Multi-user video conference
Hello, I have following requirements for multi-user video conference: 1. One Teacher and up to 10 Students are in the same room. 2. Teacher hears everyone in the room and sees video streams from every student (total bandwidth = 1 audios stream + n video streams). 3. Student hears everyone in the room and sees video stream from Teacher only (total bandwidth = 1 audio stream + 1 video stream). 4. There is no CPU intensive video encoding process on the server. Teacher receives video streams as-is plus muxed audio stream. 5. Conference is auto-recorded on the server. (I don't know exactly how, mp4 with multiple video streams?) Is it possible to implement this with FreeSwitch? If no what requirements should be relaxed? -- Best regards, I. Sokolov
What you are describing is a video sfu, which we are not. We would accomplish the same while using less bandwidth but more cpu. On Saturday, May 28, 2016, happy.neko <happy.neko@gmail.com> wrote:
Hello,
I have following requirements for multi-user video conference:
1. One Teacher and up to 10 Students are in the same room. 2. Teacher hears everyone in the room and sees video streams from every student (total bandwidth = 1 audios stream + n video streams). 3. Student hears everyone in the room and sees video stream from Teacher only (total bandwidth = 1 audio stream + 1 video stream). 4. There is no CPU intensive video encoding process on the server. Teacher receives video streams as-is plus muxed audio stream. 5. Conference is auto-recorded on the server. (I don't know exactly how, mp4 with multiple video streams?)
Is it possible to implement this with FreeSwitch? If no what requirements should be relaxed?
-- Best regards, I. Sokolov
I know this question is impossible to answer exactly but what roughly estimated CPU usage could I expect for conference with five 480p streams transcoded at 15 fps. One room with common core i5 quad CPU, 5 simultaneous rooms, 10 rooms? Also I am a bit puzzled about differences between "mux" and "transcode" video modes. "Mux" is required to place different sources on canvas (compose completely new video stream) while "transcode" just reencodes video source with different codec, am I right? 2016-05-29 22:45 GMT+03:00 Michael Jerris <mike@jerris.com>:
What you are describing is a video sfu, which we are not. We would accomplish the same while using less bandwidth but more cpu.
On Saturday, May 28, 2016, happy.neko <happy.neko@gmail.com> wrote:
Hello,
I have following requirements for multi-user video conference:
1. One Teacher and up to 10 Students are in the same room. 2. Teacher hears everyone in the room and sees video streams from every student (total bandwidth = 1 audios stream + n video streams). 3. Student hears everyone in the room and sees video stream from Teacher only (total bandwidth = 1 audio stream + 1 video stream). 4. There is no CPU intensive video encoding process on the server. Teacher receives video streams as-is plus muxed audio stream. 5. Conference is auto-recorded on the server. (I don't know exactly how, mp4 with multiple video streams?)
Is it possible to implement this with FreeSwitch? If no what requirements should be relaxed?
-- Best regards, I. Sokolov
_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com
Official FreeSWITCH Sites http://www.freeswitch.org http://confluence.freeswitch.org http://www.cluecon.com
FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
On May 29, 2016, at 5:33 PM, happy.neko <happy.neko@gmail.com> wrote:
I know this question is impossible to answer exactly but what roughly estimated CPU usage could I expect for conference with five 480p streams transcoded at 15 fps. One room with common core i5 quad CPU, 5 simultaneous rooms, 10 rooms?
Correct, it is indeed. Try it out and see.
Also I am a bit puzzled about differences between "mux" and "transcode" video modes. "Mux" is required to place different sources on canvas (compose completely new video stream) while "transcode" just reencodes video source with different codec, am I right?
Transcode is similar to 1x1 layout in mux mode, uses a basic video follows audio approach. It does indeed re-encode the stream.
2016-05-29 22:45 GMT+03:00 Michael Jerris <mike@jerris.com <mailto:mike@jerris.com>>: What you are describing is a video sfu, which we are not. We would accomplish the same while using less bandwidth but more cpu.
