An On-line Course in Parallel Programming
OVERVIEW:
“Why is that this e-book completely different from all different parallel programming books?”
- Appropriate for both college students or professionals.
- Sensible viewpoint:
- There may be little or no theoretical evaluation of parallel
algorithms, akin to O() evaluation, most theoretical speedup,
acyclic graphs and so forth. - Intensive protection of “wizardry” facets, i.e. materials recognized
to skilled practitioners however typically not in books, akin to
protection of loop iteration scheduling, reminiscence results of storing
giant arrays and so forth. - Appendices cowl methods background, essential in utilized work
however all the time simply “assumed” to be data possessed by the readers. - Appreciable consideration is paid to strategies for debugging.
- There may be little or no theoretical evaluation of parallel
- Makes use of the primary parallel platforms—OpenMP, CUDA and
MPI—rather than languages that at this stage are largely
experimental, such because the elegant-but-not-yet-mainstream Cilk. - Begins with actual parallel code straight away in Chapter 1, with
examples from pthreads, OpenMP and MPI.
Consistently evolving:
Like all my open supply textbooks, this one is consistently
evolving. I proceed so as to add new subjects, new examples, extra timing
analyses, and so forth, and naturally repair bugs and enhance the exposition.
Stipulations: The coed should be easonably adept in
programming, and have math background by means of linear algebra. (An
appendix to the e-book evaluations the components of the latter wanted for this
e-book.)
LICENSING:
This
work is licensed beneath a Creative
Commons Attribution-No Derivative Works 3.0 United States License.
Copyright is retained by N. Matloff in all non-U.S. jurisdictions,
however permission to make use of these supplies in educating continues to be granted,
supplied the authorship and licensing info right here is displayed.
I’d admire being notified in case you use this e-book for educating,
simply in order that I do know the supplies are being put to make use of, however that is
not required.
DOWNLOAD THE BOOK:
The e-book is freely accessible, topic to the situations above, and may
be downloaded from
http://heather.cs.ucdavis.edu/~matloff/158/PLN/ParProcBook.pdf.
AUTHOR’S BIO:
Norm Matloff
is a Professor of Laptop Science on the College of
California, Davis. He was previously a statistics professor at that
college. His analysis has included numerous numerous areas,
and he has been a recipient of the college’s Distinguished
Educating Award. He was born and raised in Los Angeles, and
earned his PhD in theoretical arithmetic (chance
concept/purposeful evaluation) at UCLA.