1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
|
<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML//EN">
<html>
<head>
<title>COS Event Channel Configuration</title>
<!-- $Id$ -->
</head>
<body>
<CENTER>
<h1>COS Event Channel Configuration</h1>
</CENTER>
<H3>Run-time Configuration</H3>
<P>The new implementation of the COS Event Service uses a
factory to build all the objects and strategies it requires.
The factory can be dynamically loaded using ACE Service
Configurator, this is extremely convenient because the factory
can also parse options in the Service Configurator script file.
</P>
<P>The current implementation provides a default implementation
for this Factory, this document describes the options used by
this default implementation.
Users can define their own implementation, with new ad-hoc
strategies or with pre-selected strategies.
</P>
<H3>The configuration file</H3>
<P>The COS channel uses the same service configurator
file that the ORB uses, the default name for this file is
<CODE>svc.conf</CODE>, but the ORB option
<CODE>-ORBSvcConf</CODE> can be used to override this.
The format of the file is described in detail in
the service configurator documentation, but the relevant section
for the event channel looks like this:
</P>
<PRE>
# Comments go here...
# More comments if you want to...
static CEC_Factory "-CECDispatching reactive ....."
</PRE>
<P>All the event service factory options start with
<CODE>-CEC</CODE></P>
<H3>The options</H3>
<P><TABLE BORDER="2" CELLSPACING="2" CELLPADDING="0">
<TR>
<TH>Option</TH>
<TH>Description</TH>
</TR>
<!-- <TR NAME="CECDispatching"> -->
<TR>
<TD><CODE>-CECDispatching</CODE>
<EM>dispatching_strategy</EM>
</TD>
<TD>Select the dispatching strategy used by the COS
event service. A <EM>reactive</EM> strategy will use the
same thread that received the event from the supplier to
push the event to all the consumers.
The <EM>mt</EM> strategy will also use a pool of threads,
but the thread to dispatch is randomly selected.
</TD>
</TR>
<!-- <TR NAME="CECDispatchingThreads"> -->
<TR>
<TD><CODE>-CECDispatchingThreads</CODE>
<EM>number_of_threads</EM>
</TD>
<TD>Select the number of threads used by the <EM>mt</EM>
dispatching strategy.
</TD>
</TR>
<!-- <TR NAME="CECProxyConsumerLock"> -->
<TR>
<TD><CODE>-CECProxyConsumerLock</CODE>
<EM>lock_type</EM>
</TD>
<TD>Select the lock type (<EM>null</EM>, <EM>thread</EM> or
<EM>recursive</EM>) to synchronize access to the
ProxyPushConsumer state.
</TD>
</TR>
<!-- <TR NAME="CECProxySupplierLock"> -->
<TR>
<TD><CODE>-CECProxySupplierLock</CODE>
<EM>lock_type</EM>
</TD>
<TD>Select the lock type (<EM>null</EM>, <EM>thread</EM> or
<EM>recursive</EM>) to synchronize access to the
ProxyPushSupplier state.
</TD>
</TR>
<!-- <TR NAME="CECUseORBId"> -->
<TR>
<TD><CODE>-CECUseORBId</CODE>
<EM>orbid</EM>
</TD>
<TD>Set the name of the ORB used by the event service, only
useful in applications that create multiple ORBs and
activate the event service in one of them.
</TD>
</TR>
<!-- <TR NAME="CECConsumerControl"> -->
<TR>
<TD><CODE>-CECConsumerControl</CODE>
<EM>policy</EM>
</TD>
<TD>Select the consumer control policy (<EM>null</EM> or
<EM>reactive</EM>) to detect and discard broken consumers.
</TD>
</TR>
<!-- <TR NAME="CECSupplierControl"> -->
<TR>
<TD><CODE>-CECSupplierControl</CODE>
<EM>policy</EM>
</TD>
<TD>Select the supplier control policy (<EM>null</EM> or
<EM>reactive</EM>) to detect and discard broken suppliers.
</TD>
</TR>
<!-- <TR NAME="CECConsumerControlPeriod"> -->
<TR>
<TD><CODE>-CECConsumerControlPeriod</CODE>
<EM>period</EM>
</TD>
<TD>Set the period (in microseconds) used by the reactive
consumer control policy to poll the state of the
consumers.
</TD>
</TR>
<!-- <TR NAME="CECSupplierControlPeriod"> -->
<TR>
<TD><CODE>-CECSupplierControlPeriod</CODE>
<EM>period</EM>
</TD>
<TD>Set the period (in microseconds) used by the reactive
supplier control policy to poll the state of the
suppliers.
</TD>
</TR>
<!-- <TR NAME="CECReactivePullingPeriod"> -->
<TR>
<TD><CODE>-CECReactivePullingPeriod</CODE>
<EM>period</EM>
</TD>
<TD>Set the period (in microseconds) used by the reactive
pulling strategy to poll all the PullSuppliers for an
event.
</TD>
</TR>
<!-- <TR NAME="CECProxyConsumerCollection"> -->
<TR>
<TD><CODE>-CECProxyConsumerCollection</CODE>
<EM>flag[:flags]</EM>
</TD>
<TD><P>Configure the data structure and strategies used to
implement collections of
<CODE>ProxyPushConsumers</CODE> and
<CODE>ProxyPullConsumers</CODE>.
The argument is a colon separated list of flags, with
the following semantics:
<TABLE>
<TR><TH>Flag</TH><TH>Description</TH>
</TR>
<TR>
<TD>MT</TD>
<TD>Use regular mutexes and/or condition
variables for serialization.
</TD>
<TD>ST</TD>
<TD>Use null mutexes and/or condition
variables for serialization.
