… | |
… | |
65 | technically possible. |
65 | technically possible. |
66 | |
66 | |
67 | Of course, if you want lots of policy (this can arguably be somewhat |
67 | Of course, if you want lots of policy (this can arguably be somewhat |
68 | useful) and you want to force your users to use the one and only event |
68 | useful) and you want to force your users to use the one and only event |
69 | model, you should I<not> use this module. |
69 | model, you should I<not> use this module. |
70 | |
|
|
71 | |
70 | |
72 | =head1 DESCRIPTION |
71 | =head1 DESCRIPTION |
73 | |
72 | |
74 | L<AnyEvent> provides an identical interface to multiple event loops. This |
73 | L<AnyEvent> provides an identical interface to multiple event loops. This |
75 | allows module authors to utilise an event loop without forcing module |
74 | allows module authors to utilise an event loop without forcing module |
… | |
… | |
141 | =head2 I/O WATCHERS |
140 | =head2 I/O WATCHERS |
142 | |
141 | |
143 | You can create an I/O watcher by calling the C<< AnyEvent->io >> method |
142 | You can create an I/O watcher by calling the C<< AnyEvent->io >> method |
144 | with the following mandatory key-value pairs as arguments: |
143 | with the following mandatory key-value pairs as arguments: |
145 | |
144 | |
146 | C<fh> the Perl I<file handle> (I<not> file descriptor) to watch for |
145 | C<fh> the Perl I<file handle> (I<not> file descriptor) to watch |
147 | events. C<poll> must be a string that is either C<r> or C<w>, which |
146 | for events. C<poll> must be a string that is either C<r> or C<w>, |
148 | creates a watcher waiting for "r"eadable or "w"ritable events, |
147 | which creates a watcher waiting for "r"eadable or "w"ritable events, |
149 | respectively. C<cb> is the callback to invoke each time the file handle |
148 | respectively. C<cb> is the callback to invoke each time the file handle |
150 | becomes ready. |
149 | becomes ready. |
|
|
150 | |
|
|
151 | Although the callback might get passed parameters, their value and |
|
|
152 | presence is undefined and you cannot rely on them. Portable AnyEvent |
|
|
153 | callbacks cannot use arguments passed to I/O watcher callbacks. |
151 | |
154 | |
152 | The I/O watcher might use the underlying file descriptor or a copy of it. |
155 | The I/O watcher might use the underlying file descriptor or a copy of it. |
153 | You must not close a file handle as long as any watcher is active on the |
156 | You must not close a file handle as long as any watcher is active on the |
154 | underlying file descriptor. |
157 | underlying file descriptor. |
155 | |
158 | |
156 | Some event loops issue spurious readyness notifications, so you should |
159 | Some event loops issue spurious readyness notifications, so you should |
157 | always use non-blocking calls when reading/writing from/to your file |
160 | always use non-blocking calls when reading/writing from/to your file |
158 | handles. |
161 | handles. |
159 | |
|
|
160 | Although the callback might get passed parameters, their value and |
|
|
161 | presence is undefined and you cannot rely on them. Portable AnyEvent |
|
|
162 | callbacks cannot use arguments passed to I/O watcher callbacks. |
|
|
163 | |
162 | |
164 | Example: |
163 | Example: |
165 | |
164 | |
166 | # wait for readability of STDIN, then read a line and disable the watcher |
165 | # wait for readability of STDIN, then read a line and disable the watcher |
167 | my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub { |
166 | my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub { |
… | |
… | |
174 | |
173 | |
175 | You can create a time watcher by calling the C<< AnyEvent->timer >> |
174 | You can create a time watcher by calling the C<< AnyEvent->timer >> |
