Currently, I am teaching a graduate class on data stream systems at Umaine. After motivation, data models and query language constructs, we now enter the ‘hands-on’ phase in the class: programming with data streams systems.
As a system, we chose Stanford STREAM system since it has versions for Windows/Linux and Mac, and runs flawlessly on Macs. It has a nice graphical user interface to see query trees, and is open source code, so we are planning to play around with adaptive processing later in the class.
There were some struggles to get it to run on Windows/Ubuntu, but we got it to compile and run eventually.