On Saturday, May 28, 2016, happy.neko <happy.neko@gmail.com <mailto:happy.neko@gmail.com>> wrote: Hello,
I have following requirements for multi-user video conference:
1. One Teacher and up to 10 Students are in the same room. 2. Teacher hears everyone in the room and sees video streams from every student (total bandwidth = 1 audios stream + n video streams). 3. Student hears everyone in the room and sees video stream from Teacher only (total bandwidth = 1 audio stream + 1 video stream). 4. There is no CPU intensive video encoding process on the server. Teacher receives video streams as-is plus muxed audio stream. 5. Conference is auto-recorded on the server. (I don't know exactly how, mp4 with multiple video streams?)
Is it possible to implement this with FreeSwitch? If no what requirements should be relaxed?
-- Best regards, I. Sokolov
_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org <mailto:consulting@freeswitch.org> http://www.freeswitchsolutions.com <http://www.freeswitchsolutions.com/>
Official FreeSWITCH Sites http://www.freeswitch.org <http://www.freeswitch.org/> http://confluence.freeswitch.org <http://confluence.freeswitch.org/> http://www.cluecon.com <http://www.cluecon.com/>
FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org <mailto:FreeSWITCH-users@lists.freeswitch.org> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users <http://lists.freeswitch.org/mailman/listinfo/freeswitch-users> UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users <http://lists.freeswitch.org/mailman/options/freeswitch-users> http://www.freeswitch.org <http://www.freeswitch.org/> _________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com
Official FreeSWITCH Sites http://www.freeswitch.org http://confluence.freeswitch.org http://www.cluecon.com
FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
2016-05-31 18:03 GMT+03:00 Michael Jerris <mike@jerris.com>:
On May 29, 2016, at 5:33 PM, happy.neko <happy.neko@gmail.com> wrote:
I know this question is impossible to answer exactly but what roughly estimated CPU usage could I expect for conference with five 480p streams transcoded at 15 fps. One room with common core i5 quad CPU, 5 simultaneous rooms, 10 rooms?
Correct, it is indeed. Try it out and see.
Will SIPp do the task or your can recommend another (maybe simpler?) benchmark tool?
We do not provide assistance with load testing as it is too much of a drain on resources to help people do it properly. By far the most accurate way is with real traffic.
On May 31, 2016, at 11:46 AM, happy.neko <happy.neko@gmail.com> wrote:
2016-05-31 18:03 GMT+03:00 Michael Jerris <mike@jerris.com <mailto:mike@jerris.com>>:
On May 29, 2016, at 5:33 PM, happy.neko <happy.neko@gmail.com <mailto:happy.neko@gmail.com>> wrote:
I know this question is impossible to answer exactly but what roughly estimated CPU usage could I expect for conference with five 480p streams transcoded at 15 fps. One room with common core i5 quad CPU, 5 simultaneous rooms, 10 rooms?
Correct, it is indeed. Try it out and see.
Will SIPp do the task or your can recommend another (maybe simpler?) benchmark tool?
Maybe this could help: Intel whitepaper on ffmpeg mosaic transcoding of multiple Full-HD streams at 30fps using a recent Xeon E3 CPU. https://www-ssl.intel.com/content/dam/www/public/us/en/documents/white-paper... François. https://www-ssl.intel.com/content/dam/www/public/us/en/documents/white-paper... On 2016-05-31 18:06, Michael Jerris wrote:
We do not provide assistance with load testing as it is too much of a drain on resources to help people do it properly. By far the most accurate way is with real traffic.
On May 31, 2016, at 11:46 AM, happy.neko <happy.neko@gmail.com <mailto:happy.neko@gmail.com>> wrote:
2016-05-31 18:03 GMT+03:00 Michael Jerris <mike@jerris.com <mailto:mike@jerris.com>>:
On May 29, 2016, at 5:33 PM, happy.neko <happy.neko@gmail.com <mailto:happy.neko@gmail.com>> wrote:
I know this question is impossible to answer exactly but what roughly estimated CPU usage could I expect for conference with five 480p streams transcoded at 15 fps. One room with common core i5 quad CPU, 5 simultaneous rooms, 10 rooms?
Correct, it is indeed. Try it out and see.
Will SIPp do the task or your can recommend another (maybe simpler?) benchmark tool?
_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com
Official FreeSWITCH Sites http://www.freeswitch.org http://confluence.freeswitch.org http://www.cluecon.com
FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
participants (3)
-
François
-
happy.neko
-
Michael Jerris