… | |
… | |
899 | I<destroy> is the time, in microseconds, that it takes to destroy a single |
899 | I<destroy> is the time, in microseconds, that it takes to destroy a single |
900 | watcher. |
900 | watcher. |
901 | |
901 | |
902 | =head2 Results |
902 | =head2 Results |
903 | |
903 | |
904 | name watcher bytes create invoke destroy comment |
904 | name watchers bytes create invoke destroy comment |
905 | EV/EV 400000 244 0.56 0.46 0.31 EV native interface |
905 | EV/EV 400000 244 0.56 0.46 0.31 EV native interface |
906 | EV/Any 100000 610 3.52 0.91 0.75 EV + AnyEvent watchers |
906 | EV/Any 100000 610 3.52 0.91 0.75 EV + AnyEvent watchers |
907 | CoroEV/Any 100000 610 3.49 0.92 0.75 coroutines + Coro::Signal |
907 | CoroEV/Any 100000 610 3.49 0.92 0.75 coroutines + Coro::Signal |
908 | Perl/Any 16000 654 4.64 1.22 0.77 pure perl implementation |
908 | Perl/Any 100000 513 4.91 0.92 1.15 pure perl implementation |
909 | Event/Event 16000 523 28.05 21.38 0.86 Event native interface |
909 | Event/Event 16000 523 28.05 21.38 0.86 Event native interface |
910 | Event/Any 16000 943 34.43 20.48 1.39 Event + AnyEvent watchers |
910 | Event/Any 16000 943 34.43 20.48 1.39 Event + AnyEvent watchers |
911 | Glib/Any 16000 1357 96.99 12.55 55.51 quadratic behaviour |
911 | Glib/Any 16000 1357 96.99 12.55 55.51 quadratic behaviour |
912 | Tk/Any 2000 1855 27.01 66.61 14.03 SEGV with >> 2000 watchers |
912 | Tk/Any 2000 1855 27.01 66.61 14.03 SEGV with >> 2000 watchers |
913 | POE/Event 2000 6644 108.15 768.19 14.33 via POE::Loop::Event |
913 | POE/Event 2000 6644 108.15 768.19 14.33 via POE::Loop::Event |
914 | POE/Select 2000 6343 94.69 807.65 562.69 via POE::Loop::Select |
914 | POE/Select 2000 6343 94.69 807.65 562.69 via POE::Loop::Select |
915 | |
915 | |
916 | =head2 Discussion |
916 | =head2 Discussion |
917 | |
917 | |
918 | The benchmark does I<not> measure scalability of the event loop very |
918 | The benchmark does I<not> measure scalability of the event loop very |
919 | well. For example, a select-based event loop (such as the pure perl one) |
919 | well. For example, a select-based event loop (such as the pure perl one) |
… | |
… | |
921 | file descriptors grows high. In this benchmark, only a single filehandle |
921 | file descriptors grows high. In this benchmark, only a single filehandle |
922 | is used (although some of the AnyEvent adaptors dup() its file descriptor |
922 | is used (although some of the AnyEvent adaptors dup() its file descriptor |
923 | to worka round bugs). |
923 | to worka round bugs). |
924 | |
924 | |
925 | C<EV> is the sole leader regarding speed and memory use, which are both |
925 | C<EV> is the sole leader regarding speed and memory use, which are both |
926 | maximal/minimal, respectively. Even when going through AnyEvent, there is |
926 | maximal/minimal, respectively. Even when going through AnyEvent, there are |
927 | only one event loop that uses less memory (the C<Event> module natively), and |
927 | only two event loops that use slightly less memory (the C<Event> module |
928 | no faster event model, not event C<Event> natively. |
928 | natively and the pure perl backend), and no faster event models, not even |
|
|
929 | C<Event> natively. |
929 | |
930 | |
930 | The pure perl implementation is hit in a few sweet spots (both the |
931 | The pure perl implementation is hit in a few sweet spots (both the |
931 | zero timeout and the use of a single fd hit optimisations in the perl |
932 | zero timeout and the use of a single fd hit optimisations in the perl |
932 | interpreter and the backend itself). Nevertheless tis shows that it |
933 | interpreter and the backend itself). Nevertheless tis shows that it |
933 | adds very little overhead in itself. Like any select-based backend its |
934 | adds very little overhead in itself. Like any select-based backend its |