</TD>
<TD>LIST</TD>
<TD>Implement the collection using an ordered list,
fast for iteration (i.e. during event dispatching),
but slow for insertion and removal (i.e. when
clients connect and disconnect from the EC).
</TD>
<TD>RB_TREE</TD>
<TD>Implement the collection using a Red-Black tree,
slow for iteration (i.e. during event dispatching),
but fast for insertion and removal (i.e. when
clients connect and disconnect from the EC).
</TD>
<TD>IMMEDIATE</TD>
<TD>Threads block until they can execute a change on
the data structure, the system must use other
approaches to guarantee that the iterators are not
invalidated during event dispatching. For example,
use a separate dispatching thread.
</TD>
<TD>COPY_ON_READ</TD>
<TD>Before initiating an iteration to dispatch events
(or similar tasks) a copy of the complete collection
is performed.
This solves most of the synchronization problems,
but introduces a significant source of overhead and
priority inversions on the critical path.
</TD>
<TD>COPY_ON_WRITE <STRONG>UNIMPLEMENTED</STRONG></TD>
<TD>Similar to the previous one, but the copy is only
performed when needed.
</TD>
<TD>DELAYED</TD>
<TD>Threads that need to change the collection can
detect if that change will invalidate iterators used
by other threads.
If so, the thread posts the change on a queue that
is executed once the collection is no longer in
use.
</TD>
</TR>
</TABLE>
</P>
</TD>
</TR>
<!-- <TR NAME="CECProxySupplierCollection"> -->
<TR>
<TD><CODE>-CECProxySupplierCollection</CODE>
<EM>flag[:flags]</EM>
</TD>
<TD><P>Configure the data structure and strategies used to
implement collections of <CODE>ProxyPushSupplier</CODE>
and <CODE>ProxyPullSupplier</CODE> objects.
Use the same arguments as with the
<CODE>CECProxyConsumerCollection</CODE> option.
</P>
</TD>
</TR>
</TABLE>
</P>
<H3>The constructor</H3>
<P>The <CODE>TAO_CEC_EventChannel</CODE> class implements the
<CODE>CosEventChannelAdmin::EventChannel</CODE> interface;
this class takes one mandatory and two optional parameters in
its constructor:
</P>
<PRE>
TAO_CEC_EventChannel (const TAO_CEC_EventChannel_Attributes& attributes,
TAO_CEC_Factory* factory = 0,
int own_factory = 0);
</PRE>
<P>The <CODE>factory</CODE> is an optional parameter to override
the default strategy factory used by the event channel,
the event channel will destroy the factory if the
<CODE>own_factory</CODE> argument is true.
</P>
<P>The <CODE>attributes</CODE> parameter can be used to fine tune
some of the algorithms and strategies used by the event channel,
the default values are probably OK for most applications.
Notice that the attributes include the POA used to activate the
ConsumerAdmin, SupplierAdmin, ProxyPushConsumer,
ProxyPushSupplier, ProxyPullConsumer and the ProxyPullSupplier
objects;
this POAs must have the <CODE>IMPLICIT_ACTIVATION</CODE> and the
<CODE>SYSTEM_ID</CODE> policies (as the RootPOA does).
</P>
<P><TABLE BORDER="2" CELLSPACING="2" CELLPADDING="0">
<TR><TH>Attribute</TH><TH>Description</TH></TR>
<TR><TD><CODE>consumer_reconnect</CODE></TD>
<TD>If the attribute is not zero then the same consumer can
call <CODE>connect_push_consumer</CODE> on its
ProxyPushSupplier multiple times to change its
subscriptions;
this is usually more efficient that disconnecting and
connecting again.
</TD>
</TR>
<TR><TD><CODE>supplier_reconnect</CODE></TD>
<TD>If the attribute is not zero then the same supplier can
call <CODE>connect_push_supplier</CODE> on its
ProxyPushConsumer multiple times to change its
publications;
this is usually more efficient that disconnecting and
connecting again.
</TD>
</TR>
<TR><TD><CODE>disconnect_callbacks</CODE></TD>
<TD>It not zero the event channel will send disconnect
callbacks when a disconnect method is called on a Proxy.
In other words, if a consumer calls
<CODE>disconnect_push_supplier()</CODE> on its proxy the
EC will invoke <CODE>disconnect_push_consumer()</CODE>
on the consumer. A similar thing is done for suppliers.
It is a matter of debate what the spec requires for the
regular event service.
</TD>
</TR>
<TR><TD><CODE>busy_hwm</CODE></TD>
<TD>When the delayed flag is set on proxy collections,
this flag controls the maximum
number of threads that can simultaneously iterate over the
collection before blocking.
It can be used to avoid starvation in delayed updates on
the collection.
</TD>
</TR>
<TR><TD><CODE>max_write_delay</CODE></TD>
<TD>When the delayed flag is set on proxy collections,
this flag controls the maximum
number of threads that will initiate dispatching
<B>after</B> a change has been posted.
Any thread after that is blocked until the operations are
performed.
It can be used to completely stop starvation of delayed
updates on the collection.
</TD>
</TR>
</TABLE>
</P>
<P><HR><P>
Back to the TAO <A HREF="index.html">components documentation</A>.
<!--#include virtual="/~schmidt/cgi-sig.html" -->
<address><a href="mailto:coryan@cs.wustl.edu">Carlos O'Ryan</a></address>
<!-- Created: Thu Jul 1 21:44:28 CDT 1999 -->
<!-- hhmts start -->
Last modified: Thu Mar 23 19:10:16 PST 2000
<!-- hhmts end -->
</body>
</html>
|