names["larry"] = 1;
names["david"] = 2;
names["juanita"] = 3;
map<string,int>::iterator iter = names.find("daniel");
On the other hand, in the following code, find() will likely check 135+ characters:
map<string,int> names;
names["/usr/local/lib/fancy-pants/share/etc/doc/foobar/longpath/yadda/yadda/wilma"] = 1;
names["/usr/local/lib/fancy-pants/share/etc/doc/foobar/longpath/yadda/yadda/fred"] = 2;
names["/usr/local/lib/fancy-pants/share/etc/doc/foobar/longpath/yadda/yadda/barney"] = 3;
map<string,int>::iterator iter = names.find("/usr/local/lib/fancy-pants/share/etc/doc/foobar/longpath/yadda/yadda/betty");
That's because the string comparisons have to search deeper to find a match since the beginning of each string is the same.
Using size() in your comparison for equality won't help you much here since your data set is so small. A std::map is kept sorted so its elements can be searched with a binary search. Each call to find should result in less than 5 string comparisons for a miss, and an average of 2 comparisons for a hit. But it does depend on your data. If most of your path strings are of different lengths, then a size check like Motti describes could help a lot.
Something to consider when thinking of alternative algorithms is how many many "hits" you get. Are most of your find() calls returning end() or a hit? If most of your find()s return end() (misses) then you are searching the entire map every time (2logn string compares).
Hash_map is a good idea; it should cut your search time in about half for hits; more for misses.
A custom algorithm may be called for because of the nature of path strings, especially if your data set has common ancestry like in the above code.
Another thing to consider is how you get your search strings. If you are reusing them, it may help to encode them into something that is easier to compare. If you use them once and discard them, then this encoding step is probably too expensive.
I used something like a Huffman coding tree once (a long time ago) to optimize string searches. A binary string search tree like that may be more efficient in some cases, but its pretty expensive for small sets like yours.
Finally, look into alternative std::map implementations. I've heard bad things about some of VC's stl code performance. The DEBUG library in particular is bad about checking you on every call. StlPort used to be a good alternative, but I haven't tried it in a few years. I've always loved Boost too.
As Even said the operator used in a set is < not ==.
If you don't care about the order of the strings in your set you can pass the set a custom comparator that performs better than the regular less-than.
For example if a lot of your strings have similar prefixes (but they vary in length) you can sort by string length (since string.length is constant speed).
If you do so beware a common mistake:
struct comp {
bool operator()(const std::string& lhs, const std::string& rhs)
{
if (lhs.length() < rhs.length())
return true;
return lhs < rhs;
}
};
This operator does not maintain a strict weak ordering, as it can treat two strings as each less than the other.
string a = "z";
string b = "aa";
Follow the logic and you'll see that comp(a, b) == true and comp(b, a) == true.
The correct implementation is:
struct comp {
bool operator()(const std::string& lhs, const std::string& rhs)
{
if (lhs.length() != rhs.length())
return lhs.length() < rhs.length();
return lhs < rhs;
}
};
The first thing is to try using a hash_map if that's possible - you are right that the standard string compare doesn't first check for size (since it compares lexicographically), but writing your own map code is something you'd be better off avoiding. From your question it sounds like you do not need to iterate over ranges; in that case map doesn't have anything hash_map doesn't.
It also depends on what sort of keys you have in your map. Are they typically very long? Also what does "a little slow" mean? If you have not profiled the code it's quite possible that it's a different part taking time.
Update: Hmm, the bottleneck in your program is a map::find, but the map always has less than 15 elements. This makes me suspect that the profile was somehow misleading, because a find on a map this small should not be slow, at all. In fact, a map::find should be so fast, just the overhead of profiling could be more than the find call itself. I have to ask again, are you sure this is really the bottleneck in your program? You say the strings are paths, but you're not doing any sort of OS calls, file system access, disk access in this loop? Any of those should be orders of magnitude slower than a map::find on a small map. Really any way of getting a string should be slower than the map::find.
You can try to use a sorted vector (here's one sample), this may turn out to be faster (you'll have to profile it to make sure of-course).
Reasons to think it'll be faster:
Less memory allocations and deallocations (the vector will expand to the maximal size used and then reuse freed memory).
Binary find with random access should be faster than tree traversal (espacially due to data locality).
Reasons to think it'll be slower:
Deleations and additions will mean moving strings around in memory, since string's swap is efficiant and the size of the data set is small this may not be an issue.
std::map's comparator isn't std::equal_to it's std::less, I'm not sure what the best way to short circuit a < compare so that it would be faster than the built in one.
If there are always < 15 elems, perhaps you could use a key besides std::string?
Motti has a good solution. However, I'm pretty sure that for your < 15 elements a map isn't the right way because its overhead will always be greater than that of a simple lookup table with an appropriate hashing scheme. In your case, it might even be enough to hash by length alone, and if that still produces collisions, use a linear search through all entries of the same length.
To establish if I'm right, a benchmark is of course required but I'm quite sure of its outcome.
