| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: Suggestions / Feedback
Thread: when will 64 bit support come |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 23
|
|
| Author |
|
|
zombie67 [MM]
Senior Cruncher USA Joined: May 26, 2006 Post Count: 228 Status: Offline Project Badges: 
|
Hi Movieman! Long time!
----------------------------------------Anyway, it is all about the app. Few benefit from 64 bit computing. So someone would need to look at each of the many project apps that run under WCG, to see if any would benefit from 64 bit. I'm assuming that someone already did the analysis, and determined that 64 bit would not speed things up. It would be foolish to just ignore it, after all.  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi zombie67,
64-bit applications would run faster. The question is whether or not this would be worth the additional man hours required to maintain the additional queues. At some point, we will probably run a 64-bit application for a project. I don't expect it to be soon. Lawrence |
||
|
|
Movieman
Veteran Cruncher Joined: Sep 9, 2006 Post Count: 1042 Status: Offline |
Hi zombie67, 64-bit applications would run faster. The question is whether or not this would be worth the additional man hours required to maintain the additional queues. At some point, we will probably run a 64-bit application for a project. I don't expect it to be soon. Lawrence Hi Lawrence, Just curious. How many machines would it take before doing a project in 64 bit became a viable alternative? I'm also curious as to how much more work would be done as I asked earlier. If it's 5-10% then I'd agree, not worth the effort. 30% or more then it's something that should be considered IMHO. Maybe one project just done in 64 bit to see what the results would be. I know if you put a medical app like the HCC or FAAH projects on 64 bit I'd be there in a heartbeat and I can think of another 1200+ machines that would be also. I'd love to see and participate in a test of this. Thanks and Happy Holidays to you and yours.  [Edit 1 times, last edit by Movieman at Dec 25, 2008 3:28:34 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi zombie67, 64-bit applications would run faster. The question is whether or not this would be worth the additional man hours required to maintain the additional queues. At some point, we will probably run a 64-bit application for a project. I don't expect it to be soon. Lawrence Hi Lawrence, Just curious. How many machines would it take before doing a project in 64 bit became a viable alternative? I'm also curious as to how much more work would be done as I asked earlier. If it's 5-10% then I'd agree, not worth the effort. 30% or more then it's something that should be considered IMHO. I would think it is more about the improvement in performance than the number of machines, since 64bit machines run 32bit apps just fine. The quick test is the ratio of FLOPS to IOPS, if it is higher than .9 there is no point in looking farther because the maximum performance gain will be about 11%. You gain a little from being able to use larger registers and larger chunks of memory but the big difference is the integer ALU. |
||
|
|
Exar Kun [HoloNet]
Advanced Cruncher France Joined: May 30, 2007 Post Count: 52 Status: Offline Project Badges: 
|
Don't forget that many persons (including me ^^) will install a 64-bit OS if there is an improvement of crunching time. I have one PC under Xp 32 bits, but I will install a 64-bits linux as soon as WCG will complete a 64-bit project...
----------------------------------------  |
||
|
|
GIBA
Ace Cruncher Joined: Apr 25, 2005 Post Count: 5374 Status: Offline |
I wonder see 64 bits releases available to use in WCG projects.
----------------------------------------
Cheers ! GIB@
  Join BRASIL - BRAZIL@GRID team and be very happy ! http://www.worldcommunitygrid.org/team/viewTeamInfo.do?teamId=DF99KT5DN1 |
||
|
|
pball1224
Cruncher USA Joined: Nov 22, 2005 Post Count: 45 Status: Offline Project Badges: 
|
It seems to me that if a particular project decided to write their WU's for 64 bit then they would only be able to be run on a 64bit CPU, OS, and Client, which would be a sever loss in the total number of clients available.
I would venture to say it would be a much greater loss of overall throughput than what might be gained by whatever increase running in 64 bit would allow. |
||
|
|
JmBoullier
Former Community Advisor Normandy - France Joined: Jan 26, 2007 Post Count: 3716 Status: Offline Project Badges: 
|
pball1224,
----------------------------------------You are right about the possible consequences, but there would be a higher possibility that WCG could offer it sooner than what members are really asking. Generally members are not asking for a specific project to be 64-bit exclusively, they are asking (ideally) for a 64-bit version of each application for each platform, which means offering twice more different application versions, with all the feeders, maintenance, etc... which would go with them. That is where WCG is not exactly ready to go nowadays.  Cheers. Jean. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I wonder see 64 bits releases available to use in WCG projects. I hope that some of the other requests that have been posted will come first, such as additional badges, new projects, updated forum system, revivial of DDDT and an update to CEP software. While 64bit will help improve throughput for some users, I believe that projects that benefit ALL users should be considered first, especially if the benefits include lower support requirements. 64bit will double the support requirement immediately and only benefit whatever proportion of users are using 64bit. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I wonder see 64 bits releases available to use in WCG projects. I hope that some of the other requests that have been posted will come first, such as additional badges, new projects, updated forum system, revivial of DDDT and an update to CEP software. While 64bit will help improve throughput for some users, I believe that projects that benefit ALL users should be considered first, especially if the benefits include lower support requirements. 64bit will double the support requirement immediately and only benefit whatever proportion of users are using 64bit. I could give a !@#%^*$%$T$*&%^#%$^&# about badges or new forums. I crunch to advance the science.64 bit would help a lot in my case.Revival of DDT is out of the hands of anyone here and is in the hands of the project scientists.CEP is a total trainwreck. |
||
|
|
|