[ViewVC] Diff of: cvs/AnyEvent/lib/AnyEvent.pm

Comparing AnyEvent/lib/AnyEvent.pm (file contents):
Revision 1.84 by root, Fri Apr 25 13:48:42 2008 UTC vs.
Revision 1.98 by root, Sun Apr 27 16:31:48 2008 UTC

…		…
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		70
		71	#TODO#
		72
		73	Net::IRC3
		74	AnyEvent::HTTPD
		75	AnyEvent::DNS
		76	IO::AnyEvent
		77	Net::FPing
		78	Net::XMPP2
		79	Coro
		80
		81	AnyEvent::IRC
		82	AnyEvent::HTTPD
		83	AnyEvent::DNS
		84	AnyEvent::Handle
		85	AnyEvent::Socket
		86	AnyEvent::FPing
		87	AnyEvent::XMPP
		88	AnyEvent::SNMP
		89	Coro
71		90
72	=head1 DESCRIPTION	91	=head1 DESCRIPTION
73		92
74	L<AnyEvent> provides an identical interface to multiple event loops. This	93	L<AnyEvent> provides an identical interface to multiple event loops. This
75	allows module authors to utilise an event loop without forcing module	94	allows module authors to utilise an event loop without forcing module
…		…
141	=head2 I/O WATCHERS	160	=head2 I/O WATCHERS
142		161
143	You can create an I/O watcher by calling the C<< AnyEvent->io >> method	162	You can create an I/O watcher by calling the C<< AnyEvent->io >> method
144	with the following mandatory key-value pairs as arguments:	163	with the following mandatory key-value pairs as arguments:
145		164
146	C<fh> the Perl I<file handle> (I<not> file descriptor) to watch for	165	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	166	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,	167	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	168	respectively. C<cb> is the callback to invoke each time the file handle
150	becomes ready.	169	becomes ready.
		170
		171	Although the callback might get passed parameters, their value and
		172	presence is undefined and you cannot rely on them. Portable AnyEvent
		173	callbacks cannot use arguments passed to I/O watcher callbacks.
151		174
152	The I/O watcher might use the underlying file descriptor or a copy of it.	175	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	176	You must not close a file handle as long as any watcher is active on the
154	underlying file descriptor.	177	underlying file descriptor.
155		178
156	Some event loops issue spurious readyness notifications, so you should	179	Some event loops issue spurious readyness notifications, so you should
157	always use non-blocking calls when reading/writing from/to your file	180	always use non-blocking calls when reading/writing from/to your file
158	handles.	181	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		182
164	Example:	183	Example:
165		184
166	# wait for readability of STDIN, then read a line and disable the watcher	185	# 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 {	186	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {
…		…
174		193
175	You can create a time watcher by calling the C<< AnyEvent->timer >>	194	You can create a time watcher by calling the C<< AnyEvent->timer >>
176	method with the following mandatory arguments:	195	method with the following mandatory arguments:
177		196
178	C<after> specifies after how many seconds (fractional values are	197	C<after> specifies after how many seconds (fractional values are
179	supported) should the timer activate. C<cb> the callback to invoke in that	198	supported) the callback should be invoked. C<cb> is the callback to invoke
180	case.	199	in that case.
		200
		201	Although the callback might get passed parameters, their value and
		202	presence is undefined and you cannot rely on them. Portable AnyEvent
		203	callbacks cannot use arguments passed to time watcher callbacks.
181		204
182	The timer callback will be invoked at most once: if you want a repeating	205	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	206	timer you have to create a new watcher (this is a limitation by both Tk
184	and Glib).	207	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		208
190	Example:	209	Example:
191		210
192	# fire an event after 7.7 seconds	211	# fire an event after 7.7 seconds
193	my $w = AnyEvent->timer (after => 7.7, cb => sub {	212	my $w = AnyEvent->timer (after => 7.7, cb => sub {
…		…
234		253
235	You can watch for signals using a signal watcher, C<signal> is the signal	254	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	255	I<name> without any C<SIG> prefix, C<cb> is the Perl callback to
237	be invoked whenever a signal occurs.	256	be invoked whenever a signal occurs.
238		257
		258	Although the callback might get passed parameters, their value and
		259	presence is undefined and you cannot rely on them. Portable AnyEvent
		260	callbacks cannot use arguments passed to signal watcher callbacks.
		261
239	Multiple signal occurances can be clumped together into one callback	262	Multiple signal occurances can be clumped together into one callback
240	invocation, and callback invocation will be synchronous. synchronous means	263	invocation, and callback invocation will be synchronous. synchronous means
241	that it might take a while until the signal gets handled by the process,	264	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.	265	but it is guarenteed not to interrupt any other callbacks.
243		266
…		…
257		280
