Andy Ayers: Understanding the Phoenix Compiler Framework
Description
Dan Reed is Microsoft's Director of Scalable/Multi-Core Systems Research and head of the
recently formed Universal Parallel Computing Research Centers (UPCRC): one at the University of California at Berkeley (UC-Berkeley) and a second at the University of Illinois
at Urbana-Champaign (UIUC). Since we've been focusing a bit recently on the Concurrency and Parallelism Software Revolution we figured Dan would be another great technical guru to talk to about Multi/Many-Core's impact
on the future of general purpose computing.
The angle of this conversation focuses attention primarily on the server-side parallelism problem which is distinct from the client problem (as addressed by Burton Smith
here) but part of the same wide-angle general purpose solution to the complex (and arguably fractal) general problem that spans microblips in DRAM to massive data centers.
Certainly the computation Cloud of the future must not only be scalable and highly performant, but also adaptive and homeostatic in how it reacts to frequent perturbation.
What are some of the challenges on the server side with respect to concurrent processing and massive scalability? Clustered server computing environments have traditionally been very good at parallel computation (compared to the general purpose client) so what's
Dan and Microsoft working on to ensure our Cloud scales to ManyCore? Is machine learning being incorporated into clustered computing software adaptation and evolution?
Dan has a very interesting biography:
I am also a member of the President's Council of Advisors on Science and Technology (PCAST) and chair of the Computing Research Association (CRA)"
Dan was the head of CS at Illinois during the birth of the web browser Mosaic which changed the way people interact with the Internet forever... We talk about where the web is today (including browsers) versus what Mosaic enabled when it arrived.
Enjoy. This is another great discussion with a supercomputing stalwart whose main focus these days is on ensuring we are prepared for the highly parallel future of general purpose computation in the sky.
Low res file here.
Download
Download this episode
- Low Quality MP4 (152.3 MB)
- High Quality MP4 (334.7 MB)
- Mid Quality MP4 (233.6 MB)
- Mid Quality WMV
- MP3
- WMV
More episodes in this series
Abolade Gbadegesin: Live Mesh Architecture
Related episodes
Expert to Expert: Erik Meijer and Bertrand Meyer - Objects, Contracts, Concurrency,…
MSR Cambridge Tour: Machine Learning Group, Computer Vision and F#
DryadLINQ
Expert to Expert: Inside Concurrent Basic (CB)
CHESS: An Automated Concurrency Testing Tool
Burton Smith: On General Purpose Super Computing and the History and Future of Parallelism
Anders Hejlsberg, Herb Sutter, Erik Meijer, Brian Beckman: Software Composability and the…
Don Syme: Introduction to F#, Part 2
Don Syme: Introduction to F#, Part 1
Programming in the Age of Concurrency: Software Transactional Memory
The Discussion
-
Minh awesome video!
-
Charles Minh wrote:awesome video!
Glad you enjoyed it! Thanks for watching.
C -
CPrest Great Video !
not just welcome to microsoft (for us non MS employees).
Welcome to channel 9 !
-
JoshRoss What is this dry-ah thing? Or at the very least, how do you spell it? -
Charles JoshRoss wrote:What is this dry-ah thing? Or at the very least, how do you spell it?
"Dryad is an infrastructure which allows a programmer to use the resources of a computer cluster or a data center for running data-parallel programs. A Dryad programmer can use thousands of machines, each of them with multiple processors or cores, without knowing anything about concurrent programming." -
staceyw Charles wrote:
JoshRoss wrote:
What is this dry-ah thing? Or at the very least, how do you spell it?
"Dryad is an infrastructure which allows a programmer to use the resources of a computer cluster or a data center for running data-parallel programs. A Dryad programmer can use thousands of machines, each of them with multiple processors or cores, without knowing anything about concurrent programming."
Andrew Birrell is involved in that project - it must be good. He also has an Automatic Mutual Exclusion (AME) project for alternative to explicit locks that looks interesting. -
Charles staceyw wrote:Charles wrote:
JoshRoss wrote:
What is this dry-ah thing? Or at the very least, how do you spell it?
"Dryad is an infrastructure which allows a programmer to use the resources of a computer cluster or a data center for running data-parallel programs. A Dryad programmer can use thousands of machines, each of them with multiple processors or cores, without knowing anything about concurrent programming."
Andrew Birrell is involved in that project - it must be good. He also has an Automatic Mutual Exclusion (AME) project for alternative to explicit locks that looks interesting.
I sense a Going Deep with Andrew in future...
C -
staceyw Charles wrote:staceyw wrote:
Charles wrote:
JoshRoss wrote:
What is this dry-ah thing? Or at the very least, how do you spell it?
"Dryad is an infrastructure which allows a programmer to use the resources of a computer cluster or a data center for running data-parallel programs. A Dryad programmer can use thousands of machines, each of them with multiple processors or cores, without knowing anything about concurrent programming."
Andrew Birrell is involved in that project - it must be good. He also has an Automatic Mutual Exclusion (AME) project for alternative to explicit locks that looks interesting.
I sense a Going Deep with Andrew in future...
C
That would be awesome Charles!!
I have been big fan ever sense his "Programming Threads in C#" paper. The first, and best, paper on managed threads and locks (granted I have not got Joe Duffy's book yet).
Hmmm. That gives me a crazy dream. How about a panel talk on Concurrency futures with Joe Duffy, Andrew Birrell, Don Box, Dan Reed, and George Chrysanthakopoulos. Don tossed in for salt
-
Charles staceyw wrote:Charles wrote:
staceyw wrote:
Charles wrote:
JoshRoss wrote:
What is this dry-ah thing? Or at the very least, how do you spell it?
"Dryad is an infrastructure which allows a programmer to use the resources of a computer cluster or a data center for running data-parallel programs. A Dryad programmer can use thousands of machines, each of them with multiple processors or cores, without knowing anything about concurrent programming."
Andrew Birrell is involved in that project - it must be good. He also has an Automatic Mutual Exclusion (AME) project for alternative to explicit locks that looks interesting.
I sense a Going Deep with Andrew in future...
C
That would be awesome Charles!!
I have been big fan ever sense his "Programming Threads in C#" paper. The first, and best, paper on managed threads and locks (granted I have not got Joe Duffy's book yet).
Hmmm. That gives me a crazy dream. How about a panel talk on Concurrency futures with Joe Duffy, Andrew Birrell, Don Box, Dan Reed, and George Chrysanthakopoulos. Don tossed in for salt
Excellent idea. Let me if see if I can make this happen. I'd want to add Herb Sutter, Erik Meijer and Anders as well. And three cameras.
Thank you for the suggestion. Let's make this happen.
C -
staceyw Charles wrote:
Excellent idea. Let me if see if I can make this happen. I'd want to add Herb Sutter, Erik Meijer and Anders as well. And three cameras.
Thank you for the suggestion. Let's make this happen.
C
Oops. Forgot about Erik and Anders. Many apologies. They are a must. That would be so extreme - The Concurrency Dream Team (CDT). -
ymasuda_ Differentiating computing unit scale in new decade of Wintel?
Microsoft has server specific core parallel computing. In other words, Microsoft is developing high scale computing area of mainframe computer architecture, with Intel. Mainframe computers have value differentiation of multi-processors management, and Microsoft has multi-cores management. System buyers may guess past long decade of large scale mainframe versus small scale personal computing.
Innovative processors development is in different side of hardware business, and given development point is how intelligent software solution may add generic values to multi-processor units under Windows resource management. Let's evaluate how Windows became bigger beside mainframe computers.