News for Week 48
If you're new here and like what you read, you may want to subscribe to my RSS blog feed or free email updates. Thanks for visiting!
Once again, it is time for a short newscast. Notable this time is the fact, that more and more stuff seems to be written about
multi-cores, threading and parallelism in general. It feels like people do start realizing that the concurrent era is here and that they need to improve their knowledge in this field. A very encouraging sign! Or maybe its just because of Supercomputing and the recent coverage will die down again. Anyways, without further ado, here are some of the articles I found interesting lately:
- An article about GPGPU-programming - an overview of the current state of affairs, for the people who already know the idea the most interesting piece in there is probably the claimed speedups.
- James Gosling tells his views about why MPI is not such a great idea on multi-cores - and I could not agree more…
- David Dice compares parallelism to drugs in his parallelism manifesto - and wishes for a system where parallelism is described without using threads - a system, where the runtime system maps this parallelism onto threads as needed. I wonder what he thinks about Cilk?
- Closely related to the last point, I wonder how he will like the tasking-system that will be present in OpenMP 3.0? A short overview of the features to come are in this presentation, and tasking is definitely the big thing (TM) for this version of the spec.
- An article for the Java-folks out there: Billy Newport worries about Java’s and Java EE’s performance in the multi-core era. I am not sure I agree with his conclusions, though - if you look into the comments for this article, many valid objections are raised.
- My last item is neither new nor has it anything to do with parallelism, but since I imagine this blog is visited frequently by real programmers, they might be interested in the article describing themselves. My favorite quote from this article:
Besides, the determined Real Programmer can write Fortran programs in any language.
There is a follow up article that shows an even more real programmer. Very entertaining!
Hope you found something interesting among them!
