c# - Why does multithreading give a worse performance here? -

i enumerate through bitarray setting every second bit false.

now i'd speed splitting 2 threads.. weird reason though, time per thread half amount of work takes 64% more time, , wonder why's that?

could due kind of cpu caching effect? how do properly?

i have tried 8 threads lambda expressions around ~1400 ms, in single threading constandly got 850 ms. when let single thread work, took me 830 ms. don't understand, knowing cause here?

code:

    class program     {         static int count = 0x10000000;         static int half = count / 2;         static bitarray bitarray = new bitarray(count);          static unsafe void main(string[] args)         {             stopwatch sw = stopwatch.startnew();  #if single             (int = 0; < bitarray.count; += 2)                 bitarray.set(i, true); #else             thread thread1 = new thread(thread1);             thread thread2 = new thread(thread2);             thread1.start();             thread2.start();             thread1.join();             thread2.join(); #endif             sw.stop();              console.writeline(sw.elapsedmilliseconds);             console.readline();         }          static void thread1()         {             stopwatch sw = stopwatch.startnew();             (int = 0; < half; += 2)                 bitarray.set(i, true);             sw.stop();             console.writeline("thread1: {0}", sw.elapsedmilliseconds);         }          static void thread2()         {             stopwatch sw = stopwatch.startnew();             (int = half; < count; += 2)                 bitarray.set(i, true);             sw.stop();             console.writeline("thread2: {0}", sw.elapsedmilliseconds);         }     } 

bitarray not thread-safe class. should not use that. , in fact, beside correctness cause of slowness. here's why:

if @ source code of bitarray, contains int version field, updated @ every operation, notably set(), call.

this means every thread continuously updates same memory location, huge perf killer because cores have communicate , synchronize when accessing location. in condition makes perfect sense multithreaded solution has worse performance single core one.