258	The child process is specified by the C<pid> argument (if set to C<0>, it	281	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	282	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	283	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	284	signal handler for C<SIGCHLD>. The callback will be called with the pid
262	and exit status (as returned by waitpid).	285	and exit status (as returned by waitpid), so unlike other watcher types,
		286	you I<can> rely on child watcher callback arguments.
263		287
264	There is a slight catch to child watchers, however: you usually start them	288	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	289	I<after> the child process was created, and this means the process could
266	have exited already (and no SIGCHLD will be sent anymore).	290	have exited already (and no SIGCHLD will be sent anymore).
267		291
…		…
889	});	913	});
890		914
891	$quit->wait;	915	$quit->wait;
892		916
893		917
894	=head1 BENCHMARK	918	=head1 BENCHMARKS
895		919
896	To give you an idea of the performance and overheads that AnyEvent adds	920	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	921	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	922	of various event loops I prepared some benchmarks.
899	event models natively and with anyevent. The benchmark creates a lot of	923
900	timers (with a zero timeout) and I/O watchers (watching STDOUT, a pty, to	924	=head2 BENCHMARKING ANYEVENT OVERHEAD
		925
		926	Here is a benchmark of various supported event models used natively and
		927	through anyevent. The benchmark creates a lot of timers (with a zero
		928	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	929	which it is), lets them fire exactly once and destroys them again.
902	them again.
903		930
904	Rewriting the benchmark to use many different sockets instead of using	931	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	932	distribution.
906	(socket creation is expensive), but qualitatively the same figures, so it
907	was not used.
908		933
909	=head2 Explanation of the columns	934	=head3 Explanation of the columns
910		935
911	I<watcher> is the number of event watchers created/destroyed. Since	936	I<watcher> is the number of event watchers created/destroyed. Since
912	different event models feature vastly different performances, each event	937	different event models feature vastly different performances, each event
913	loop was given a number of watchers so that overall runtime is acceptable	938	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	939	and similar between tested event loop (and keep them from crashing): Glib
…		…
930	signal the end of this phase.	955	signal the end of this phase.
931		956
932	I<destroy> is the time, in microseconds, that it takes to destroy a single	957	I<destroy> is the time, in microseconds, that it takes to destroy a single
933	watcher.	958	watcher.
934		959
935	=head2 Results	960	=head3 Results
936		961
937	name watchers bytes create invoke destroy comment	962	name watchers bytes create invoke destroy comment
938	EV/EV 400000 244 0.56 0.46 0.31 EV native interface	963	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	964	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	965	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	966	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	967	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	968	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	969	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	970	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	971	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	972	POE/Select 2000 6343 94.13 809.12 565.96 via POE::Loop::Select
948		973
949	=head2 Discussion	974	=head3 Discussion
950		975
951	The benchmark does I<not> measure scalability of the event loop very	976	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)	977	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	978	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	979	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	980	the same time, so select/poll-based implementations get an unnatural speed
956	boost.	981	boost.
957		982
		983	Also, note that the number of watchers usually has a nonlinear effect on
		984	overall speed, that is, creating twice as many watchers doesn't take twice
		985	the time - usually it takes longer. This puts event loops tested with a
		986	higher number of watchers at a disadvantage.
		987
		988	To put the range of results into perspective, consider that on the
		989	benchmark machine, handling an event takes roughly 1600 CPU cycles with
		990	EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000 CPU
		991	cycles with POE.
		992
958	C<EV> is the sole leader regarding speed and memory use, which are both	993	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	994	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	995	far less memory than any other event loop and is still faster than Event
961	natively.	996	natively.
962		997
963	The pure perl implementation is hit in a few sweet spots (both the	998	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	999	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	1000	interpreter and the backend itself). Nevertheless this shows that it
