| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Active Research
Forum: Smash Childhood Cancer
Thread: SCC work units for FLI-1 and MyoD1 targets |
| No member browsing this thread |
|
Thread Status: Active Thread Type: Sticky Thread Total posts in this thread: 24
|
|
| Author |
|
|
Cyclops
Senior Cruncher Joined: Jun 13, 2022 Post Count: 295 Status: Offline |
Hi Aperture_Science_Innovators and alanb1951, the earlier comment was a miscommunication on my part, sorry for the confusion. It was explained to me that there is a system of delayed feedback where the number of workunits that end up going out is directly related to how many are coming back over time, but there is a minimum amount of work generated. When we first start sending out work, it makes sense that there would be a lag in availability and we also will monitor and adjust the rate of churning through batches to produce workunits for download. Now that SCC1 workunits are being returned, we expect an increase in the amount that will be sent out again.
|
||
|
|
alanb1951
Veteran Cruncher Joined: Jan 20, 2006 Post Count: 873 Status: Offline Project Badges: 
|
Cyclops,
Thanks for that information on the flow-control - it looks like a good system (provided there aren't lots of job-hogs delaying the return of work!) Cheers - Al. |
||
|
|
Dayle Diamond
Senior Cruncher Joined: Jan 31, 2013 Post Count: 450 Status: Offline Project Badges: 
|
Cyclops, what's the point of ensuring that supply and demand meet?
What's the cost of loading for more or even all batches into the system and exceeding demand? |
||
|
|
Speedy51
Veteran Cruncher New Zealand Joined: Nov 4, 2005 Post Count: 1265 Status: Offline Project Badges: 
|
Dayle to your 2nd question I am going to say they possibly don't have the storage space to load all of the batches or maybe they need batches returned before more can be created
---------------------------------------- [Edit 1 times, last edit by Speedy51 at Aug 23, 2023 9:22:24 PM] |
||
|
|
Dayle Diamond
Senior Cruncher Joined: Jan 31, 2013 Post Count: 450 Status: Offline Project Badges:
|
I thought the former might be true, but then the limit would be a hard gigabyte one, not an adjustable throughput one.
The latter is true for ARP, but if its not true for SCC or any other project, then they'd just load all those in one go. |
||
|
|
sptrog1
Master Cruncher Joined: Dec 12, 2017 Post Count: 1557 Status: Offline Project Badges: 
|
I like the idea of a adjustable thruput. But I just came within 35 seconds of an empty list. I have signed up for all task categories and my event list says none are available. I am going to blame the job hogs. I , however, do not want to see a return to constipation.
|
||
|
|
Hans Sveen
Veteran Cruncher Norge Joined: Feb 18, 2008 Post Count: 802 Status: Offline Project Badges:
|
Hi!
Seems like SCC1_000xxxx_MyoD1 is on pause for the time being, new targets is issued:I just got some SCC1_0004221_FLI1-A and SCC1_0004222_FLI1-A. H.Sveen |
||
|
|
alanb1951
Veteran Cruncher Joined: Jan 20, 2006 Post Count: 873 Status: Offline Project Badges:
|
I wondered when there'd be work for the FLI-1 target, and here it is at last :-)
I note that this receptor is about 2/3 the size of the MyoD1 ones, so it's no surprise that these tasks seem to take a lot less time than the MyoD1 WUs. I reckon we'll get through a batch of these quite quickly :-) I'd be interested to know what, if anything, is ever going to happen with the MyoD1-C target that had the ill-formed data file; do the scientists plan to fix it, or has that particular case been abandoned? Cheers - Al. |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7580 Status: Recently Active Project Badges: 
|
I note that this receptor is about 2/3 the size of the MyoD1 ones, so it's no surprise that these tasks seem to take a lot less time than the MyoD1 WUs. I reckon we'll get through a batch of these quite quickly :-) You are indeed correct. I have noticed these new SCC units are running very quickly. On my I7-2600 (Linux Mint) many are finishing in 30-45 minutes. Even on my slower systems they are only taking 60 to 90 minutes. I am guessing on quick modern system, they may take 10 to 15 minutes a piece (maybe even less). I wonder how long they may last. Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
nivrip
Senior Cruncher North Yorkshire Joined: Sep 13, 2007 Post Count: 262 Status: Offline Project Badges:
|
Haven't had any SCC WUs for over a week now. Thought that they had stopped.
----------------------------------------
ЮРКШИР КРУНЧЕР
|
||
|
|
|