FYI, Abstract of late Argonne Paper. - C.
-------- Original Message -------- Subject: Re: [Pc-ad-workshop] any submissions for AD@ICCS06 close to the wire? Date: Tue, 13 Dec 2005 09:00:40 -0600 From: Paul Hovland hovland@mcs.anl.gov To: Christian Bischof bischof@rz.rwth-aachen.de References: 439E59AD.5000508@rz.rwth-aachen.de 439EDAC9.4040601@mcs.anl.gov 439EE16F.3010600@rz.rwth-aachen.de
Sure, here it is.
Automatic Differentiation is the process of translating one program that computes a function $f$ and generating a different program that computes the derivative of that function, $f'$. Activity analysis is important for AD. Our results show that a dynamic activity analysis, checking at run-time, incurs an average overhead of 27% when all independent variable are active. When as few as half of the independent variables are active, dynamic activity analysis enables an average speedup of 28%. We investigate static activity analysis combined with dynamic activity analysis as a technique for reducing the overhead of dynamic activity analysis.
Jean's familiar with this project, so he can be a good arbiter of who would be appropriate referees.
Paul
Christian Bischof wrote:
Paul,
could you please send me an abstract of the paper that you want to hand in late? We want to use this as a guideline in tomorrows phone meeting where we want to assign referees. Of course we would only send the final version then to the referees.
Thanks, Chris
Paul Hovland wrote:
Chris,
I know of no papers in this state other than the one for which I already requested an extension. Sorry.
Paul
Christian Bischof wrote:
Dear Program Committee Members,
at the present time (6am European time, midnight Eastern Time in the US) we have 12 submissions to the AD workshop.
There is consensus among the members of the organizing committee that we do not want to be dogmatic about the deadline, but in light of the tight schedule for review we cannot afford more than a one or at most a two-day extension.
As many of you are aware of paper writing activities geared toward the AD workshop at ICCS06, I would ask you to let me know whether there are papers "close to the wire" that you are aware of that would mature into a paper fit for submission in the period indicated. I will have a telephone conf. with Uwe, Jean and Shaun later today and it would be helpful for us if we had a complete picture of the paper lineup.
With best regards, Chris --
Prof. Christian Bischof, Ph.D.
Institute for Scientific Computing and Center for Computing and Communications
RWTH Aachen University
Seffenter Weg 23, 52074 Aachen, Germany
Tel. +49-241-8029110, Fax +49-241-8022241
bischof@rz.rwth-aachen.de mailto:bischof@rz.rwth-aachen.de, www.sc.rwth-aachen.de http://www.sc.rwth-aachen.de/, www.rz.rwth-aachen.de http://www.rz.rwth-aachen.de/
Pc-ad-workshop mailing list Pc-ad-workshop@lists.rwth-aachen.de http://mailman.rwth-aachen.de/mailman/listinfo/pc-ad-workshop
--
Prof. Christian Bischof, Ph.D.
Institute for Scientific Computing and Center for Computing and Communications
RWTH Aachen University
Seffenter Weg 23, 52074 Aachen, Germany
Tel. +49-241-8029110, Fax +49-241-8022241
bischof@rz.rwth-aachen.de mailto:bischof@rz.rwth-aachen.de, www.sc.rwth-aachen.de http://www.sc.rwth-aachen.de/, www.rz.rwth-aachen.de http://www.rz.rwth-aachen.de/