You might consider pre-computing a hash for a string, and saving that in your map. Doing so gives the advantage of hash compares instead of string compares during the search through the std::map tree.
class HashedString
{
unsigned m_hash;
std::string m_string;
public:
HashedString(const std::string& str)
: m_hash(HashString(str))
, m_string(str)
{};
// ... copy constructor and etc...
unsigned GetHash() const {return m_hash;}
const std::string& GetString() const {return m_string;}
};
This has the benefits of computing a hash of the string once, on construction. After this, you could implement a comparison function:
struct comp
{
bool operator()(const HashedString& lhs, const HashedString& rhs)
{
if(lhs.GetHash() < rhs.GetHash()) return true;
if(lhs.GetHash() > rhs.GetHash()) return false;
return lhs.GetString() < rhs.GetString();
}
};
Since hashes are now computed on HashedString construction, they are stored that way in the std::map, and so the compare can happen very quickly (an integer compare) in an astronomically high percentage of the time, falling back on standard string compares when the hashes are equal.
Maybe you could reverse the strings prior to using them as keys in the map? That could help if the first few letters of each string are identical.
Here are some things you can consider:
0) Are you sure this is where the performance bottleneck is? Like the results from Quantify, Cachegrind, gprof or something like that? Because lookups on such a smap map should be fairly fast...
1) You can override the functor used to compare the keys in std::map<>, there is a second template parameter to do that. I doubt you can do much better than operator<, however.
2) Are the contents of the map changing a lot? If not, and given the very small size of your map, maybe using a sorted vector and binary search could yield better results (for example because you can exploit memory locality better.
3) Are the elements known at compile time? You could use a perfect hash function to improve lookup times if that is the case. Search for gperf on the web.
4) Do you have a lot of lookups that fail to find anything? If so, maybe comparing with the first and last elements in the collection may eliminate many mismatches quicker than a full search every time.
These have been suggested already, but in more detail:
5) Since you have so few strings, maybe you could use a different key. For example, are your keys all the same size? Can you use a class containing a fixed-length array of characters? Can you convert your strings to numbers or some data structure with only numbers?
Depending on the usage cases, there are some other techniques you can use. For example we had an application that needed to keep up with over a million different file paths. The problem with that there were thousands of objects that needed to keep small maps of these file paths.
Since adding new file paths to the data set was an infrequent operation, when path was added to the system, a master map was searched. If the path was not found, then it was added and a new sequenced integer (starting at 1) was returned. If the path already existed, then the previously assigned integer was returned. Then each map maintained by each object was converted from a string based map to an integer map. Not only did this greatly improve performance, it reduced memory usage by not having so many duplicate copies of the strings.
Sure, this is a very specific optimization. But when it comes to performance improvements, you often find yourself having to make tailored solutions to specific problems.
And I hate strings :) Not are they slow to compare, but they can really trash your CPU caches on high performance software.
Try std::tr1::unordered_map (found in the header <tr1/unordered_map>). This is a hash map, and, while it doesn't maintain a sorted order of elements, will likely be far faster than a regular map.
If your compiler doesn't support TR1, get a newer version. MSVC and gcc both support TR1, and I believe the newest versions of most other compilers also have support. Unfortunately, a lot of the library reference sites haven't been updated, so TR1 remains a largely-unknown piece of technology.
I hope C++0x isn't the same way.
EDIT: Note that the default hashing method for tr1::unordered_map is tr1::hash, which needs to be specialized to work on a UDT, probably.
Where you have long common substrings, a trie might be a better data structure than a map or a hash_map. I said "might", though - a hash_map already only traverses the key once per lookup, so should be fairly fast. I won't discuss it further since others already have.
You could also consider a splay tree if some keys are more frequently looked up than others, but of course this makes the worst-case lookup worse than a balanced tree, and lookups are mutating operations, which may matter to you if you're using e.g. a reader-writer lock.
If you care about the performance of lookups more than modifications, you might do better with an AVL tree than a red-black, which I think is what STL implementations generally use for map. An AVL tree is typically better balanced and so will on average require fewer comparisons per lookup, but the difference is marginal.
Finding an implementation of these that you're happy with might be an issue. A search on the Boost main page suggests they have a splay and AVL tree but not a trie.
You mentioned in a comment that you never have a lookup that fails to find anything. So you could in theory skip the final comparison, which in a tree of 15 < 2^4 elements could give you something like a 20-25% speedup without doing anything else. In fact, maybe more than that, since equal strings are the slowest to compare. Whether it's worth writing your own container just for this optimisation is another question.
You might also consider locality of reference - I don't know whether you could avoid the occasional page miss by allocating the keys and the nodes out of a small heap. If you only need about 15 entries at a time, then assuming a file name limit below 256 bytes you could ensure that everything accessed during a lookup fits into a single 4k page (apart from the key being looked up, of course). It may be that comparing the strings is insignificant compared with a couple of page loads. However, if this is your bottleneck there must be an enormous number of lookups going on, so I'd guess that everything is reasonably close to the CPU. Worth checking, maybe.
Another thought: if you are using pessimistic locking on a structure where there's a lot of contention (you said in a comment the program is massively multi-threaded) then regardless of what the profiler tells you (what code the CPU cycles are spent in), it might be costing you more than you think by effectively limiting you to 1 core. Try a reader-writer lock?
hash_map is not standard, try using unordered_map available in tr1 (which is available in boost if your tool chain doesn't already have it).
For small numbers of strings you might be better using vector, as map is typically implemented as a tree.
Why don't you use a hashtable instead? boost::unordered_map could do. Or you can roll out your own solution, and store the crc of a string instead of the string itself. Or better yet, put #defines for the strings, and use those for lookup, e.g.,
#define "STRING_1" STRING_1