ndt-users list created

[Jeremy Palmer <>]

The throughput numbers could be correct. I see similar results when testing 
from
hosts on our network at one site to our NDT server at a different site (10/100
local LANs at both sites connected by OC-48 backbone).
If your running the test from a Windows host, try installing a real-time
bandwidth meter app on this machine to verify the throughput.
The two I'm familiar with are DUMeter (not free but time limited demo 
available,
I use this on all my Windows machines)
http://www.dumeter.com/
MyDUMeter (open source DUMeter clone)
http://www.codeproject.com/csharp/mydumeter.asp
Also, a coworker pointed out they use NetMeter (I've never used this one 
myself)
http://www.metal-machine.de/readerror/

-- 
Jeremy Palmer
Network Engineering
ViaWest, Inc.
Office: 720.891.1045
Fax: 303-874-5236
www.viawest.net

Jeremy Schafer wrote:
> Is it possible that I have bogus information or is it correct? Here is
> what the Java applet spit out as information when I did a scan:
> 
> TCP/Web100 Network Diagnostic Tool v5.5.1
> click START to begin
> Connected to 172.28.7.136 -- Using IPv4 address
> Checking for Middleboxes ..... Done
> checking for firewalls ...... Done
> running 10s outbound test (client-to-server [C2S]) .... 93.98Mb/s
> running 10s inbound test (server-to-client [S2C]) ..... 94.09Mb/s
> The slowest link in the end-to-end path is a 1.0 Gbps Gigabit Ethernet
> subnet.
> 
> Both the throughput numbers and the "slowest link" output are
> interesting. The NDT client and server are on different networks but
> they are both connected to their local switch at 100Mb/s full-duplex.
> Are these numbers still expected? They seem awful high for the speed
> they should be running at.
> 
> Also, is it possible to change the test from 10s to 30s? If so, do I
> cause any problems by doing it? I appreciate the help. Thanks.
>