| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 3593
|
|
| Author |
|
|
spRocket
Senior Cruncher Joined: Mar 25, 2020 Post Count: 280 Status: Offline Project Badges:
|
Currently I have an 096 priority job and an 097 non-priority on my main cruncher, with an 071 on-deck but deep in the queue. One of my other crunchers also has an 096 on priority, and there's a 100 on one laptop and an 095 on the other.
----------------------------------------The 095 has been absolutely slogging. More than once, I've seen the estimated time extend past the deadline, only to get reeled in later. Currently, the estimate is 4.6 hours on a 5.9 hour deadline. EDIT: The 095 finally completed, after 75.98 hours of elapsed time, and came back valid. Talk about a squeaker! [Edit 1 times, last edit by spRocket at Oct 20, 2021 10:17:03 AM] |
||
|
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12594 Status: Offline Project Badges:
|
I have a 104 come in 2 hours ago and a 069 last night.
Mike |
||
|
|
MJH333
Senior Cruncher England Joined: Apr 3, 2021 Post Count: 301 Status: Offline Project Badges:
|
Mike,
097 is now priority. Cheers, Mark |
||
|
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12594 Status: Offline Project Badges:
|
Thanks, Mark.
Hopefully we will get an update from Kevin before I do my next update at the weekend. Mike |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges:
|
Latest stats
Average generation: 98.4 Pace (average time to complete a generation): 4.00 (14-day average) first_indexed generation num_units_currently_on_generation num_units_completed_last_day |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges:
|
One item of note - the extreme laggards were not moving fast enough to be able to catch up before the "average generation" reaches end of project. This would cause a long wind-down at the end of the project and a delay in the scientists being able to do their final analysis. In order to get them moving faster we are now sending out 3 copies of those extreme laggards (defined as 25 generations behind the lead generation).
You can see the impact of this in the data above as it shows that the 6 units furthest behind units all advanced a generation in the past 24 hours. At a larger scale you can see the overall impact of the three strategies (sending extreme laggards to 3 reliable hosts, sending regular laggards to 2 reliable hosts and then treating the rest as normal) in the following data which is based on data from units completed in the past 48 hours. The avg_hrs_to_complete refers to the time from when we loaded the unit and made it available to be processed until the time that it was validated and the next generation was created. num_completed avg_hrs_to_complete type This increase of 0.2% in the runtime needed by the project is significantly helping catch up those units that are furthest behind. |
||
|
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12594 Status: Offline Project Badges:
|
Thank you, Kevin.
13 days have passed since the last report. In that time, the 6 hindmost laggards has moved on 8 generations to generation 017, whereas the last of the main batches has moved on 6 generations. At those rates, catch-up will occur in about 6 weeks. 119,588 units have been validated in the last 13 days (9,199.1 per day), compared with 237,872 results returned, so we have 1,304 results awaiting validation. 097 is the latest generation identified as labelled 'priority'. I will base this response on that. If anyone knows of a later batch please advise and I will adjust my figures. 37.6% of the units are in the priority generations. There are now 3,013,178 units to be crunched to finish the project. The average number of results returned over the last 10 days is up significantly to 18,422.9 per day, so my target for completion is now calculated as 12 September 2022. This assumes no invalids or errors are included in those returns. I suspect we are unlikely to see the project finish much before the end of 2022. Mike |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Kevin,
Can you assign the extreme laggards to the fastest returning machines (mine isn't the fastest I'm sure) or a specific machine. I run 64 concurrently and they all finish within about 18 hours (boinc client has them estimated at 17:23), give or take an hour or so. I would be willing to dedicate the machine to them. I count about 51 WUs identified as extreme laggards so they all could run concurrently. Machine runs 24x7. |
||
|
|
MJH333
Senior Cruncher England Joined: Apr 3, 2021 Post Count: 301 Status: Offline Project Badges:
|
Kevin,
If you wished to take up entity's suggestion of having dedicated machines, I'd be happy to dedicate my PCs which together run 30 ARP concurrently 24/7. Cheers, Mark |
||
|
|
Unixchick
Veteran Cruncher Joined: Apr 16, 2020 Post Count: 1298 Status: Recently Active Project Badges:
|
Thank you Kevin and Mike for the data and stats. It is great to see the small steps of progress we are making. I really like the plan you came up with for the extreme laggards. It is great that you can actually see it working.
|
||
|
|
|