934 | performance becomes really bad with lots of file descriptors, of course, |
935 | performance becomes really bad with lots of file descriptors, of course, |
935 | but this was not subjetc of this benchmark. |
936 | but this was not subject of this benchmark. |
936 | |
937 | |
937 | The C<Event> module has a relatively high setup and callback invocation cost, |
938 | The C<Event> module has a relatively high setup and callback invocation cost, |
938 | but overall scores on the third place. |
939 | but overall scores on the third place. |
939 | |
940 | |
940 | C<Glib>'s memory usage is quite a bit bit higher, features a faster |
941 | C<Glib>'s memory usage is quite a bit bit higher, but it features a |
941 | callback invocation and overall lands in the same class as C<Event>. |
942 | faster callback invocation and overall ends up in the same class as |
|
|
943 | C<Event>. However, Glib scales extremely badly, doubling the number of |
|
|
944 | watchers increases the processing time by more than a factor of four, |
|
|
945 | making it completely unusable when using larger numbers of watchers |
|
|
946 | (note that only a single file descriptor was used in the benchmark, so |
|
|
947 | inefficiencies of C<poll> do not account for this). |
942 | |
948 | |
943 | The C<Tk> adaptor works relatively well, the fact that it crashes with |
949 | The C<Tk> adaptor works relatively well. The fact that it crashes with |
944 | more than 2000 watchers is a big setback, however, as correctness takes |
950 | more than 2000 watchers is a big setback, however, as correctness takes |
945 | precedence over speed. Nevertheless, its performance is surprising, as the |
951 | precedence over speed. Nevertheless, its performance is surprising, as the |
946 | file descriptor is dup()ed for each watcher. This shows that the dup() |
952 | file descriptor is dup()ed for each watcher. This shows that the dup() |
947 | employed by some adaptors is not a big performance issue (it does incur a |
953 | employed by some adaptors is not a big performance issue (it does incur a |
948 | hidden memory cost inside the kernel, though). |
954 | hidden memory cost inside the kernel, though, that is not reflected in the |
|
|
955 | figures above). |
949 | |
956 | |
950 | C<POE>, regardless of underlying event loop (wether using its pure perl |
957 | C<POE>, regardless of underlying event loop (wether using its pure perl |
951 | select-based backend or the Event module) shows abysmal performance and |
958 | select-based backend or the Event module) shows abysmal performance and |
952 | memory usage: Watchers use almost 30 times as much memory as EV watchers, |
959 | memory usage: Watchers use almost 30 times as much memory as EV watchers, |
953 | and 10 times as much memory as both Event or EV via AnyEvent. Watcher |
960 | and 10 times as much memory as both Event or EV via AnyEvent. Watcher |
… | |
… | |
962 | Using EV through AnyEvent is faster than any other event loop, but most |
969 | Using EV through AnyEvent is faster than any other event loop, but most |
963 | event loops have acceptable performance with or without AnyEvent. |
970 | event loops have acceptable performance with or without AnyEvent. |
964 | |
971 | |
965 | The overhead AnyEvent adds is usually much smaller than the overhead of |
972 | The overhead AnyEvent adds is usually much smaller than the overhead of |
966 | the actual event loop, only with extremely fast event loops such as the EV |
973 | the actual event loop, only with extremely fast event loops such as the EV |
967 | adds Anyevent significant overhead. |
974 | adds AnyEvent significant overhead. |
968 | |
975 | |
969 | And you should simply avoid POE like the plague if you want performance or |
976 | And you should simply avoid POE like the plague if you want performance or |
970 | reasonable memory usage. |
977 | reasonable memory usage. |
971 | |
978 | |
972 | |
979 | |