Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
![]() |
World Community Grid Forums
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
No member browsing this thread |
Thread Status: Active Total posts in this thread: 81
|
![]() |
Author |
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
For the Human Proteome Folding Project (HPF) look at http://www.nyu.edu/fas/biology/faculty/bonneau/hpf-results.pdf at Bonneau Lab ( http://www.nyu.edu/fas/biology/faculty/bonneau/ ). It is still being reviewed.
For FightAIDS@Home look at Dr. Olson's Molecular Graphics Lab at http://www.scripps.edu/mb/olson/ The Publications list at http://www.scripps.edu/mb/olson/publications.html does not show any publications for FightAIDS@Home ( http://fightaidsathome.scripps.edu/ ). For a list of update posts, look in Start Here at http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=4925#51997 Lawrence |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
As far as I know, the first Human Proteome Folding papers are still in peer review. They will be announced here on the forums, and will be listed on the project archive page.
All the HPF papers will be published in open access journals. We don't know the publication dates yet. The FightAIDS@Home project isn't ready to publish as far as I know. They are still busy crunching numbers here on the grid. Read the latest update from the project scientists for how things are progressing. They are ready to begin lab testing based on the results of Phase 1a. |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Today we have reached 15,800 years of processing of Phase II of the FAAH project (30,200 total years of Phase I & II processing - 14,400 years of Phase I processing). From the time that the 43% completion mark was reached (http://fightaidsathome.scripps.edu/) it was estimated it would take 25,000 years of processing to complete Phase II. So at this time we have completed ~63% of Phase II and we have about 9,200 years of data left to process. At the present rate of processing it should take about three more months to finsh Phase II, which should occur sometime next January.
----------------------------------------[Edit 1 times, last edit by Former Member at Oct 5, 2006 3:43:55 PM] |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks for the Info Mate.
|
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Maybe they don’t measure completion by finished years but by finished work units. I don’t know, I’m really new to this. Don’t some people have faster computers than others? Also arnt we gaining new members to help with the work units as time goes by? Shouldnth this speed up the project? I don’t know if those are already consiterd in the time till completion equation.
|
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
The estimated finishing time for Phase II of the FAAH project is based on the average of 100 years per day processing. The day when FAAH Phase II reached 43 per cent the number of years that had been processed up to that point was divided by 43 which produced an estimated total processing time of 25,000 years. Keep in mind this is only an estimate based on avaliable data. No information has been provided by the FAAH personnel about any other alternative method of calculating estimated completion times.
|
||
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Sir, since FAAH is quite steady in processing time for an individual computer, I'd be inclined to say that the number of validated work units returned to your lab from the total population you're issuing / going to issue for phase II would give an accurate estimate. The CPU time given the different speeds of computers is subject to tremendous bandwidth. The value on global stats is reported in a mix of wall-clock time and CPU time depending on whether UD Agent or BOINC processed a work unit and at what CPU clock speed and the momentary user load put on a 'device'. The grid now processes phase II on global stats per the box below:
----------------------------------------If u can say for instance: 27mp / 3q / (12mt / 100 ) = 75% it would likely be a better bet. mp = million processed work units mt = millions total net library of work units q - quorum of 3 to get 1 validated net work unit in library thx FightAIDS@Home Project Statistics Statistics Last Updated: 10/08/2006 06:06:03 (UTC) [0 hour(s) ago] Totals: Run Time (y:d:h:m:s) 30,513:268:13:47:38 Points Generated 7,861,698,118 Results Returned 26,956,574 Averages: Run Time Per Calendar Day (y:d:h:m:s) 94:065:01:00:20 Run Time Per Result (y:d:h:m:s) 0:000:09:54:57 Points Per Hour of Run Time 29.41148 Points Per Calendar Day 24,264,500.36420 Points Per Result 291.64307 Results Per Calendar Day 83,199.30247 Yesterday: Total Run Time (y:d:h:m:s) 104:223:03:47:01 Points Generated 29,647,834 Results Returned 80,615
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 3 times, last edit by Sekerob at Oct 8, 2006 6:40:34 AM] |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Sir, since FAAH is quite steady in processing time for an individual computer, I'd be inclined to say that the number of validated work units returned to your lab from the total population you're issuing / going to issue for phase II would give an accurate estimate. The CPU time given the different speeds of computers is subject to tremendous bandwidth. The value on global stats is reported in a mix of wall-clock time and CPU time depending on whether UD Agent or BOINC processed a work unit and at what CPU clock speed and the momentary user load put on a 'device'. The grid now processes phase II on global stats per the box below: If u can say for instance: 27mp / 3q / (12mt / 100 ) = 75% it would likely be a better bet. mp = million processed work units mt = millions total net library of work units q - quorum of 3 to get 1 validated net work unit in library thx FightAIDS@Home Project Statistics Statistics Last Updated: 10/08/2006 06:06:03 (UTC) [0 hour(s) ago] Totals: Run Time (y:d:h:m:s) 30,513:268:13:47:38 Points Generated 7,861,698,118 Results Returned 26,956,574 Averages: Run Time Per Calendar Day (y:d:h:m:s) 94:065:01:00:20 Run Time Per Result (y:d:h:m:s) 0:000:09:54:57 Points Per Hour of Run Time 29.41148 Points Per Calendar Day 24,264,500.36420 Points Per Result 291.64307 Results Per Calendar Day 83,199.30247 Yesterday: Total Run Time (y:d:h:m:s) 104:223:03:47:01 Points Generated 29,647,834 Results Returned 80,615 The 27 million processed work units includes both Phase I and Phase II, not just Phase II. Also, it has never been stated how many "top hits" from Phase I are being processed in Phase II. This is an unknown thus is an unknown number of work units. Plus, during Phase II the quorum requirements were dropped from 5 to 3. This would speed things up a bit, along with more computers being added daily to the Grid, but by how much is unknown. 75% completion of Phase II at this time is possible, but by simple "dead reckoning" from the last updated per centage completed we are at least 63 per cent complete. [Edit 1 times, last edit by Former Member at Oct 8, 2006 5:25:05 PM] |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
63% ?! Here is my reasoning for 52%.
http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=9112#72362 |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
63% ?! Here is my reasoning for 52%. http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=9112#72362 Its possible that we are at 52% based on the work units returned method. This assumes that for each molecule tested requires the same number of work units. This might actually vary from one molecule to another. Also, quorum requirements during Phase II have changed, so this might alter the percentage completed. As for the elasped processing time method, this method assumes the same computers make similar efforts processing the work units over time (and future work units take as long to process as previously completed workunits). During Phase II the default CPU utilization dropped from 100% to 60%. So now due to the change what was 1 year of processing will now be 1.67 years. This can skew estimated completion times. Also, I am assuming an average of 100 years of processing per day and so far this is staying relatively consistant. Regardless, 52% completion by the elasped processing time method is possible. Either method of calculating the estimated % completion of Phase II might be equally valid but both are making assumptions that can produce results that can vary from reality. [Edit 2 times, last edit by Former Member at Oct 9, 2006 6:43:53 PM] |
||
|
|
![]() |