176 | method with the following mandatory arguments: |
175 | method with the following mandatory arguments: |
177 | |
176 | |
178 | C<after> specifies after how many seconds (fractional values are |
177 | C<after> specifies after how many seconds (fractional values are |
179 | supported) should the timer activate. C<cb> the callback to invoke in that |
178 | supported) the callback should be invoked. C<cb> is the callback to invoke |
180 | case. |
179 | in that case. |
|
|
180 | |
|
|
181 | Although the callback might get passed parameters, their value and |
|
|
182 | presence is undefined and you cannot rely on them. Portable AnyEvent |
|
|
183 | callbacks cannot use arguments passed to time watcher callbacks. |
181 | |
184 | |
182 | The timer callback will be invoked at most once: if you want a repeating |
185 | The timer callback will be invoked at most once: if you want a repeating |
183 | timer you have to create a new watcher (this is a limitation by both Tk |
186 | timer you have to create a new watcher (this is a limitation by both Tk |
184 | and Glib). |
187 | and Glib). |
185 | |
|
|
186 | Although the callback might get passed parameters, their value and |
|
|
187 | presence is undefined and you cannot rely on them. Portable AnyEvent |
|
|
188 | callbacks cannot use arguments passed to time watcher callbacks. |
|
|
189 | |
188 | |
190 | Example: |
189 | Example: |
191 | |
190 | |
192 | # fire an event after 7.7 seconds |
191 | # fire an event after 7.7 seconds |
193 | my $w = AnyEvent->timer (after => 7.7, cb => sub { |
192 | my $w = AnyEvent->timer (after => 7.7, cb => sub { |
… | |
… | |
234 | |
233 | |
235 | You can watch for signals using a signal watcher, C<signal> is the signal |
234 | You can watch for signals using a signal watcher, C<signal> is the signal |
236 | I<name> without any C<SIG> prefix, C<cb> is the Perl callback to |
235 | I<name> without any C<SIG> prefix, C<cb> is the Perl callback to |
237 | be invoked whenever a signal occurs. |
236 | be invoked whenever a signal occurs. |
238 | |
237 | |
|
|
238 | Although the callback might get passed parameters, their value and |
|
|
239 | presence is undefined and you cannot rely on them. Portable AnyEvent |
|
|
240 | callbacks cannot use arguments passed to signal watcher callbacks. |
|
|
241 | |
239 | Multiple signal occurances can be clumped together into one callback |
242 | Multiple signal occurances can be clumped together into one callback |
240 | invocation, and callback invocation will be synchronous. synchronous means |
243 | invocation, and callback invocation will be synchronous. synchronous means |
241 | that it might take a while until the signal gets handled by the process, |
244 | that it might take a while until the signal gets handled by the process, |
242 | but it is guarenteed not to interrupt any other callbacks. |
245 | but it is guarenteed not to interrupt any other callbacks. |
243 | |
246 | |
… | |
… | |
257 | |
260 | |
258 | The child process is specified by the C<pid> argument (if set to C<0>, it |
261 | The child process is specified by the C<pid> argument (if set to C<0>, it |
259 | watches for any child process exit). The watcher will trigger as often |
262 | watches for any child process exit). The watcher will trigger as often |
260 | as status change for the child are received. This works by installing a |
263 | as status change for the child are received. This works by installing a |
261 | signal handler for C<SIGCHLD>. The callback will be called with the pid |
264 | signal handler for C<SIGCHLD>. The callback will be called with the pid |
262 | and exit status (as returned by waitpid). |
265 | and exit status (as returned by waitpid), so unlike other watcher types, |
|
|
266 | you I<can> rely on child watcher callback arguments. |
