Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[ammoman]

Hi all,
I tested my link with a tester from Glasnost and it gave me the following result -

Our measurements found the following characteristics for your link (hostXX-XXX-XX-27.rangeXX-151.btcentralplus.com - XX.XXX.XX.27)
 Downstream Bandwidth: 1496 Kbps    Downstream Queue: 1025 ms
 Upstream Bandwidth: 397 Kbps    Upstream Queue: 90 ms

Notes:
 Your downstream queue is larger than 300ms and thus larger than average round-trip times in the Internet (e.g., 50ms for US coast-to-coast delays and 150ms for transatlantic delays) which is the recommended size for queues.

I assume that the large downstream queue means that I have a lower download speed than my line is capable of and if so, is there anything I can do to change it.
I'm using a Homehub 1 on an up to 8mb connection and here are my stats -
DSL Connection
Link Information

Uptime:
0 days, 13:50:48

Modulation:
G.992.1 annex A

Bandwidth (Up/Down)
[kbps/kbps]:
448 / 1,760

Data Transferred
(Sent/Received) [MB/MB]:
14.16 / 134.79

Output Power (Up/Down)
[dBm]:
11.5 / 17.5

Line Attenuation
(Up/Down) [dB]:
27.5 / 53.0

SN Margin (Up/Down)
[dB]:
20.0 / 8.0

Vendor ID (Local/Remote):
TMMB / TSTC

Loss of Framing
(Local/Remote):
0 / 0

Loss of Signal
(Local/Remote):
5 / 0

Loss of Power
(Local/Remote):
0 / 0

Loss of Link (Remote):
0

Error Seconds
(Local/Remote):
510 / 0

FEC Errors (Up/Down):
0 / 4,941,739

CRC Errors (Up/Down):
0 / 1,044

HEC Errors (Up/Down):
0 / 702

Line Profile:
Interleaved

Any help appreciated.

[roseway]

I'm afraid I don't know what the downstream queue is, but I just did a Glasnost test myself and it wasn't even able to complete the link test, although the P2P tests all passed.

[ammoman]

Cheers Eric,
Hopefully one of our gurus will know!

[roseway]

I'm getting interested in this now, and I found a different route to the link test (probably the one which you used http://loki10.mpi-sws.mpg.de/bb/bb.php ), with this result:

 Our measurements found the following characteristics for your link (port-252.xxx.th.newnet.co.uk - 80.175.68.252)

*    Downstream Bandwidth:    657 Kbps         *    Downstream Queue:    308 ms
*    Upstream Bandwidth:    397 Kbps         *    Upstream Queue:    1705 ms
Notes:

* Your downstream queue is larger than 300ms and thus larger than average round-trip times in the Internet (e.g., 50ms for US coast-to-coast delays and 150ms for transatlantic delays) which is the recommended size for queues.

* Your upstream queue is so large that packets can get delayed for more than a second. This can be a problem for real-time traffic such as VoIP or online games!

Quite different from yours, but still terrible according to them. And the result isn't representative any way, because the measured downstream speed is much lower than it really is (about 3.9 Mbps currently with speedtest.net).

[general disquiet]

Ammoman,  did you use the UK server or one of the US ones?  The high queue time might reflect a US server, and a connection losing quite a few packets in transmission.  That might be a reflection of the high SNR margin on your line.  (Just checked and Glasnost has now been changed - it will be a US server it is linking to)

Try the 1st and 3rd test here http://www.measurementlab.net/measurement-lab-tools on the same page as Glasnost and see if they pick anything unusual up.  The first one is fairly straightforward, but you'll need to play around a little to get the third one to work properly.  I'd start off by setting your target speed to 1 Mbps and the other figure (latency) to 350 ms.  If that fails, it will suggest values you can try.

- these tests do let you pick which server you use.

Good luck