Message boards :
Number crunching :
New credits
Message board moderation
| Author | Message |
|---|---|
[AF>HFR>RR] Jim PROFITSend message Joined: 3 Jun 07 Posts: 107 Credit: 31,331,137 RAC: 0 Level ![]() Scientific publications ![]() ![]() ![]() ![]() ![]()
|
Credits change if i look my results since today CPU time (sec) 88,513.57 claimed credit 3,681.72 granted credit 4,398.94 Can we have explanation for the new calculation of credits? |
GDFSend message Joined: 14 Mar 07 Posts: 1958 Credit: 629,356 RAC: 0 Level ![]() Scientific publications ![]() ![]() ![]() ![]() ![]() |
Credits change if i look my results since today So, before the credits were assigned based on the wu by us, which meant we had to compute the approximate time for each. This is a problem every time we change the workunit to another one of different length. The current method gives credits based on the number of seconds that your client records for the wu multiplied by a number given by the application. This number should be always the same. There have been at least a couple of cases in which there were strange results, like someone computed for 80,000 seconds and was granted only 88 credits :(,. This is time is the contrary. It is odd because the server should give what it is claimed provided that the result is valid. We can come back to the previous method which seemed more reliable until we find a better way. |
©2025 Universitat Pompeu Fabra