263 | |
267 | |
264 | There is a slight catch to child watchers, however: you usually start them |
268 | There is a slight catch to child watchers, however: you usually start them |
265 | I<after> the child process was created, and this means the process could |
269 | I<after> the child process was created, and this means the process could |
266 | have exited already (and no SIGCHLD will be sent anymore). |
270 | have exited already (and no SIGCHLD will be sent anymore). |
267 | |
271 | |
… | |
… | |
452 | might chose the wrong one unless you load the correct one yourself. |
456 | might chose the wrong one unless you load the correct one yourself. |
453 | |
457 | |
454 | You can chose to use a rather inefficient pure-perl implementation by |
458 | You can chose to use a rather inefficient pure-perl implementation by |
455 | loading the C<AnyEvent::Impl::Perl> module, which gives you similar |
459 | loading the C<AnyEvent::Impl::Perl> module, which gives you similar |
456 | behaviour everywhere, but letting AnyEvent chose is generally better. |
460 | behaviour everywhere, but letting AnyEvent chose is generally better. |
|
|
461 | |
|
|
462 | =head1 OTHER MODULES |
|
|
463 | |
|
|
464 | The following is a non-exhaustive list of additional modules that use |
|
|
465 | AnyEvent and can therefore be mixed easily with other AnyEvent modules |
|
|
466 | in the same program. Some of the modules come with AnyEvent, some are |
|
|
467 | available via CPAN. |
|
|
468 | |
|
|
469 | =over 4 |
|
|
470 | |
|
|
471 | =item L<AnyEvent::Util> |
|
|
472 | |
|
|
473 | Contains various utility functions that replace often-used but blocking |
|
|
474 | functions such as C<inet_aton> by event-/callback-based versions. |
|
|
475 | |
|
|
476 | =item L<AnyEvent::Handle> |
|
|
477 | |
|
|
478 | Provide read and write buffers and manages watchers for reads and writes. |
|
|
479 | |
|
|
480 | =item L<AnyEvent::Socket> |
|
|
481 | |
|
|
482 | Provides a means to do non-blocking connects, accepts etc. |
|
|
483 | |
|
|
484 | =item L<AnyEvent::HTTPD> |
|
|
485 | |
|
|
486 | Provides a simple web application server framework. |
|
|
487 | |
|
|
488 | =item L<AnyEvent::DNS> |
|
|
489 | |
|
|
490 | Provides asynchronous DNS resolver capabilities, beyond what |
|
|
491 | L<AnyEvent::Util> offers. |
|
|
492 | |
|
|
493 | =item L<AnyEvent::FastPing> |
|
|
494 | |
|
|
495 | The fastest ping in the west. |
|
|
496 | |
|
|
497 | =item L<Net::IRC3> |
|
|
498 | |
|
|
499 | AnyEvent based IRC client module family. |
|
|
500 | |
|
|
501 | =item L<Net::XMPP2> |
|
|
502 | |
|
|
503 | AnyEvent based XMPP (Jabber protocol) module family. |
|
|
504 | |
|
|
505 | =item L<Net::FCP> |
|
|
506 | |
|
|
507 | AnyEvent-based implementation of the Freenet Client Protocol, birthplace |
|
|
508 | of AnyEvent. |
|
|
509 | |
|
|
510 | =item L<Event::ExecFlow> |
|
|
511 | |
|
|
512 | High level API for event-based execution flow control. |
|
|
513 | |
|
|
514 | =item L<Coro> |
|
|
515 | |
|
|
516 | Has special support for AnyEvent. |
|
|
517 | |
|
|
518 | =item L<IO::Lambda> |
|
|
519 | |
|
|
520 | The lambda approach to I/O - don't ask, look there. Can use AnyEvent. |
|
|
521 | |
|
|
522 | =item L<IO::AIO> |
|
|
523 | |
|
|
524 | Truly asynchronous I/O, should be in the toolbox of every event |
|
|
525 | programmer. Can be trivially made to use AnyEvent. |
|
|
526 | |
|
|
527 | =item L<BDB> |
|
|
528 | |
|
|
529 | Truly asynchronous Berkeley DB access. Can be trivially made to use |
|
|
530 | AnyEvent. |
|
|
531 | |
|
|
532 | =back |
457 | |
533 | |
458 | =cut |
534 | =cut |
459 | |
535 | |
460 | package AnyEvent; |
536 | package AnyEvent; |
461 | |
537 | |
… | |
… | |
889 | }); |
965 | }); |
890 | |
966 | |
891 | $quit->wait; |
967 | $quit->wait; |
