• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

SIMAP Stats 12/23/05

rise

rise

Diamond Member
project info.

TeAm rank #22- currently only 97 teams so we should be able to get top 10 in a hurry. the top 5 seems pretty tough but who knows? 🙂

TeAm stats as of 12/22/05

i have been running this project off and on for about two weeks now. it seems to run pretty well. two work units will finish in about 5 hours on my rig in sig and i haven't had any sort of computation errors or hung units.

the one thing i did notice is that it seems to dominate my rosetta project. i had originally set it to run 10% of my boinc production (rosetta got the other 90%) but i noticed that rosetta would never start and simap would just keep crunching away until i suspended it.

as a work around, i basically just suspend simap for all but the times i want to run it. the minute i resume simap, rosetta gets preempted and remains idle until i suspend simap again.

so if any of you have experienced, or solved this, let me know 🙂

at any rate, just wanted to get a thread started for one of younger TeAms as an easy place we can find stats/discussion. 🙂
 
frankly, i'm not sure 😛

i know it sounds stupid but i originally joined to "support the TeAm!!" as puddy would say 🙂

then i realized i had a bunch of wu's due tomorrow so i figured i'd finish them up at least. but aside from it's desire to muscle out other boinc projects, it runs very well. the units finish like clockwork, always ~ 5 hours each.

i'll do some more looking (or wait for RD to come by and fill us all in 😛 )

thanks for the links :beer:
 
Originally posted by: rise4310
so if any of you have experienced, or solved this, let me know 🙂
Click to expand...

I thought BOINC watched for when WUs need to be returned and would run those that need to be returned soonest. It also keeps track of percent assigned to each project and how much has been done for each project. So after finishing sooner due WUs will work on other project until the percent assigned evens out.

I think if you let it run long enough it will end up doing what you want it to. Well if it's working right anyway.
 
hmm, interesting. i didn't know it did that and it didn't seem to be doing that when i ran seti with rosetta. it seemed to switch between projects based soley on percentage assigned to each. that was last month so maybe i'm remembering it wrong.

all my simap are due tomorrow so that would make sense but from day one it took over my queue so i think its something else.
 
BOINC will switch into nearest deadline-mode if needed, and can also disable work-request if overcommitted.

One of the reasons to switch into deadline-mode is if there's less than 2N to deadline, there N is your cache-setting.


The important to remember is, even short-term you'll often run in deadline-mode so one project gets nearly all cpu-cycles, the BOINC client keeps track of your long-term dept, and after crunching a low-resource-share project exclusively till out of work it can take days/weeks/months till next time asks for more work.


So, the highly recommended is to just let BOINC do it's job, and not try to micro-manage by suspending individual projects or individual results.


Also, many recommends to use a short cache-setting, since the larger the cache-size and the more projects you're running, the longer before things stabilizes to your preferred resource-shares.
 
Originally posted by: Rattledagger
BOINC will switch into nearest deadline-mode if needed, and can also disable work-request if overcommitted.

One of the reasons to switch into deadline-mode is if there's less than 2N to deadline, there N is your cache-setting.


The important to remember is, even short-term you'll often run in deadline-mode so one project gets nearly all cpu-cycles, the BOINC client keeps track of your long-term dept, and after crunching a low-resource-share project exclusively till out of work it can take days/weeks/months till next time asks for more work.


So, the highly recommended is to just let BOINC do it's job, and not try to micro-manage by suspending individual projects or individual results.


Also, many recommends to use a short cache-setting, since the larger the cache-size and the more projects you're running, the longer before things stabilizes to your preferred resource-shares.
Click to expand...

Yeah, that's what I was trying to say 😉

He is soooo patient with us and sooooooo helpful :thumbsup:

Why isn't he "Elite"? Well anyway he is in my book 🙂

Thanks Rattledagger!
 
jeesh, i didn't realize he hasn't been granted the elite title yet by the powers that be. 😕

well, matters not, RD is elite in my book too 🙂
 
ok, we're 200 small points away from our next overtake, "chicopee" so they may go down at the next update 🙂

also a big welcome to ingleside giving us 8 members :beer:
 
well, we passed chicopee as i thought but, i dunno if it was a statistical glitch, bavaria united threw in a monster day to keep us mired in 21st place.

we still have a few teams on the overtake target 🙂 progress, not perfection 😉
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top