| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Active Research
Forum: Mapping Cancer Markers Forum
Thread: Updated science application to 7.28 for Mapping Cancer Markers |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 37
|
|
| Author |
|
|
gb009761
Master Cruncher Scotland Joined: Apr 6, 2005 Post Count: 3010 Status: Offline Project Badges: 
|
Just thought I'd pop by and thank the Techs (and everyone else - beta testers et al), for getting this new release up and running.
----------------------------------------I now feel a heck of a lot safer for when it comes to restarting my system   |
||
|
|
ryan222h
Senior Cruncher Joined: Sep 4, 2006 Post Count: 425 Status: Offline |
I have a number of 7.28 workunits running over 12 hours on a 4.2 ghz i7 machine. Also noticed the machine only was granted 3 days of runtime when normal is ~8.
----------------------------------------Switched to FAAH until the bugs get worked out of these MCM units. Maybe its just my machine or what?  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
7 days standard deadline, 35% of that for repairs. FAICS there are no bugs of note. Just let it run and keep a small cache... Or, increase the Minimum buffer to above 2.45 days. Theory has it you'd only get the 7 day deadline tasks. None I've heard or seen have run 100 hours on 4.2Ghz machines and yet to report even one late, and if late, there's ~24 hour grace. Thing is, if yours runs long and now, the extra copy on No Reply wont possibly come back before yours.
|
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges: 
|
It'd be interested to hear some internal stats from the techs how MCM is doing in e.g. invalid in quantity and percent of old v new. There's a total results validated reduction for WCG as whole by several 100K from Saturday to Sunday, yet we ran an absolute record this weekend for both Saturday [573 years] and Sunday [572 years] even surpassing last week Monday all time record. Possibly many more venturing onto MCM feeling 'safe' from invalids [MCM did 25 years more than FAAH.... 283 years on Sunday]. (Getting more 7.26 repairs than hoped for... _3 and _4 copies... pushing them ahead to get the system clear of these, as interrupting these is not an option here... power-outs are always lurking, with beautiful weather as we have now had for nearly 2 weeks though unlikely, since Cyclone Nettune wrecked havoc here 2 weeks ago :( ). The reason there was a dip in results validated was due to FAHV running out of work for about 8 hours on Saturday morning. This caused more FAAH and MCM1 work to be sent. These numbers should be back to normal. As for actual stats, those take lots of time to query against the database, I have kicked off the query for 7.28 stats and i'll provide some shortly. Thanks, -Uplinger |
||
|
|
KWSN - A Shrubbery
Master Cruncher Joined: Jan 8, 2006 Post Count: 1585 Status: Offline |
7 days standard deadline, 35% of that for repairs. FAICS there are no bugs of note. Just let it run and keep a small cache... Or, increase the Minimum buffer to above 2.45 days. Theory has it you'd only get the 7 day deadline tasks. None I've heard or seen have run 100 hours on 4.2Ghz machines and yet to report even one late, and if late, there's ~24 hour grace. Thing is, if yours runs long and now, the extra copy on No Reply wont possibly come back before yours. Yup, of all the tasks I've done the longest ran for 50.6 hours on a 2.6 ghz machine. No way a 4.2 wouldn't finish in time unless it was turned off.  Distributed computing volunteer since September 27, 2000 |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
Sek,
We have had just over 300,000 results returned with success status. Of these 177,000 results were validated with less than 100 being invalid. Compared to previous versions where invalids would be in the thousands. Note, these are just stats of results that were either returned or updated within the past 24 hours for version 7.28. Thanks, -Uplinger |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Well think that's enough of a strong number to qualify as less than insignificant, but don't have you stopped from remaining in the flow to fix the
----------------------------------------Good Job (Said Hancock) [Edit 1 times, last edit by Former Member at Dec 17, 2013 2:11:40 PM] |
||
|
|
rbotterb
Senior Cruncher United States Joined: Jul 21, 2005 Post Count: 401 Status: Offline Project Badges: 
|
I cut back my cache to 0.5 days to help my laptop catch up and not have so many MCM1 WUs in panic mode. Looking at my results for the day of crunching, it appears I got a couple 7.28 WUs done OK, though when I looked at their details, I noticed that some of them went through as many as 4 runs of invalids before I got them and ran them to completion. We appear to be crunching through at least some of these WUs, but it does look like some additional 'fine tuning' may be needed before all the kinks are worked out. An I still have a handful of 7.26 WUs to finish up probably before tomorrow before I'm completely onto the new 7.28 WUs going forward.
Oh well, I might get to do some more beta work later this month to help work through some of these loose ends before everything is a-ok for everyone.... |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
rbotterb,
There are always loose ends with applications. Unfortunately not every computer is created without fault. Also, there are always the fun issues that arise due to security software on a members machine. One of the hardest things for us when debugging an issue is recreating it. The invalids were something we finally were able to recreate and thus fix. We will be monitoring the application going forward as we do with all applications to see if there is a bug and attempt to fix it. Like you said, more beta attempts :) Thank you for your patience! -Uplinger |
||
|
|
bieberj
Senior Cruncher United States Joined: Dec 2, 2004 Post Count: 406 Status: Offline Project Badges: 
|
Uplinger,
Thank you for fixing the invalid issue. I did get an invalid caused by yesterday's forced reboot due to automatic patch installation. I doubled checked and they are still 7.26. All applications now being done or waiting to be acknowledged is 7.28 so hopefully that is behind us. Time will tell at next reboot (probably Saturday). Will double check results. -bieberj |
||
|
|
|