892 | |
968 | |
893 | |
969 | |
894 | =head1 BENCHMARK |
970 | =head1 BENCHMARKS |
895 | |
971 | |
896 | To give you an idea of the performance and overheads that AnyEvent adds |
972 | To give you an idea of the performance and overheads that AnyEvent adds |
897 | over the event loops themselves (and to give you an impression of the |
973 | over the event loops themselves and to give you an impression of the speed |
898 | speed of various event loops), here is a benchmark of various supported |
974 | of various event loops I prepared some benchmarks. |
899 | event models natively and with anyevent. The benchmark creates a lot of |
975 | |
900 | timers (with a zero timeout) and I/O watchers (watching STDOUT, a pty, to |
976 | =head2 BENCHMARKING ANYEVENT OVERHEAD |
|
|
977 | |
|
|
978 | Here is a benchmark of various supported event models used natively and |
|
|
979 | through anyevent. The benchmark creates a lot of timers (with a zero |
|
|
980 | timeout) and I/O watchers (watching STDOUT, a pty, to become writable, |
901 | become writable, which it is), lets them fire exactly once and destroys |
981 | which it is), lets them fire exactly once and destroys them again. |
902 | them again. |
|
|
903 | |
982 | |
904 | Rewriting the benchmark to use many different sockets instead of using |
983 | Source code for this benchmark is found as F<eg/bench> in the AnyEvent |
905 | the same filehandle for all I/O watchers results in a much longer runtime |
984 | distribution. |
906 | (socket creation is expensive), but qualitatively the same figures, so it |
|
|
907 | was not used. |
|
|
908 | |
985 | |
909 | =head2 Explanation of the columns |
986 | =head3 Explanation of the columns |
910 | |
987 | |
911 | I<watcher> is the number of event watchers created/destroyed. Since |
988 | I<watcher> is the number of event watchers created/destroyed. Since |
912 | different event models feature vastly different performances, each event |
989 | different event models feature vastly different performances, each event |
913 | loop was given a number of watchers so that overall runtime is acceptable |
990 | loop was given a number of watchers so that overall runtime is acceptable |
914 | and similar between tested event loop (and keep them from crashing): Glib |
991 | and similar between tested event loop (and keep them from crashing): Glib |
… | |
… | |
930 | signal the end of this phase. |
1007 | signal the end of this phase. |
931 | |
1008 | |
932 | I<destroy> is the time, in microseconds, that it takes to destroy a single |
1009 | I<destroy> is the time, in microseconds, that it takes to destroy a single |
933 | watcher. |
1010 | watcher. |
934 | |
1011 | |
935 | =head2 Results |
1012 | =head3 Results |
936 | |
1013 | |
937 | name watchers bytes create invoke destroy comment |
1014 | name watchers bytes create invoke destroy comment |
938 | EV/EV 400000 244 0.56 0.46 0.31 EV native interface |
1015 | EV/EV 400000 244 0.56 0.46 0.31 EV native interface |
939 | EV/Any 100000 244 2.50 0.46 0.29 EV + AnyEvent watchers |
1016 | EV/Any 100000 244 2.50 0.46 0.29 EV + AnyEvent watchers |
940 | CoroEV/Any 100000 244 2.49 0.44 0.29 coroutines + Coro::Signal |
1017 | CoroEV/Any 100000 244 2.49 0.44 0.29 coroutines + Coro::Signal |
941 | Perl/Any 100000 513 4.92 0.87 1.12 pure perl implementation |
1018 | Perl/Any 100000 513 4.92 0.87 1.12 pure perl implementation |
942 | Event/Event 16000 516 31.88 31.30 0.85 Event native interface |
1019 | Event/Event 16000 516 31.88 31.30 0.85 Event native interface |
943 | Event/Any 16000 936 39.17 33.63 1.43 Event + AnyEvent watchers |
1020 | Event/Any 16000 590 35.75 31.42 1.08 Event + AnyEvent watchers |
944 | Glib/Any 16000 1357 98.22 12.41 54.00 quadratic behaviour |