966	same time). Nevertheless this shows that it adds very little overhead in	1001	adds very little overhead in itself. Like any select-based backend its
967	itself. Like any select-based backend its performance becomes really bad	1002	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	1003	them active), of course, but this was not subject of this benchmark.
969	this was not subject of this benchmark.
970		1004
971	The C<Event> module has a relatively high setup and callback invocation cost,	1005	The C<Event> module has a relatively high setup and callback invocation
972	but overall scores on the third place.	1006	cost, but overall scores in on the third place.
973		1007
974	C<Glib>'s memory usage is quite a bit bit higher, but it features a	1008	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	1009	faster callback invocation and overall ends up in the same class as
976	C<Event>. However, Glib scales extremely badly, doubling the number of	1010	C<Event>. However, Glib scales extremely badly, doubling the number of
977	watchers increases the processing time by more than a factor of four,	1011	watchers increases the processing time by more than a factor of four,
978	making it completely unusable when using larger numbers of watchers	1012	making it completely unusable when using larger numbers of watchers
979	(note that only a single file descriptor was used in the benchmark, so	1013	(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	1016	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	1017	more than 2000 watchers is a big setback, however, as correctness takes
984	precedence over speed. Nevertheless, its performance is surprising, as the	1018	precedence over speed. Nevertheless, its performance is surprising, as the
985	file descriptor is dup()ed for each watcher. This shows that the dup()	1019	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	1020	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	1021	hidden memory cost inside the kernel which is not reflected in the figures
988	figures above).	1022	above).
989		1023
990	C<POE>, regardless of underlying event loop (wether using its pure perl	1024	C<POE>, regardless of underlying event loop (whether using its pure
991	select-based backend or the Event module) shows abysmal performance and	1025	perl select-based backend or the Event module, the POE-EV backend
		1026	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,	1027	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	1028	EV watchers, and 10 times as much memory as Event (the high memory
		1029	requirements are caused by requiring a session for each watcher). Watcher
994	invocation is almost 900 times slower than with AnyEvent's pure perl	1030	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	1031	implementation. The design of the POE adaptor class in AnyEvent can not
996	really account for this, as session creation overhead is small compared	1032	really account for this, as session creation overhead is small compared
997	to execution of the state machine, which is coded pretty optimally within	1033	to execution of the state machine, which is coded pretty optimally within
998	L<AnyEvent::Impl::POE>. POE simply seems to be abysmally slow.	1034	L<AnyEvent::Impl::POE>. POE simply seems to be abysmally slow.
999		1035
1000	=head2 Summary	1036	=head3 Summary
1001		1037
		1038	=over 4
		1039
1002	Using EV through AnyEvent is faster than any other event loop, but most	1040	=item * Using EV through AnyEvent is faster than any other event loop
1003	event loops have acceptable performance with or without AnyEvent.	1041	(even when used without AnyEvent), but most event loops have acceptable
		1042	performance with or without AnyEvent.
1004		1043
1005	The overhead AnyEvent adds is usually much smaller than the overhead of	1044	=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	1045	the actual event loop, only with extremely fast event loops such as EV
1007	adds AnyEvent significant overhead.	1046	adds AnyEvent significant overhead.
1008		1047
1009	And you should simply avoid POE like the plague if you want performance or	1048	=item * You should avoid POE like the plague if you want performance or
1010	reasonable memory usage.	1049	reasonable memory usage.
		1050
		1051	=back
		1052
		1053	=head2 BENCHMARKING THE LARGE SERVER CASE
		1054
		1055	This benchmark atcually benchmarks the event loop itself. It works by
		1056	creating a number of "servers": each server consists of a socketpair, a
		1057	timeout watcher that gets reset on activity (but never fires), and an I/O
		1058	watcher waiting for input on one side of the socket. Each time the socket
		1059	watcher reads a byte it will write that byte to a random other "server".
		1060
		1061	The effect is that there will be a lot of I/O watchers, only part of which
		1062	are active at any one point (so there is a constant number of active
		1063	fds for each loop iterstaion, but which fds these are is random). The
		1064	timeout is reset each time something is read because that reflects how
		1065	most timeouts work (and puts extra pressure on the event loops).
		1066
