| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Help Cure Muscular Dystrophy - Phase 2 Forum
Thread: WU 170% (NOT) done? |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 14
|
|
| Author |
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1931 Status: Offline Project Badges: 
|
Anyone care to explain this
---------------------------------------- : Stopping and restarting the science application from the BOINC manager did not solve this. For now, I have suspended this WU, a few others have finished fine and or are processing for now as usual... Ralf  |
|||
|
||||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Let it run out to 12 hrs CPU, suspect 8:20 will finish, happens from time to time when running over 6 hr "soft stop"
|
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1931 Status: Offline Project Badges:
|
Let it run out to 12 hrs CPU, suspect 8:20 will finish, happens from time to time when running over 6 hr "soft stop" I have seen several WU's going over 6hrs, but never this +100%...Will start it back up again, but somehow I have the feeling that it still won't finish, would be happy if I am wrong though (as that would count nicely against my attempt to get Sapphire for HCMD2  )Ralf |
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1931 Status: Offline Project Badges:
|
Well, I'll be darned, I was wrong. After 9h and 200% being done it finished and said it was ready to report.
----------------------------------------Wonder if I now get 200% of run time and points credited...  Ralf [Edit 1 times, last edit by TPCBF at Jul 19, 2012 7:42:54 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Nope, but you will get 9 hrs and points as per system, ;) is 100 % for first 60% and 100% of the remaining 40%. Depending on the wingmans times could be good points, but don't bet on it
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
The screenshot already told the story flat: 7:37:29 was the last checkpoint save, and the task had run CPU time of 7:37:54 up to that point with an elapsed of 7:39 (high efficiency indicator)... zero worries.
It's been an all project long fluke of HCMD2 in particular. It's because of the 60% positions complete at the 6 hours soft-break and the client having at times having a hard time equating where the task progress is. Main rule: Watch checkpoints saving in the BOINC Manager Task properties or much easier BOINCTasks which has a column showing how much time has passed since last checkpoint [and how many checkpoints were made since start/restart. Since the last days are upon us... |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello SekeRob
The last days are upon us...this sounds good (or ominous?) I am such a newbie here that I barely understand the language...perhaps a quick summary for a hopeful dad? |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
 Good question, kismetix. HCMD2 is a unique type of project. We run through a dataset and return results for the first part after a few hours. Based on the percentage returned, the remaining unprocessed dataset is chopped up into smaller pieces and sent out again. Each time a dataset is chopped up, the smaller pieces are called a new generation. Eventually, a work unit is so small that it is completely processed. Eventally all the results returned are assembled into a complete result for the original dataset and returned to the project scientists. This is complicated and means that percentage completed refers to the time that the program expects to run before returning a partial result. But the algorithm may decide to run shorter or longer. TPCBF thought that his work unit was taking longer than expected because it had gotten caught in an endless loop. However, SekeRob pointed out that the screenshot said that the program had reached a checkpoint in the code only seconds before TPCBF made the screenshot, so it was still running well, not looping.  That is why Sekerob urged TPCBF to continue running it. And sure enough, it completed in about 2 more hours.The original size estimate is just a guess in HCMD2. Lawrence |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello Lawrence
Thanks for the update, greatly appreciated. I am curious, with all the results returned from the grid, is there an actual model of the protein (dystrophin (?)) being built as a -say- computer assisted drafting model (CAD) somewhere? In other words, as the results are being returned from the grid, is this information used to construct a 3D model of the protein? Is this the image we see on the simple view of the World Community Grid Window along with the other pictures? |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello kismetix,
The original HPF1 screensaver showed pretty 3-D models of various proteins which I would spend minutes watching. The results from HCMD2 are much more mathematical than graphical. It is a pity since pretty pictures are a great advertisement. Lawrence |
||
|
|
|