1021 | Glib/Any 16000 1357 98.22 12.41 54.00 quadratic behaviour |
945 | Tk/Any 2000 1860 26.97 67.98 14.00 SEGV with >> 2000 watchers |
1022 | Tk/Any 2000 1860 26.97 67.98 14.00 SEGV with >> 2000 watchers |
946 | POE/Event 2000 6644 108.64 736.02 14.73 via POE::Loop::Event |
1023 | POE/Event 2000 6644 108.64 736.02 14.73 via POE::Loop::Event |
947 | POE/Select 2000 6343 94.13 809.12 565.96 via POE::Loop::Select |
1024 | POE/Select 2000 6343 94.13 809.12 565.96 via POE::Loop::Select |
948 | |
1025 | |
949 | =head2 Discussion |
1026 | =head3 Discussion |
950 | |
1027 | |
951 | The benchmark does I<not> measure scalability of the event loop very |
1028 | The benchmark does I<not> measure scalability of the event loop very |
952 | well. For example, a select-based event loop (such as the pure perl one) |
1029 | well. For example, a select-based event loop (such as the pure perl one) |
953 | can never compete with an event loop that uses epoll when the number of |
1030 | can never compete with an event loop that uses epoll when the number of |
954 | file descriptors grows high. In this benchmark, all events become ready at |
1031 | file descriptors grows high. In this benchmark, all events become ready at |
955 | the same time, so select/poll-based implementations get an unnatural speed |
1032 | the same time, so select/poll-based implementations get an unnatural speed |
956 | boost. |
1033 | boost. |
957 | |
1034 | |
|
|
1035 | Also, note that the number of watchers usually has a nonlinear effect on |
|
|
1036 | overall speed, that is, creating twice as many watchers doesn't take twice |
|
|
1037 | the time - usually it takes longer. This puts event loops tested with a |
|
|
1038 | higher number of watchers at a disadvantage. |
|
|
1039 | |
|
|
1040 | To put the range of results into perspective, consider that on the |
|
|
1041 | benchmark machine, handling an event takes roughly 1600 CPU cycles with |
|
|
1042 | EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000 CPU |
|
|
1043 | cycles with POE. |
|
|
1044 | |
958 | C<EV> is the sole leader regarding speed and memory use, which are both |
1045 | C<EV> is the sole leader regarding speed and memory use, which are both |
959 | maximal/minimal, respectively. Even when going through AnyEvent, it uses |
1046 | maximal/minimal, respectively. Even when going through AnyEvent, it uses |
960 | far less memory than any other event loop and is still faster than Event |
1047 | far less memory than any other event loop and is still faster than Event |
961 | natively. |
1048 | natively. |
962 | |
1049 | |
963 | The pure perl implementation is hit in a few sweet spots (both the |
1050 | The pure perl implementation is hit in a few sweet spots (both the |
964 | zero timeout and the use of a single fd hit optimisations in the perl |
1051 | constant timeout and the use of a single fd hit optimisations in the perl |
965 | interpreter and the backend itself, and all watchers become ready at the |
1052 | interpreter and the backend itself). Nevertheless this shows that it |
966 | same time). Nevertheless this shows that it adds very little overhead in |
1053 | adds very little overhead in itself. Like any select-based backend its |
967 | itself. Like any select-based backend its performance becomes really bad |
1054 | performance becomes really bad with lots of file descriptors (and few of |
968 | with lots of file descriptors (and few of them active), of course, but |
1055 | them active), of course, but this was not subject of this benchmark. |
969 | this was not subject of this benchmark. |
|
|
970 | |
1056 | |
971 | The C<Event> module has a relatively high setup and callback invocation cost, |
1057 | The C<Event> module has a relatively high setup and callback invocation |
