Posts by MontagsMeeting
|
41)
Message boards :
Number crunching :
WU abuse
(Message 16282)
Posted 20 Mar 2009 by MontagsMeeting Post:
Please explain that! Your conclusion doesn't have anything to do with what you rely on, or am i wrong ??? |
|
42)
Message boards :
Number crunching :
WU abuse
(Message 16267)
Posted 20 Mar 2009 by MontagsMeeting Post:
it is allowed and wanted to build your own apps if they do the work correctly this is part of any boinc project |
|
43)
Message boards :
Number crunching :
WU abuse
(Message 16264)
Posted 20 Mar 2009 by MontagsMeeting Post: Some go to the north pole, others to the equator in their holidays is it really that hard to understand that not all people like the same I like it to keep my rigs at 100% all the time and "can't understand" why someone can shut down their computers you're not bad because of shutting down your PC in my eyes - it is your way why are the people bad in your eyes when they don't do it your way And it isn't competition, i have done it this way all the time and there was a lot of time when i wasn't competetive. The competition that counts is me against the technic, i like it, it's fun for me. If it isn't important for you why doing MW? Let some more WUs for me - for me it's important ;) |
|
44)
Message boards :
Number crunching :
WU abuse
(Message 16253)
Posted 20 Mar 2009 by MontagsMeeting Post: Yes, it's all about keeping things up and running and not malevolence that someone else is getting more than he deserves |
|
45)
Message boards :
Number crunching :
WU abuse
(Message 16250)
Posted 20 Mar 2009 by MontagsMeeting Post: Surely this calculation only applies to a user who crunches for a single project! I run seti alongside, but as MW is actually the only serious project using ATI-graphics there's nothing to switch to. If seti goes ATI i'm back, they're only whining when servers go down, as it should be |
|
46)
Message boards :
Number crunching :
WU abuse
(Message 16246)
Posted 20 Mar 2009 by MontagsMeeting Post: If they are going to write a script they certainly will make it request work as much as possible. made some self-experiments? :D |
|
47)
Message boards :
Number crunching :
WU abuse
(Message 16230)
Posted 20 Mar 2009 by MontagsMeeting Post: So if I understand you, if you have one place to obtain water wherever you live and you decide that your obtaining water should take precedence over anyone else's and so you decide to try to get every drop of water that you possibly can by whatever means neccessary and then you tell me that I should either do the same or get out of the way, I would be the one who was being greedy? No you didn't understand It is actually the situation, that faster rigs run dry very often if they don't correct the behavior of BOINC while slower rigs have no really problems keeping their client running. if i do 100 WUs and then have to wait 1 hour to get new WUs this will make my rigs utilization about 50% (100WUs/hour) if a slow rig will do 100 WUs (1 WU/hour) and then waits 1 hour it will have a utilization of 99% That's not fair and will favor slow rigs? Your example with the waterhole doesn't fit the situation. I would stop if the waterhole gets hurt, but the waterhole can live without water. To correct your waterhole story there are people who need more water than others, for whatever reason. Let's say there are people who need 100l/h while others need only 1l/h the hole produces only people*50l/h and everybody can carry 20l so the people who need 100l/h have to come 5times/h to get what they need for the next 12 Minutes while the other people need only to come once in 20 hours to fill their needs for the next 20 hours. if it is now only allowed to come once per hour ... Should people be allowed to go to the waterhole if they need to? And now who is greedy? And as i've written i only correct BOINCs behavior to increase time between 2 requests if they aren't successful. Which is meant to reduce serverload in case there's a problem which isn't the case here, so this behavior is wrong and doesn't match the situation. I will either increase repetition or stop completely my script if there will be any problems with the servers. My script actually runs every 5 Minutes and i try this weekend to run it only if boinc can't handle it by itself. And if you follow my posts, i only tried to optimize my rigs. My rigs run dry often enough, but can you see me whining and complaining? I don't do bad to the project nor do i badmouth something or someone which was the intension of this thread and the reason why i write here. Statements like someone should be banned (or whatever else) are antipathetic to me. They are for small people in small worlds and say much more about them as their allegation of a "Me First" attitude. especially if there is no reason for that other than their selfishness. They don't care about the project as there's no problem, so everything left is that they care for themselves only and hide it behind a specious argument. BTW i don't think one minute or less is good for the project, but it's not to me to judge it or even to ban someone. If the servers can handle it, it's the peoples decision to do it or not. |
