| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: The Clean Energy Project - Phase 2 Forum
Thread: Research Log: Updates from the Harvard Team |
| No member browsing this thread |
|
Thread Status: Active Thread Type: Sticky Thread Total posts in this thread: 1840
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Well, Harvard closed early today, public transport is suspended, all travel on the roads banned, state of emergency declared, and we’ve got white-out conditions outside. But aside from that we are all tucked in warm and cosy at home and enjoy a round of quiet and peaceful coding. Now, where’s the hot chocolate?
http://www.boston.com/metrodesk/2013/02/08/na...HUKvWpeDP5hf9K/story.html |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
It’s still snowing though not that hard anymore and we also didn’t lose power or heat. So, all is well. And one has to say, it’s really pretty out there – in an Alaska kind of way
---------------------------------------- . Have a great weekend, everybody and enjoy some impressions from winter-wonderland!http://www.boston.com/news/local/massachusett...h6RUuSJR327K/gallery.html [Edit 1 times, last edit by Former Member at Feb 9, 2013 2:49:24 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
1C of global warming is 7% more water vapor retention capacity of the atmosphere. Combine that with aerosols, microdust etc from whatever HSS source [the catalysts to generate rain/snow], and the MGS fork that has developed over the last few years, you've got a big snow/rain generation increase capacity on the US east. Factual, we have 0.8C+ already in GW.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
A great Sunday to all the CEP crunchers with something funny from xkcd. Been there, done that. Enjoy!
http://xkcd.com/1089/ |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi SekeRob,
yes, the climate is fragile as it is quite and dumping gigantic quantities of greenhouse gases and aerosols into the atmosphere should strike everyone as a very bad idea... Best wishes from Your Harvard CEP team |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Here is an interesting discussion provided by Sekerob concerning the new BOINC version and tweaking app_config.xml for concurrent jobs control. Check it out!
http://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,34705 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
We are working together with our friends at Q-Chem to improve some algorithmic issues (concerning bootstrapping and orbital projection) that have popped up in the quantum chemistry code that we use. Nothing wrong but somewhat suboptimally designed. CEP is a great testbed for these issues.
|
||
|
|
Rickjb
Veteran Cruncher Australia Joined: Sep 17, 2006 Post Count: 666 Status: Offline Project Badges: 
|
A question related to optimisations in the Q-Chem code:
----------------------------------------Are the Q-Chem x86 binaries generated purely by high-level programming-language compilers (C, FORTRAN, etc), or do they also contain hand-optimised code that uses some of the x86 SIMD instructions? I'm thinking that using function calls to routines like those in Intel® Math Kernel Library (Intel® MKL) could provide significant speedups. I have observed that running CEP2 consumes more power and on overclocked CPUs requires more core voltage than when running only the other WCG sciences (perhaps excepting those running CHARMM), suggesting that the SIMD instructions may already be used. However, this may be only because Q-CHEM & CHARMM comprise highly-optimised but vanilla FORTRAN code, rather than code generated by something less obsessed with the runtime speed of numerical calculations, like C/C++. [Edit 2 times, last edit by Rickjb at Feb 12, 2013 1:52:28 AM] |
||
|
|
Amelior Scout
Cruncher United States Joined: Dec 20, 2011 Post Count: 17 Status: Offline Project Badges: 
|
Look what was found on today's Google alert:
----------------------------------------Cheap Solar Cell Spray Paint http://www.futurity.org/science-technology/fo...ar-cells-use-spray-paint/
~Life is ever-changing~
----------------------------------------[Edit 3 times, last edit by Wolf834 at Feb 12, 2013 1:07:15 AM] |
|||
|
||||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
A question related to optimisations in the Q-Chem code: Are the Q-Chem x86 binaries generated purely by high-level programming-language compilers (C, FORTRAN, etc), or do they also contain hand-optimised code that uses some of the x86 SIMD instructions? I'm thinking that using function calls to routines like those in Intel® Math Kernel Library (Intel® MKL) could provide significant speedups. I have observed that running CEP2 consumes more power and on overclocked CPUs requires more core voltage than when running only the other WCG sciences (perhaps excepting those running CHARMM), suggesting that the SIMD instructions may already be used. However, this may be only because Q-CHEM & CHARMM comprise highly-optimised but vanilla FORTRAN code, rather than code generated by something less obsessed with the runtime speed of numerical calculations, like C/C++. Q-Chem is a proprietary program, which by the token of WCG-IBM not having permitted it to run on the grid unless, had it's code divulged. The programmers/techs found some areas to improve upon, but they could not share that under their NDA. |
||
|
|
|