In this video episode of Teach Parallel, Susan Rivoire, Assistant Professor of Computer Science, Sonoma State University discusses her undergraduate survey course that covered a number of different programming models.
The course not only exposed to students to parallelism, it showed how they could directly affect performance through (relatively) simple parallel models and tools. One obstacle she faced was the lack of teaching resources, specifically her need for manycore platforms. You can learn more about the course through the paper “A Breadth-first Course in Multicore and Manycore Programming” she delivered at SIGCSE 2010. Both the Intel Academic Community’s Manycore Testing Lab for education and the EAPF’s LittleFe Hardware Grants can help provide these kind of resources. The Intel microgrant awards for parallel content can also help provide funds for the creation of content for teaching parallelism.