| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Community
Forum: Chat Room
Thread: Average # of results going down, average days of cpu time going up for me |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 5
|
|
| Author |
|
|
RicktheBrick
Senior Cruncher Joined: Sep 23, 2005 Post Count: 206 Status: Offline Project Badges: 
|
December 6th of last year I had 39 days of cpu time. I had close to half of the days with more than 30 days of cpu time. I had only an average of 112 results per day. In September of last year I had much fewer days of over 30 days of cpu time but I had an average of 158 results per day. So even though I donated about 15% more cpu time in December I had about 70% of the number of results. I have noticed that the number of results that take over 15 hours has increased dramatically for me. So that is why I figure my cpu time is going up and the results are going down. As a whole there has not been that great of a swing in the number of results so I can only figure that I have a good record over the last 5 years so they are sending me work that requires more cpu time than the average user. Doing a rough calculation I figure my average time per result went from 4.8 to 6.8 hours.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
WCG extends or shrink WU sizes as the servers allow, based on performance. Usually the first to have this done are FAAH & HPF2. That's the only reason you see [average] duration changes on the same devices. In general [very simplistically], there's no special function that checks if any device can handle long or short, only if deadlines can be met [before the next scheduled reconnect, if that is controlled via the "connect ever xx days" setting.
Here's the mean run time for the past 3 months, per science, showing how run times vary. http://bit.ly/WCGART. Clearly you see major step change on HPF2/FAAH/C4CW, others go flat with a slow reduction as a result of ever more powerful devices joining the crunch, some move wildly based on the toughness of targets and proteins/ligands to process. Summary, nothing to do with your devices, maybe/surely a substantial element of the science mix that your devices are processing [according your device profile selection] --//-- |
||
|
|
GeraldRube
Master Cruncher United States Joined: Nov 20, 2004 Post Count: 2153 Status: Offline Project Badges: 
|
WCG extends or shrink WU sizes as the servers allow, based on performance. Usually the first to have this done are FAAH & HPF2. That's the only reason you see [average] duration changes on the same devices. In general [very simplistically], there's no special function that checks if any device can handle long or short, only if deadlines can be met [before the next scheduled reconnect, if that is controlled via the "connect ever xx days" setting. Here's the mean run time for the past 3 months, per science, showing how run times vary. http://bit.ly/WCGART. Clearly you see major step change on HPF2/FAAH/C4CW, others go flat with a slow reduction as a result of ever more powerful devices joining the crunch, some move wildly based on the toughness of targets and proteins/ligands to process. Summary, nothing to do with your devices, maybe/surely a substantial element of the science mix that your devices are processing [according your device profile selection] --//-- That is correct and it also depends on the operating system  |
||
|
|
RicktheBrick
Senior Cruncher Joined: Sep 23, 2005 Post Count: 206 Status: Offline Project Badges:
|
I import stats into a spreadsheet in order to find out trends and other information. Since cpu time is in the format yy:dd:hh:ss and because that is not a variable that my spreadsheet will sum, I have to first import the data into a word processor. I do this so that I can replace all the : with spaces. I than import this data into the spreadsheet since I can than tell it use a space to determine the columns. I just wish there was an easier way since that takes up a lot of time if I want to import a lot of stats.
|
||
|
|
wplachy
Senior Cruncher Joined: Sep 4, 2007 Post Count: 423 Status: Offline |
I import stats into a spreadsheet in order to find out trends and other information. Since cpu time is in the format yy:dd:hh:ss and because that is not a variable that my spreadsheet will sum, I have to first import the data into a word processor. I do this so that I can replace all the : with spaces. I than import this data into the spreadsheet since I can than tell it use a space to determine the columns. I just wish there was an easier way since that takes up a lot of time if I want to import a lot of stats. This thread offers a number of solutions to the conversion problem.
Bill P
 |
||
|
|
|