Table of contents
Remember that, as part of homework 6, each term project team is required to give a 5-minute proposal talk in class this Thurs. I'll let you choose the order.
On parellel.ecse, /parallel-class/openmp/rpi/cpuvs.wall.cc is a simple program that shows how wallclock time shrinks as the number of threads rises (up to a point) but CPU time rises from the start.
That is because one way that a thread waits to run is to burn cycles. For small delays, this is easier and has less overhead than explicitly suspending. The GNU version of OpenMP lets you choose the crossover point between buring cycles and syspending a thread.
As part of getting the Xeon Phi coprocessor working, I'll need to reboot parallel from time to time. That takes a few minutes and you're welcome to connect again when it comes up.
I have to do this when I'm physically at the machine.
I'll try to do this when it's not actively being used, and will try to send a message to connected people.