972 | but overall scores on the third place. |
1058 | cost, but overall scores in on the third place. |
973 | |
1059 | |
974 | C<Glib>'s memory usage is quite a bit bit higher, but it features a |
1060 | C<Glib>'s memory usage is quite a bit higher, but it features a |
975 | faster callback invocation and overall ends up in the same class as |
1061 | faster callback invocation and overall ends up in the same class as |
976 | C<Event>. However, Glib scales extremely badly, doubling the number of |
1062 | C<Event>. However, Glib scales extremely badly, doubling the number of |
977 | watchers increases the processing time by more than a factor of four, |
1063 | watchers increases the processing time by more than a factor of four, |
978 | making it completely unusable when using larger numbers of watchers |
1064 | making it completely unusable when using larger numbers of watchers |
979 | (note that only a single file descriptor was used in the benchmark, so |
1065 | (note that only a single file descriptor was used in the benchmark, so |
… | |
… | |
982 | The C<Tk> adaptor works relatively well. The fact that it crashes with |
1068 | The C<Tk> adaptor works relatively well. The fact that it crashes with |
983 | more than 2000 watchers is a big setback, however, as correctness takes |
1069 | more than 2000 watchers is a big setback, however, as correctness takes |
984 | precedence over speed. Nevertheless, its performance is surprising, as the |
1070 | precedence over speed. Nevertheless, its performance is surprising, as the |
985 | file descriptor is dup()ed for each watcher. This shows that the dup() |
1071 | file descriptor is dup()ed for each watcher. This shows that the dup() |
986 | employed by some adaptors is not a big performance issue (it does incur a |
1072 | employed by some adaptors is not a big performance issue (it does incur a |
987 | hidden memory cost inside the kernel, though, that is not reflected in the |
1073 | hidden memory cost inside the kernel which is not reflected in the figures |
988 | figures above). |
1074 | above). |
989 | |
1075 | |
990 | C<POE>, regardless of underlying event loop (wether using its pure perl |
1076 | C<POE>, regardless of underlying event loop (whether using its pure |
991 | select-based backend or the Event module) shows abysmal performance and |
1077 | perl select-based backend or the Event module, the POE-EV backend |
|
|
1078 | couldn't be tested because it wasn't working) shows abysmal performance |
992 | memory usage: Watchers use almost 30 times as much memory as EV watchers, |
1079 | and memory usage: Watchers use almost 30 times as much memory as |
993 | and 10 times as much memory as both Event or EV via AnyEvent. Watcher |
1080 | EV watchers, and 10 times as much memory as Event (the high memory |
|
|
1081 | requirements are caused by requiring a session for each watcher). Watcher |
994 | invocation is almost 900 times slower than with AnyEvent's pure perl |
1082 | invocation speed is almost 900 times slower than with AnyEvent's pure perl |
995 | implementation. The design of the POE adaptor class in AnyEvent can not |
1083 | implementation. The design of the POE adaptor class in AnyEvent can not |
996 | really account for this, as session creation overhead is small compared |
1084 | really account for this, as session creation overhead is small compared |
997 | to execution of the state machine, which is coded pretty optimally within |
1085 | to execution of the state machine, which is coded pretty optimally within |
998 | L<AnyEvent::Impl::POE>. POE simply seems to be abysmally slow. |
1086 | L<AnyEvent::Impl::POE>. POE simply seems to be abysmally slow. |
999 | |
1087 | |
1000 | =head2 Summary |
1088 | =head3 Summary |
1001 | |
1089 | |
|
|
1090 | =over 4 |
|
