[PLUG] O.T.VoIP and Satellite

[Chuck Hast]

Here is another I just pulled, it is bad, the downlink has higher jitter
too, so it must be loading. Also note the speeds have dropped.



Speed test statistics
---------------------
Download speed: 2655 kbps
Upload speed: 743 kbps
Download consistency of service: 19 %
Upload consistency of service: -- %
Download test type: socket
Upload test type: POST
Maximum TCP delay: 990 ms
Average download pause: 12 ms
Minimum round trip time to server: 1 ms
Average round trip time to server: 1 ms
Estimated download bandwidth: 11177 kbps
Route concurrency: 4.2097044
Download TCP forced idle: 0 %
Maximum route speed: 524280 kbps

VoIP test statistics
--------------------
Jitter: you --> server: 173.5 ms
Jitter: server --> you: 23.0 ms
Packet loss: you --> server: 0.0 %
Packet loss: server --> you: 0.0 %
Packet discards: 74.2 %
Packets out of order: 0.0 %
Estimated MOS score: 1.1


General information
-------------------
IP address: 72.168.141.72
Local time: Dec 30, 2014 6:16:44 PM
Test server: http://voiptest.8x8.com:82/


On Tue, Dec 30, 2014 at 2:56 PM, Mike C. <mconnors1 at gmail.com> wrote:

> >
> > Here is another test of the link the jitter has dropped down quite a
> > bit still not good enough for VoIP, but better.
> >
> >
> > Speed test statistics
> > ---------------------
> > Download speed: 3864 kbps
> > Upload speed: 894 kbps
> > Download consistency of service: 55 %
> > Upload consistency of service: -- %
> > Download test type: socket
> > Upload test type: POST
> > Maximum TCP delay: 933 ms
> > Average download pause: 7 ms
> > Minimum round trip time to server: 1 ms
> > Average round trip time to server: 1 ms
> > Estimated download bandwidth: 13010 kbps
> > Route concurrency: 3.3666103
> > Download TCP forced idle: 0 %
> > Maximum route speed: 524280 kbps
> >
> > VoIP test statistics
> > --------------------
> > Jitter: you --> server: 166.8 ms
> > Jitter: server --> you: 3.1 ms
> > Packet loss: you --> server: 0.0 %
> > Packet loss: server --> you: 0.0 %
> > Packet discards: 73.0 %
> > Packets out of order: 0.0 %
> > Estimated MOS score: 1.1
> >
> >
> > General information
> > -------------------
> > IP address: 72.168.141.72
> > Local time: Dec 30, 2014 11:19:37 AM
> > Test server: http://voiptest.8x8.com:82/
> >
> > Here it is with a 729 codec simulation:
> >
> >
> > Speed test statistics
> > ---------------------
> > Download speed: 3557 kbps
> > Upload speed: 896 kbps
> > Download consistency of service: 21 %
> > Upload consistency of service: -- %
> > Download test type: socket
> > Upload test type: POST
> > Maximum TCP delay: 1242 ms
> > Average download pause: 8 ms
> > Minimum round trip time to server: 1 ms
> > Average round trip time to server: 1 ms
> > Estimated download bandwidth: 4777 kbps
> > Route concurrency: 1.3429916
> > Download TCP forced idle: 0 %
> > Maximum route speed: 524280 kbps
> >
> > VoIP test statistics
> > --------------------
> > Jitter: you --> server: 106.9 ms
> > Jitter: server --> you: 2.9 ms
> > Packet loss: you --> server: 0.0 %
> > Packet loss: server --> you: 0.0 %
> > Packet discards: 65.8 %
> > Packets out of order: 0.0 %
> > Estimated MOS score: 1.4
> >
> >
> > General information
> > -------------------
> >
> > Note that the Jitter has dropped, not sure if that is going from 711 to
> 729
> > but I will do more testing.
>
>
> This is good information and a good start. Yes, the jitter did drop and I'd
> expect it to for g.729 since it requires less bandwidth. Even though at 90
> Kbps, it's using approx. 10% of the available bandwidth.
>
> The bigger concern for me though is "packet discards = 65.8%" The VOIP
> server will discard packets it receives too late and that it can't use.
> This indicates the delay is too high between you and the VOIP server.
>
> It's too bad that only max and not avg tcp delay is provided. As it is, max
> tcp delay of 1242 ms is way out of tolerance for VOIP.
>
> You could provide this data to HughesNet and see what they can do. Maybe
> there is some way they can clean up the sat-link?
>
> I'd be curious to see test results from say midnight or after. If there's
> congestion on the Sat uplink side, that could delay the packets enough for
> the VOIP server to discard them.
> _______________________________________________
> PLUG mailing list
> PLUG at lists.pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
>



-- 

Chuck Hast  -- KP4DJT --
Glass, five thousand years of history and getting better.
The only container material that the USDA gives blanket approval on.