| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Community
Forum: Chat Room
Thread: The Road to 600 TeraFLOPS: Need More Powah to stay Smarter Than Watson. |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 86
|
|
| Author |
|
|
astrolabe.
Senior Cruncher Joined: May 9, 2011 Post Count: 496 Status: Offline |
If there is no other way to do solarEnergy-materials research on the grid other than what the current-form of CEP2 WUs is already doing, then I'm afraid that CEP2 is fighting a losing battle on the grid and dragging WCG's over-all output with it It's always so very unhelpful to have one of the biggest experts in WCG design and implementation wade in on this subject.  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
And so it looks like today we're doing the real 450 CPU years in one day, without any not-normal validation slippage from prior days.
http://bit.ly/WCGQLK http://bit.ly/WCGFPY Crunch On --//-- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Last week it was not to be... a validator/transitioner out, but today it's "on the record", 452 CPU years and first time over half a billion points (503,977,487) without any big spillovers... that's about 720 TeraFlops
Crunch On --//-- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Wow simply amazing.
Can't wait to see what happens when gpu WUs get-out. |
||
|
|
KWSN - A Shrubbery
Master Cruncher Joined: Jan 8, 2006 Post Count: 1585 Status: Offline |
Wow simply amazing. Can't wait to see what happens when gpu WUs get-out. Easy, we'll just set the goalposts in PetaFLOPS  Distributed computing volunteer since September 27, 2000 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Think with the server upgrade and the credit inflation, there's no escape but having to align with the doubling of Cobblestones required for the same amount of performance. Was 100 Cobblestones (credit, not points) per GigaFLOPS, but since since the new system was applied at SETI increased to 200 Cobblestones/GFL. See
http://boinc.berkeley.edu/wiki/Computation_credit At any rate, this morning's first "new" statistics show that the per CPU year performance per the old way increased from 1.59 TFL to 2.08 per CPU year. At the 452 years of Monday that would have meant we've been running a 940 TFL machine...  Of course this will require some work to get all the charts modded to show the new measure, but in no particular hurry. Let the dust settle first and then work over all the tables. Maybe after the day after tomorrow, Inshallah ;P --//-- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Today we've passed the 800 TFL on 60 days running... (who needs GPU's ;?). On the go added mini graphs of all active sciences and their FLOPS per computed year. Some vary wildly, so the backend is being put under the microscope why some do this and other sciences much less. Anyway, whoever is grabbing HCMD2 jobs, the remaining ones, is getting momentarily over 50 credit per hour
 http://bit.ly/WCGFPY (zoom in or visit the many other detail charts for blow ups) --//-- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Amazing. 800 Tflops..... Great work everyone, let's keep this pace up!!!
|
||
|
|
kffitzgerald
Senior Cruncher USA Joined: Jan 29, 2011 Post Count: 222 Status: Offline Project Badges: 
|
anyone have any idea how I can determine what I am doing with my five system FLOP wise???
|
||
|
|
Bearcat
Master Cruncher USA Joined: Jan 6, 2007 Post Count: 2803 Status: Offline Project Badges: 
|
Ask snurk to make you a signature. Just make a request in his forum.
----------------------------------------
Crunching for humanity since 2007!
 |
||
|
|
|