I’m load-testing Wowza 3.0.5 on EC2 m1.small instances.
The configuration is:
-
Ubuntu 10.10 32-bit, Java -server 1.7u3
-
standard liverepeater/edge (origin is an m1.large, performance is not a problem on this machine at the moment)
-
live 128kbps mp3 streams
-
no custom modules
-
Wowza tuning guide followed (VHost.xml below)
I’m using rtmpdump on a third server to slowly build up outgoing connections to the edge. CPU usage on the edge server increases steadily (faster than I would expect to be honest) but increases after about 300 connections. I cannot get past 400 connections without CPU reaching 100% and load average sky-rocketing.
I would expect an m1.small to handle ~1000 outgoing 128kbps streams, taking the bandwidth of the incoming mediacaster streams into account.
How can I address this?
[HTML]
2
*
1935,80,443,554
true
24000
65000
true
100
cupertinostreaming,smoothstreaming,sanjosestreaming,dvrchunkstreaming
com.wowza.wms.http.HTTPCrossdomain
*crossdomain.xml
none
com.wowza.wms.http.HTTPClientAccessPolicy
*clientaccesspolicy.xml
none
com.app.wms.http.LiveServer
app_ctrl*
admin-basic
com.app.wms.http.Stats
stat*
none
1
*
8086
true
16000
16000
true
100
com.wowza.wms.http.streammanager.HTTPStreamManager
streammanager*
admin-digest
com.wowza.wms.http.HTTPServerInfoXML
serverinfo*
admin-digest
com.wowza.wms.http.HTTPConnectionInfo
connectioninfo*
admin-digest
com.wowza.wms.http.HTTPConnectionCountsXML
connectioncounts*
admin-digest
com.wowza.wms.transcoder.httpprovider.HTTPTranscoderThumbnail
transcoderthumbnail*
admin-digest
com.wowza.wms.http.HTTPProviderMediaList
medialist*
admin-digest
com.wowza.wms.http.HTTPServerVersion
*
none
smoothstreaming
smoothstreaming
cupertinostreaming
cupertinostreaming
sanjosestreaming
sanjosestreaming
dvrchunkstreaming
dvrchunkstreaming
60
40
2
100
5
2
250
true
65000
24000
true
100
2000
90000
500
75
true
1024000
65000
50
4096
true
65000
65000
50
4096
1
2
1
1
60000
12000
20000
0
60000
true
[/HTML]