| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Community
Forum: Chat Room
Thread: Run Time -vs- Results ranking |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 6
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Just wondering about the large discrepancy between my Run Time -vs- Results ranking...
It seems the ones of you that have been around a long time have more run time -vs- results or they are about equal , but I have more results than run time. What do you all think is more responsible, the fact that I am running Linux or is it because of the x980 CPU or is it something else ? I know Linux produces less points but I am really only looking at #1--results then #2--run time Total Run Time (Rank) (#6,902) Results Returned (Rank) (#2,954) Doesn't really matter but I just wanted to hear your opinions...  |
||
|
|
BobCat13
Senior Cruncher Joined: Oct 29, 2005 Post Count: 295 Status: Offline Project Badges: 
|
Most likely the fast CPU is the cause. Not all WCG volunteers have fast processors.
My dual core is 3.0Ghz, the quad core is 3.3Ghz and my rankings are: Total Run Time (#4,734) Results Returned (#1,593) |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Pretty sure it's because faster processors give more points per unit of runtime.
----------------------------------------"Faster" comes from clock speed, but also from architecture: Each generation of CPU architecture tends to be more efficient than the previous generation, in the sense that it can do more work per clock cycle. Crunchers that have been around for a while may have accumulated years of runtime on Pentium 4-era processors, which contribute just as much runtime per core per day as a 980X or Sandy Bridge does, but considerably less points and results. [Edit 1 times, last edit by Former Member at Apr 13, 2011 3:41:26 AM] |
||
|
|
anhhai
Veteran Cruncher Joined: Mar 22, 2005 Post Count: 839 Status: Offline Project Badges: 
|
Another thing to take into consideration is what projects ppl are crunching. Right now, I am crunching a lot of HFCC, but over the next few days, I will be switching over to other projects. I figure that the number of results I crunch per day can easily double.
---------------------------------------- |
||
|
|
pcwr
Ace Cruncher England Joined: Sep 17, 2005 Post Count: 10903 Status: Offline Project Badges:
|
2 Points:-
----------------------------------------* Different projects' WUs take different lengths of time to complete. If you go to "my grid" , "my statistics" you can workout the average time each projects' WUs take. faah takes alot longer to do than hcc etc. * Faster CPUs take less time to complete WUs. My fast PC takes half the time that my low PC takes to do the same type project WUs. Patrick  |
||
|
|
sk..
Master Cruncher http://s17.rimg.info/ccb5d62bd3e856cc0d1df9b0ee2f7f6a.gif Joined: Mar 22, 2007 Post Count: 2324 Status: Offline Project Badges: 
|
The runtime rank remains affected by all those people who had slow computers and ran for some time. This affect lasts longer than the points and tasks completed rank, so it takes longer to rise in the runtime rank to begin with, even with many cores/threads.
For a long time my runtime rank was way lower than the points or credit. When you get inside the top 1000 its more likely to level itself out; 617, 600, 646, but I do have a good mix of systems running. tomast, you have a good mix of project contribution; HCC probably has the fastest task turnover, and HFCC has one of the lowest, so the tasks completed it a fair reflection of your contribution. If you had just been crunching for HCC then your tasks returned rank would have been much higher, giving a more skewed view of your contribution. One other thing, if you had only been crunching for CEP2 then your runtime would be slightly less, due to relatively high discrepancy between CPU time and elapsed time for that project. |
||
|
|
|