		1067	In this benchmark, we use 10000 socketpairs (20000 sockets), of which 100
		1068	(1%) are active. This mirrors the activity of large servers with many
		1069	connections, most of which are idle at any one point in time.
		1070
		1071	Source code for this benchmark is found as F<eg/bench2> in the AnyEvent
		1072	distribution.
		1073
		1074	=head3 Explanation of the columns
		1075
		1076	I<sockets> is the number of sockets, and twice the number of "servers" (as
		1077	each server has a read and write socket end).
		1078
		1079	I<create> is the time it takes to create a socketpair (which is
		1080	nontrivial) and two watchers: an I/O watcher and a timeout watcher.
		1081
		1082	I<request>, the most important value, is the time it takes to handle a
		1083	single "request", that is, reading the token from the pipe and forwarding
		1084	it to another server. This includes deleting the old timeout and creating
		1085	a new one that moves the timeout into the future.
		1086
		1087	=head3 Results
		1088
		1089	name sockets create request
		1090	EV 20000 69.01 11.16
		1091	Perl 20000 75.28 112.76
		1092	Event 20000 212.62 257.32
		1093	Glib 20000 651.16 1896.30
		1094	POE 20000 349.67 12317.24 uses POE::Loop::Event
		1095
		1096	=head3 Discussion
		1097
		1098	This benchmark I<does> measure scalability and overall performance of the
		1099	particular event loop.
		1100
		1101	EV is again fastest. Since it is using epoll on my system, the setup time
		1102	is relatively high, though.
		1103
		1104	Perl surprisingly comes second. It is much faster than the C-based event
		1105	loops Event and Glib.
		1106
		1107	Event suffers from high setup time as well (look at its code and you will
		1108	understand why). Callback invocation also has a high overhead compared to
		1109	the C<< $_->() for .. >>-style loop that the Perl event loop uses. Event
		1110	uses select or poll in basically all documented configurations.
		1111
		1112	Glib is hit hard by its quadratic behaviour w.r.t. many watchers. It
		1113	clearly fails to perform with many filehandles or in busy servers.
		1114
		1115	POE is still completely out of the picture, taking over 1000 times as long
		1116	as EV, and over 100 times as long as the Perl implementation, even though
		1117	it uses a C-based event loop in this case.
		1118
		1119	=head3 Summary
		1120
		1121	=over 4
		1122
		1123	=item * The pure perl implementation performs extremely well, considering
		1124	that it uses select.
		1125
		1126	=item * Avoid Glib or POE in large projects where performance matters.
		1127
		1128	=back
		1129
		1130	=head2 BENCHMARKING SMALL SERVERS
		1131
		1132	While event loops should scale (and select-based ones do not...) even to
		1133	large servers, most programs we (or I :) actually write have only a few
		1134	I/O watchers.
		1135
		1136	In this benchmark, I use the same benchmark program as in the large server
		1137	case, but it uses only eight "servers", of which three are active at any
		1138	one time. This should reflect performance for a small server relatively
		1139	well.
		1140
		1141	The columns are identical to the previous table.
		1142
		1143	=head3 Results
		1144
		1145	name sockets create request
		1146	EV 16 20.00 6.54
		1147	Event 16 81.27 35.86
		1148	Glib 16 32.63 15.48
		1149	Perl 16 24.62 162.37
		1150	POE 16 261.87 276.28 uses POE::Loop::Event
		1151
		1152	=head3 Discussion
		1153
		1154	The benchmark tries to test the performance of a typical small
		1155	server. While knowing how various event loops perform is interesting, keep
		1156	in mind that their overhead in this case is usually not as important, due
		1157	to the small absolute number of watchers (that is, you need efficiency and
		1158	speed most when you have lots of watchers, not when you only have a few of
		1159	them).
		1160
		1161	EV is again fastest.
		1162
		1163	The C-based event loops Event and Glib come in second this time, as the
		1164	overhead of running an iteration is much smaller in C than in Perl (little
		1165	code to execute in the inner loop, and perl's function calling overhead is
		1166	high, and updating all the data structures is costly).
		1167
		1168	The pure perl event loop is much slower, but still competitive.
		1169
		1170	POE also performs much better in this case, but is is still far behind the
		1171	others.
		1172
		1173	=head3 Summary
		1174
		1175	=over 4
		1176
		1177	=item * C-based event loops perform very well with small number of
		1178	watchers, as the management overhead dominates.
		1179
		1180	=back
1011		1181
1012		1182
1013	=head1 FORK	1183	=head1 FORK
1014		1184
1015	Most event libraries are not fork-safe. The ones who are usually are	1185	Most event libraries are not fork-safe. The ones who are usually are

Diff Legend

-–
+Removed lines
-+
+Added lines
-<
+Changed lines
->
+Changed lines

Comparing AnyEvent/lib/AnyEvent.pm (file contents): Revision 1.84 by root, Fri Apr 25 13:48:42 2008 UTC vs. Revision 1.98 by root, Sun Apr 27 16:31:48 2008 UTC

Diff Legend

Comparing AnyEvent/lib/AnyEvent.pm (file contents):
Revision 1.84 by root, Fri Apr 25 13:48:42 2008 UTC vs.
Revision 1.98 by root, Sun Apr 27 16:31:48 2008 UTC