The second part of the TCP congestion control exercise provides a complete set of tools and infrastructure for conducting a series of experiments relative to TCP and UDP. In this part you will have the ability to run and, on-the-fly, manage various tests in a real remote testbed with reserved hosts and network resources. On this environment you will be asked to perform 6 tests that examine various aspects of the TCP congestion control. For each test you will have to complete a specific number of steps, collect the required information and create the appropriate report (e.g. a graphical representation of the results).
Please see part 1 of TCP Congestion Control course series
This page connects you to the iMinds Virtual wall. Check the side image on how you are connected to the Virtual Wall.
We are connected over IPv6 to the iMinds Virtual wall. You are not required to do so when accessing this page.
The Virtual Wall is hosted at and operated by iMinds iLab.t. The hardware which is visible through the Aggregate Manager consists of 110x dual CPU nodes (2x 6 cores), 24 GB RAM, 1x harddisk 250GB, Intel Xeon E5645 (2.40GHz), interconnected with a Force 10 C300 switch (multiple gigabit interfaces per node).
This part uses a simple TCP congestion control exercise to train you on how you can run tasks on testbeds with real hardware resources.
During this simple task a TCP client (node0) connects to a TCP server (node1) and transmits data for 30 seconds.
During this connection, in the client node, various TCP parameters are recorded in a file called “tcpout“
username: wall02guest password: wall
cd /local/forge ./install ./run_server
Just follow these steps to connect to the client machine:
username: wall01guest password: wall
cd /local/forge ./install ./monitor ./run_client
username: wall01guest password: wall logfiles: /local/forge/tcpout Machine= client1.UoPDemo.wall2-ilabt-iminds-be.wall1.ilabt.iminds.be