|
1091 | |
1002 | Using EV through AnyEvent is faster than any other event loop, but most |
1092 | =item * Using EV through AnyEvent is faster than any other event loop |
1003 | event loops have acceptable performance with or without AnyEvent. |
1093 | (even when used without AnyEvent), but most event loops have acceptable |
|
|
1094 | performance with or without AnyEvent. |
1004 | |
1095 | |
1005 | The overhead AnyEvent adds is usually much smaller than the overhead of |
1096 | =item * The overhead AnyEvent adds is usually much smaller than the overhead of |
1006 | the actual event loop, only with extremely fast event loops such as the EV |
1097 | the actual event loop, only with extremely fast event loops such as EV |
1007 | adds AnyEvent significant overhead. |
1098 | adds AnyEvent significant overhead. |
1008 | |
1099 | |
1009 | And you should simply avoid POE like the plague if you want performance or |
1100 | =item * You should avoid POE like the plague if you want performance or |
1010 | reasonable memory usage. |
1101 | reasonable memory usage. |
|
|
1102 | |
|
|
1103 | =back |
|
|
1104 | |
|
|
1105 | =head2 BENCHMARKING THE LARGE SERVER CASE |
|
|
1106 | |
|
|
1107 | This benchmark atcually benchmarks the event loop itself. It works by |
|
|
1108 | creating a number of "servers": each server consists of a socketpair, a |
|
|
1109 | timeout watcher that gets reset on activity (but never fires), and an I/O |
|
|
1110 | watcher waiting for input on one side of the socket. Each time the socket |
|
|
1111 | watcher reads a byte it will write that byte to a random other "server". |
|
|
1112 | |
|
|
1113 | The effect is that there will be a lot of I/O watchers, only part of which |
|
|
1114 | are active at any one point (so there is a constant number of active |
|
|
1115 | fds for each loop iterstaion, but which fds these are is random). The |
|
|
1116 | timeout is reset each time something is read because that reflects how |
|
|
1117 | most timeouts work (and puts extra pressure on the event loops). |
|
|
1118 | |
|
|
1119 | In this benchmark, we use 10000 socketpairs (20000 sockets), of which 100 |
|
|
1120 | (1%) are active. This mirrors the activity of large servers with many |
|
|
1121 | connections, most of which are idle at any one point in time. |
|
|
1122 | |
|
|
1123 | Source code for this benchmark is found as F<eg/bench2> in the AnyEvent |
|
|
1124 | distribution. |
|
|
1125 | |
|
|
1126 | =head3 Explanation of the columns |
|
|
1127 | |
|
|
1128 | I<sockets> is the number of sockets, and twice the number of "servers" (as |
|
|
1129 | each server has a read and write socket end). |
|
|
1130 | |
|
|
1131 | I<create> is the time it takes to create a socketpair (which is |
|
|
1132 | nontrivial) and two watchers: an I/O watcher and a timeout watcher. |
|
|
1133 | |
|
|
1134 | I<request>, the most important value, is the time it takes to handle a |
|
|
1135 | single "request", that is, reading the token from the pipe and forwarding |
|
|
1136 | it to another server. This includes deleting the old timeout and creating |
|
|
1137 | a new one that moves the timeout into the future. |
|
|
1138 | |
|
|
1139 | =head3 Results |
|
|
1140 | |
|
|
1141 | name sockets create request |
|
|
1142 | EV 20000 69.01 11.16 |
|
|
1143 | Perl 20000 73.32 35.87 |
|
|
1144 | Event 20000 212.62 257.32 |
|
|
1145 | Glib 20000 651.16 1896.30 |
|
|
1146 | POE 20000 349.67 12317.24 uses POE::Loop::Event |
|
|
1147 | |
|
|
1148 | =head3 Discussion |
|
|
1149 | |
|
|
1150 | This benchmark I<does> measure scalability and overall performance of the |
|
|
1151 | particular event loop. |
|
|
1152 | |
|
|
1153 | EV is again fastest. Since it is using epoll on my system, the setup time |
|
|
1154 | is relatively high, though. |
|
