| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Nutritious Rice for the World
Thread: BOINC underestimating RICE WU's? |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 35
|
|
| Author |
|
|
Randzo
Senior Cruncher Slovakia Joined: Jan 10, 2008 Post Count: 339 Status: Offline Project Badges: 
|
You should not undercloack you proc. It is not good idea. As far as i know CPUs when are made from silicon wafflers are testet for clock speed. So the actual CPU speed is optimal and stable you should not change it unless you need extra computational power very much. Especially do not play with it on laptops. They can adjust clock speed to current requirements.
Much better is to set CPU usage in you preferences. You can run on full speed 80% of time and have 20% of time to cool down you CPU. I use it on my laptop too. And cleaning is very simple. Remove covers and use vacuum cleaner (or power of your lungs if you are nosmoker :-)). You do not need compressed air. I clean my PC with it (vacuum not lungs) and its totally fine. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
What's the theory against early caching up? It doesn't seem to matter much, the way I'm thinking about it. Yes... once the cache is full BOINC requests more work only to match what has been crunched, to keep the cache at that level. But if you do it all at once (go from 1 day to 10 days), you may see some work unit expiry happening before things even out; if you ramp up a couple more days at a time (every couple days), then no - doing so 'early' really should not matter, if you know you'll need an extra few days to reach the next badge. I'm not increasing the cache here yet because it looks like I will make sapphire without doing so. Now, if the WU's all start disappearing and it looks like they'll run out before the 11th, then I'll increase it to ensure I have enough to make it there. :-) |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I see. Thanks.
|
||
|
|
rembertw
Senior Cruncher Belgium Joined: Nov 21, 2005 Post Count: 275 Status: Offline Project Badges: 
|
If you did buffer up to 10 days all at once, and see that your units are going into high priority, then a possible remedy is to cancel some uncrunched units with the nearest expiry date.
While canceling units is not something I would advise on principle, sometimes it can come in handy. In this case the project will refill your buffer with WU's with a later expiry date, and the project can redistribute the canceled ones without running into too long a delay. At the same time your computer can crunch without having units timing out. (as explained by ZoSo) If the buffer is increased in the 2-4-6-8-10 scheme (or similar), then the time-out dates will be spread out automatically. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
But if you have expiry when going from 1 to 10, won't you still have expiry issues if you do it incrementally.
Does it not take you ~9.75 days to get to those last work units? The expiration time is determined at the time of download. That work unit is more or less always going to be crunched 10 days after its downloaded. Since the expiration dates are further out than 10 days, there shouldn't be an issue. |
||
|
|
tombell12
Advanced Cruncher Australia Joined: Oct 8, 2009 Post Count: 87 Status: Offline Project Badges: 
|
For my laptop it's rather simple, unscrew the CPU and memory cover and some compressed air. Sitting on a notepal with 3 fans, but on really hot days I open up the DVD player slot and if necessary throttle is down further... hi season it goes to 50-60 percent, using ThreadMaster GUI or TThrottle for the automated temp control. I did manage to find a manual online for my laptop model. I may look at it a bit more. Didn't look too hard upon first glimpse :) I have tried TThrottle on full but it can't keep up with the absurd rate at which the laptop heats up on full spec hence it shuts itself down not long after. Putting the laptop on it's side worked a little longer but still succumbed to heat  |
||
|
|
tombell12
Advanced Cruncher Australia Joined: Oct 8, 2009 Post Count: 87 Status: Offline Project Badges:
|
You should not undercloack you proc. It is not good idea. As far as i know CPUs when are made from silicon wafflers are testet for clock speed. So the actual CPU speed is optimal and stable you should not change it unless you need extra computational power very much. Especially do not play with it on laptops. They can adjust clock speed to current requirements. Much better is to set CPU usage in you preferences. You can run on full speed 80% of time and have 20% of time to cool down you CPU. I use it on my laptop too. And cleaning is very simple. Remove covers and use vacuum cleaner (or power of your lungs if you are nosmoker :-)). You do not need compressed air. I clean my PC with it (vacuum not lungs) and its totally fine. I haven't had any issues with the Rightmark Underclock utility on my lappy and have used it for quite a while now. I have dust bunnies in my laptop which causes the 1.73Ghz Core Duo to run far hotter than it should at full specification. I'd be taking a risk just running at full spec under moderate load, it gets that hot I suppose my mouth could be a good option in getting dust out but I haven't ever tried these compressed air sprays for myself. I don't think I could trust a vaccum near my laptop, would probably suck the whole thing up LOL  |
||
|
|
tombell12
Advanced Cruncher Australia Joined: Oct 8, 2009 Post Count: 87 Status: Offline Project Badges:
|
If you did buffer up to 10 days all at once, and see that your units are going into high priority, then a possible remedy is to cancel some uncrunched units with the nearest expiry date. While canceling units is not something I would advise on principle, sometimes it can come in handy. In this case the project will refill your buffer with WU's with a later expiry date, and the project can redistribute the canceled ones without running into too long a delay. At the same time your computer can crunch without having units timing out. (as explained by ZoSo) If the buffer is increased in the 2-4-6-8-10 scheme (or similar), then the time-out dates will be spread out automatically. I figured that there wasn't much option other than cancelling WU's that I will not make in time. I don't feel toooo bad about it now though. They are all due on the 1st of March and with the amounts of replications used for RICE I don't know that they would need to send my unit to someone else. I would just be the one who "aborted". I've adjusted the cache to just one day for now and will likely ramp it up gradually as the end comes closer  |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Each RICE simulation requires 120,000 seeds. If YOU've aborted a task due overcaching, YOU are responsible for the delay in getting there. Aborts/Errors get immediately replaced when that 120,000 target has not been reached. The No Reply waits for reissue to find if the other 18 get to the total (Actually statistically, knreed mentioned early in the project, that something like ~17.2 results are needed, but getting all 19 enhances resolution of the target, like the difference of a 1024 pixel LCD or a 1200 pixel LCD. May even be mentioned in a help item). In short, not getting the full 19 reduces maximum definition for the scientists! Feeling guilty?
----------------------------------------
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 2 times, last edit by Sekerob at Feb 26, 2010 7:36:26 AM] |
||
|
|
rembertw
Senior Cruncher Belgium Joined: Nov 21, 2005 Post Count: 275 Status: Offline Project Badges:
|
But if you have expiry when going from 1 to 10, won't you still have expiry issues if you do it incrementally. In the end, yes of course.The difference is that a user can see when (s)he is nearing the maximum capacity for his/her system thus preventing time-outs. This also eliminates the need to cancel work units which, again, is not advisable. It is difficult to know all the internal workings of each and every project, and canceling WU's may or may not harm the project. It does however damage the reliability status of the users system in WCG, and most of the time increases the load on the network connection if the WU's are sent out a second or third time. |
||
|
|
|