NDT-DEV email list created

Revision: 335
Author: dan.blah
Date: Tue Mar 30 18:47:11 2010
Log: This is a top high level overview of the NDT test methodology by Rich  
Carlson.
http://code.google.com/p/ndt/source/detail?r=335

Added:
 /wiki/NDTTestMethodology.wiki

=======================================
--- /dev/null
+++ /wiki/NDTTestMethodology.wiki       Tue Mar 30 18:47:11 2010
@@ -0,0 +1,31 @@
+= Download Test =
+ * A block of pseudo random data is generated on the server and stored in  
memory.
+ * The server repeatedly transmits this data to the Java client for 10  
seconds.
+ * The client acknowledges receipt of this data and then discards it.
+ * The server calculates the average throughput by dividing the number of  
bytes sent by the test time.
+ * The server also makes measurements on a per-packet basis, calculating  
the capacity of the e2e path.
+ * The server measures the number of lost and delayed packets, the  
instantaneous round trip time, and examines the clients network  
configuration to determine what, if anything, caused a drop in the expected  
throughput.
+ * Multiple clients may be served simultaneously.
+
+= Upload Test =
+ * A block of pseudo random data is generated on the client and stored in  
memory.
+ * The client repeatedly transmits this data to the server for 10 seconds.
+ * The server acknowledges receipt of this data and then discards it.
+ * The server calculates the average throughput by dividing the number of  
bytes sent by the test time.
+ * The server also makes measurements on a per-packet basis, calculating  
the capacity of the e2e path.
+ * Results from the test are returned to the client at the conclusion of  
the test.
+
+= Latency/Jitter =
+ * Multiple measurements of the round trip delay and per-packet  
transmission/arrival times are collected.
+ * The delay for each measurement is summed and divided by the total  
number of measurements to generate the average round trip delay.
+
+= Advanced diagnostics =
+ * The server determines what, if anything, is limiting the maximum  
download speed
+ - Possible limits might be the clients/servers configuration, the  
clients/servers CPU utilization, or the network infrastructure
+ * The server reports if a firewall is blocking connections in one or both  
directions
+ * The client determines if a NAT box is being used
+ * The server reports abnormal conditions such as high packet loss or  
packets arriving out of order
+
+
+------------------------
+What the above says is, the NDT is a typical memory to memory  
client/server test device.  Throughput measurements closely measure the  
network performance, and ignore the disk I/O effects.  The real strength is  
in the advanced diagnostic features that are enabled by the kernel data  
automatically collected by the web100 monitoring infrastructure.  This data  
is collected during the test (at 5 msec increments) and analyzed after the  
test completes to determine what, if anything, impacted the test.  One of  
the MAJOR issues facing a commodity Internet users is the performance  
limiting host configuration settings for the Windows XP operating system.   
To illustrate this, a cable modem user with basic service (15 Mbps  
download) would MAX out at 13 Mbps with a 40 msec RTT delay.  Thus unless  
the ISP proxies content, the majority of traffic will be limited by the  
clients configuration and NOT the ISP's infrastructure.  The NDT server can  
detect and report this problem, saving consumers and ISP's dollars by  
allowing them to quickly identify where to start looking for a problem.   
The FCC really needs to understand this message, or we will not be as  
effective as we need to be.