| 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: Questions about a potential 64 Bit application for the Clean Energy Project |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 16
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
QM = Quantum Monte Carlo (?), which is what the [makes my CPU hot] QMC project does too.
And looking around, there is a Q-Chem - Charmm QM/MM interface, CHARMM being used on DDDT2 Too much minutiaen info ;-) --//-- |
||
|
|
Ingleside
Veteran Cruncher Norway Joined: Nov 19, 2005 Post Count: 974 Status: Offline Project Badges: 
|
QM = Quantum Monte Carlo (?), which is what the [makes my CPU hot] QMC project does too. And looking around, there is a Q-Chem - Charmm QM/MM interface, CHARMM being used on DDDT2 QM = Quantum Mechanics. MM = Molecular Mechanics. MC = Monte Carlo, while QMC is as mentioned Quantum Monte Carlo. MD = Molecular Dynamics.  "I make so many mistakes. But then just think of all the mistakes I don't make, although I might." |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks... it helps other readers to understand what's being referred t00 ;>)
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
WRT the amount of RAM used, 3.2/3.5GB of RAM can be easily reached when running CEP2; you just need a common operating system such as Vista or Win7 that uses over 1GB RAM within seconds of logging on, a couple of desktop applications (Outlook, Firefox) running and you’re up to 2GB. Use 4 threads and you are already pushing 1.6GB bringing you over the 3.5GB mark. 6 threads for an AMD X6 or all 8 threads of an average i7 and you are in deep trouble. Throw in a couple of GPU tasks and you are throttling the drive to the point you are wasting your time. I would not expect too many people to have a 12 threads x86 system but there might be a few. Several non WCG-Boinc projects use enough memory to test a quad on x86. Within a few months 16 to 32 thread systems will start to become common, so do you think a 64bit app for CEP2 would be useful, or would you be happy with the project remaining 32bit for the long term? You have made an error in regards to the bit-ness of the application vs the operating system. Using 64bit Windows eliminates the 3.5GB memory limit, even if you are running 32bit applications. So go ahead and run 16 to 32 CEP2 tasks at one time, as long as your operating system is 64bits and you have enough memory you will be fine, as long as no single process exceeds 2GB (4GB in certain situations). http://msdn.microsoft.com/en-us/library/aa366778%28v=vs.85%29.aspx [Edit 3 times, last edit by Former Member at Mar 19, 2011 8:19:12 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi everybody,
It is correct that 64-bit processors have 64 bit registers (16 integers and 16 double floating points), and 32 bit software can only use half of them. Since this is fast memory and since a 64-bit software can use more of it, it will indeed be faster. The advantage is limited, e.g., because the cache and RAM pipelines can be the bottleneck for both cases (after all the CPU operates on gazillions of floats). The 32-bit instructions set is still implemented in x86-64, so there is no penalty due to emulation. But 32-bit applications may lose some performance if the software infrastructure (e.g., libs) is optimized for 64-bit. Benchmarks for numerical applications (like quantum chemistry codes) still only show small differences. A 5% speedup may be possible. We also think that this is worthwhile and we would like to have a 64-bit version, in particular for the long term (32-bit is a dated technology). But it also has to be put into the context of the implementation effort. It is mostly up to the IBM team when they get around to doing it. Quantum chemists mostly love 64-bit systems because they can use gigantic amounts of RAM on a single job. We don't do this in CEP2 - individual jobs never touch the RAM limitations of a 32-bit app. So as long as the OS is 64-bit, everything should be fine for CEP2. Best wishes from Your Harvard CEP team http://en.wikipedia.org/wiki/X86-64 http://en.wikipedia.org/wiki/Processor_register ftp://ftp.software.ibm.com/software/webserver/appserv/was/64bitPerf.pdf |
||
|
|
KerSamson
Master Cruncher Switzerland Joined: Jan 29, 2007 Post Count: 1684 Status: Offline Project Badges: 
|
Hi everybody,
----------------------------------------Even if I am sharing the general opinion regarding the advantages offer by 64 bit application, I have another concern regarding the CEP2 project. I operate several hosts windows based (XP Pro 32 and 64) as well as Ubuntu based (10.04 64). The difference of efficiency between Win XP and Linux is dramatic and Win XP Pro shows the worst efficiency. For testing purposes, I operate two Phenom II x6 with respectively 8 GB and 16 GB RAM. The Boinc-Data directory is located by the both hosts on SSD. Linux needs around 20 sec for starting a CEP2 WU but Win XP Pro needs at least around 3 minutes sometimes it could take until 5 minutes. I experience similar behavior with a Xeon based host with around 4.5 minutes starting time (in this case on a HDD based system). An Athlon II x2 based system using Linux 64 only with HDD starts a CEP2 WU within less than 25 sec. Regarding the numerous Windows based hosts and the number of the daily computed WUs, an optimization of the CEP2 seems to be absolutely necessary and would increase the computation efficiency significantly. For example, based on "boinc task", WinXP Pro (32 or 64) with CEP2 shows an average efficiency of around 96% (sometimes only 92%) until Linux shows around 98.5 - 99% efficiency. Cheers, Yves |
||
|
|
|