|
1155 | |
|
|
1156 | Perl surprisingly comes second. It is much faster than the C-based event |
|
|
1157 | loops Event and Glib. |
|
|
1158 | |
|
|
1159 | Event suffers from high setup time as well (look at its code and you will |
|
|
1160 | understand why). Callback invocation also has a high overhead compared to |
|
|
1161 | the C<< $_->() for .. >>-style loop that the Perl event loop uses. Event |
|
|
1162 | uses select or poll in basically all documented configurations. |
|
|
1163 | |
|
|
1164 | Glib is hit hard by its quadratic behaviour w.r.t. many watchers. It |
|
|
1165 | clearly fails to perform with many filehandles or in busy servers. |
|
|
1166 | |
|
|
1167 | POE is still completely out of the picture, taking over 1000 times as long |
|
|
1168 | as EV, and over 100 times as long as the Perl implementation, even though |
|
|
1169 | it uses a C-based event loop in this case. |
|
|
1170 | |
|
|
1171 | =head3 Summary |
|
|
1172 | |
|
|
1173 | =over 4 |
|
|
1174 | |
|
|
1175 | =item * The pure perl implementation performs extremely well, considering |
|
|
1176 | that it uses select. |
|
|
1177 | |
|
|
1178 | =item * Avoid Glib or POE in large projects where performance matters. |
|
|
1179 | |
|
|
1180 | =back |
|
|
1181 | |
|
|
1182 | =head2 BENCHMARKING SMALL SERVERS |
|
|
1183 | |
|
|
1184 | While event loops should scale (and select-based ones do not...) even to |
|
|
1185 | large servers, most programs we (or I :) actually write have only a few |
|
|
1186 | I/O watchers. |
|
|
1187 | |
|
|
1188 | In this benchmark, I use the same benchmark program as in the large server |
|
|
1189 | case, but it uses only eight "servers", of which three are active at any |
|
|
1190 | one time. This should reflect performance for a small server relatively |
|
|
1191 | well. |
|
|
1192 | |
|
|
1193 | The columns are identical to the previous table. |
|
|
1194 | |
|
|
1195 | =head3 Results |
|
|
1196 | |
|
|
1197 | name sockets create request |
|
|
1198 | EV 16 20.00 6.54 |
|
|
1199 | Perl 16 25.75 12.62 |
|
|
1200 | Event 16 81.27 35.86 |
|
|
1201 | Glib 16 32.63 15.48 |
|
|
1202 | POE 16 261.87 276.28 uses POE::Loop::Event |
|
|
1203 | |
|
|
1204 | =head3 Discussion |
|
|
1205 | |
|
|
1206 | The benchmark tries to test the performance of a typical small |
|
|
1207 | server. While knowing how various event loops perform is interesting, keep |
|
|
1208 | in mind that their overhead in this case is usually not as important, due |
|
|
1209 | to the small absolute number of watchers (that is, you need efficiency and |
|
|
1210 | speed most when you have lots of watchers, not when you only have a few of |
|
|
1211 | them). |
|
|
1212 | |
|
|
1213 | EV is again fastest. |
|
|
1214 | |
|
|
1215 | The C-based event loops Event and Glib come in second this time, as the |
|
|
1216 | overhead of running an iteration is much smaller in C than in Perl (little |
|
|
1217 | code to execute in the inner loop, and perl's function calling overhead is |
|
|
1218 | high, and updating all the data structures is costly). |
|
|
1219 | |
|
|
1220 | The pure perl event loop is much slower, but still competitive. |
|
|
1221 | |
|
|
1222 | POE also performs much better in this case, but is is still far behind the |
|
|
1223 | others. |
|
|
1224 | |
|
|
1225 | =head3 Summary |
|
|
1226 | |
|
|
1227 | =over 4 |
|
|
1228 | |
|
|
1229 | =item * C-based event loops perform very well with small number of |
|
|
1230 | watchers, as the management overhead dominates. |
|
|
1231 | |
|
|
1232 | =back |
1011 | |
1233 | |
1012 | |
1234 | |
1013 | =head1 FORK |
1235 | =head1 FORK |
1014 | |
1236 | |
1015 | Most event libraries are not fork-safe. The ones who are usually are |
1237 | Most event libraries are not fork-safe. The ones who are usually are |