The Concurrency Runtime: Fine Grained Parallelism for C++
- Posted: Oct 08, 2008 at 12:27 PM
- 59,800 Views
- 5 Comments
We've spent a fair amount of time on Channel 9 discussing
concurrency and parallelism with various people. In particular, the folks who are writing the Parallel Computing Platform. Everything we've talked about up to this point has been targeted at .NET developers. After watching all the videos on TPL, for example,
you'd think that only .NET developers will get to benefit from Task-based concurrent programming. Of course, this is not the case. Native developers (C++) will be given an equivalent level of concurrency abstraction (and in some cases more) via the native
Concurrency Runtime and associated tools.
Here, native (C++ focused) Parallel Computing Platform team members Rick Malloy, Niklas Gustafsson, Don McCrady and Channel 9 veteran Stephen Toub join me in a conversation covering our concurrency platform tools and runtime specifically designed for C++ developers.
Enjoy!
Here, native (C++ focused) Parallel Computing Platform team members Rick Malloy, Niklas Gustafsson, Don McCrady and Channel 9 veteran Stephen Toub join me in a conversation covering our concurrency platform tools and runtime specifically designed for C++ developers.
Enjoy!
Comments Closed
Comments have been closed since this content was published more than 30 days ago, but if you'd like to continue the conversation,
please create a new thread in our Forums,
or
Contact Us and let us know.
More posts in this blog
-
Jeffrey Snover, Martin Fowler and Neil Ford:…
-
The Concurrency Runtime: Fine Grained Parallelism…
-
Parallel Computing Platform: An Integrated…
Follow the Discussion
Oops, something didn't work.
What does this mean?
Following an item on Channel 9 allows you to watch for new content and comments that you are interested in. You need to be signed in to Channel 9 to use this feature.What does this mean?
Following an item on Channel 9 allows you to watch for new content and comments that you are interested in and view them all on your notifications page.sign up for email notifications?
I also see almost no difference in the objective between managed and unmanaged worlds, notwithstanding the fact that native code is "closer to the metal" than managed code. I'm pretty sure that an algorithm that works well concurrently in C++, will also work well in C# and vica-versa (most of the time).
Is TCR: FGP "equal" on all many-core CPUs, GPUs, CPU-GPUs?
Is there a suite of performance tests? Same app/code on different chips? (Or must we wait for marketing pr wars?)
Do you have any hardware guys on your team or do you depend on industry contacts for down to the metal knowledge to inform your architecture, TCR: FGP?
Hi Art-
Glad you enjoyed the video!
1) For the initial release, the Concurrency Runtime is targeting CPUs.
2) We internally have a suite of performance tests, though I don't know at this time whether such a suite would be released publicly. There are of course industry standard suites, many of which have emerged from the HPC space, and I imagine at some point some subset of those will be publicly ported to run on the runtime once it's released.
3) To your last question, we have team members that focus on hardware, on software, and the interplay between the two.
Thanks,
Stephen
How can I dowload the video? I have trouble watching it because it keeps stopping every 5 seconds. I have a 3 mbs connection.
Choose a format from the Format dropdown under the player.
C
Remove this comment
Remove this thread
close