| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Community
Forum: Chat Room
Thread: How many years are done on the HIV/AIDS project? |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 6
|
|
| Author |
|
|
mmcphail1974@gmail.com
Cruncher Joined: Nov 23, 2005 Post Count: 4 Status: Offline Project Badges: 
|
Exactly how many supercomputer years of calculations are done on the HIV project so far? Is there any way to look on the site and see a live total of the estimated time it would take a supercomputer to do the calculations has already been done?
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
You can see how much computer run time has been spent on FightAIDS@Home by clicking Statistics - By Projects - FightAIDS@Home. But you are on your own if you want to convert it into some supercomputer equivalent.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Exactly how many supercomputer years of calculations are done on the HIV project so far? Is there any way to look on the site and see a live total of the estimated time it would take a supercomputer to do the calculations has already been done? Well since WCG *IS* essentually a supercomputer (I think we'd have the power of like a top 10 supercomputer in the world) then I'd say since the project started on Nov 19, 2005 and it's December 4, 2005 it has <16 days CPU time of a supercomputer, I say less than because people run HPF too, esp the Linux people who can ONLY run it. If you're asking how much supercomputer time has worked on it in general, not just by WCG then I'd have to say it's a mix between an elephant and a rhino |
|||
|
||||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Someone mentioned in another post that your "average" Athlon 2800+ runs at about 0.66 GFlop/s. If BlueGene/L runs at 280 TFlop/s, that gives a factor of approx 425000 (i.e. that BlueGene/L is that many times more powerful than your average PC on the Grid.
The WCG has worked on the FAAH project for a total of 1,108.5 years of "average" computer time (at time of writing). Taking that as a literal figure (not taking into account redundancy which might reduce that figure by a factor of around 5), we can calculate that BlueGene/L would have churned through the same amount of calculations in 0.0026 years, which works out to just under one day. Without data redundancy, BlueGene/L might have gotten the same results as the WGC has from 16 days of work within about 4.5 hours. But then BlueGene/L is called the fastest supercomputer in the world for a reason. By contrast, the 10th place supercomputer (a Cray XT3 system) would have gotten through the data in just over 13 days (around 2.6 days without redundancy). |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
To add to the above, since the power of the Grid has substantially increased since the beginning of the FAAH project (we are now running at about 28.4 TFlop/s total - derived from yesterday's figures and assuming that 0.66 GFlops is correct as the average), to run the same amount of data through the Grid now would take about 13.6 days instead of 16 (or 14.45 days instead of 17... how far in are we anyway?)
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Check out this page. It has some interesting stats on supercomputers and PCs.
http://uk.moneybee.net/ueber_rechenleistung.asp |
||
|
|
|