Optimization of a C++ code (that uses UnorderedMap and Vector) -

i trying optimize part of c++ code taking long time (the following part of code takes 19 seconds x amount of data, , trying finish whole process in less 5 seconds same amount of data - based on benchmarks have). have function "add" have written , copied code here. try explain as possible think needed understand code. please let me know if have missed something.

the following function add called x times x amount of data entries.

void hashtable::add(pointobject vector)   // pointobject user-defined object {     int combinedhash = hash(vector);   // function "hash" takes less 1 second x amount of data     // hashtablemap unordered_map<int, std::vector<pointobject>>     if (hashtablemap.count(combinedhash) == 0)    {         // if hashmap not contain combinedhash key,          //  add key , new vector         std::vector<pointobject> pointvectorlist;         pointvectorlist.push_back(vector);         hashtablemap.insert(std::make_pair(combinedhash, pointvectorlist));    }    else    {         // otherwise find key , corresponding vector of pointobjects , add current pointobject existing vector         auto = hashtablemap.find(combinedhash);         if (it != hashtablemap.end())         {             std::vector<pointobject> pointvectorlist = it->second;             pointvectorlist.push_back(vector);             it->second = pointvectorlist;         }    } } 

you doing lot of useless operations... if understand correctly, simplified form simply:

void hashtable::add(const pointobject& vector) {    hashtablemap[hash(vector)].push_back(vector);     } 

this works because

• a map when accessed using operator[] create default-initialized value if it's not present in map
• the value (an std::vector) returned reference can directly push_back incoming point it. std::vector either newly inserted 1 or existing 1 if key in map.

note that, depending on size of pointobject , other factors, possibly more efficient pass vector value instead of const pointobject&. kind of micro optimization requires profiling performed sensibly.