|
48)
Message boards :
Number crunching :
WU abuse
(Message 16187)
Posted 20 Mar 2009 by MontagsMeeting Post: I simply optimize my systems so they work at best efficiency, not only BOINC and its projects but nearly everything and if i can't i don't complain, i either search a solution or i will live with it. If others are better, faster, smarter, etc.. i'm happy with it or i try to keep up. Me thinks more that people who complain about others way to do things are much more "Me First" "Me right" Me Me Me I don't tell you, you do things wrong or the way you have to do - you try to tell me that your way is better, that i am wrong - who are you, that you can tell me? Seems you want to be right, better, first or whatever live and let live Holger P.S. I don't need access to server logs, etc... if there would be something wrong the project admins can tell and if they tell i will respect it and immediately stop such scripts. But at this point i have to ask if you have access to the servers that you can tell they are overloaded? So why are you complaining about something that you don't know anything about? Could only be that you think you are doing things the right way and everybody who's not doing your way is wrong. I can see servers are reponding normally and admins don't report problems -> so nothing to complain about. Sorry for the complicated way i describe things, but as my primary language is german i'm not able to bring things to a point easily. |
|
49)
Message boards :
Number crunching :
WU abuse
(Message 16173)
Posted 20 Mar 2009 by MontagsMeeting Post: Let's see it from another position - Energy - Energy is expensive, is rare, does pollute the environment, so i think if a project can't produce work for all, the least effective rigs shouldn't get any work, so the effective ones can use their potential. For the try to establish opinions i would like to say that it is not fair, that the slow rigs get enough work let's say their rigs are in use more than 95% while the fast ones are usually below that. So to be fair it would be needed that everyone gets the same percentage of what they could do if everything is fine. As BOINC isn't done for such situations it's needed to correct its behavior either by using update scripts where i see no problem, as it looks like the servers can handle that additional traffic easily, or by stopping the slow machines periodically to let the fast ones come up to the same utilization. The only thing that is really unfair and unefficient is to let the fast and efficient rigs run dry so that the slow and inefficient ones have always work. This just said because i really hate if people want to ban or doing even worse to others because they don't do the things the way they can think of in their petty world. My real opinion is that the project is the thing that is important and as long the project itself isn't hurt, everybody should be allowed to do the things as he likes. That means everybody can use update scripts or anything else as long as the server can handle it. When i started MW i had the setting <report_results_immediately> turned on as i've done seti with only Astropulse WUs which take more than 20 hours, so my rigs reported WUs not more than 10 times per day and rig. With MW this was terrible as i hammered the servers more than 2 times per Minute and that with a regular BOINC setting but i decided to not use it anymore and instead used a update script that connects every 5 minutes and if i find the time i'll write a script that will check if there is no work for MW before. I don't have a bad conscience, as the project works well and without the script i had hours of dry times which means for rigs that do WUs in an hour they would run dry for days/weeks as it would be fair if i have to wait a time where i can do 100 WUs that the other rigs have to wait 100 WUs too. Holger |
|
50)
Message boards :
Number crunching :
3rd.in - optimized apps
(Message 16056)
Posted 19 Mar 2009 by MontagsMeeting Post:
Yes, Vista 64Bit and with BOINC 6.6.17 my problem with "million" active tasks seems to be healed a little - actually i have only 8 tasks active while 3 were really crunched, none in high priority mode. seems it now depends on the number of cpus set in cc_config - still strange |
|
51)
Message boards :
Number crunching :
3rd.in - optimized apps
(Message 16014)
Posted 19 Mar 2009 by MontagsMeeting Post: 9.3 works for me with 0.19e - i had to copy/rename the 3 dlls mentioned in the readme again. |
|
52)
Message boards :
Number crunching :
Milestones
(Message 15947)
Posted 18 Mar 2009 by MontagsMeeting Post: 5 days MW - 400,000 Credit - 30,000 RAC - ~100,000 daily output as after 5 days everything runs smoothly now. BOINC (Seti & MW) - 1,400,000 Credit - 50,000 RAC |
|
53)
Message boards :
Number crunching :
3rd.in - optimized apps
(Message 15651)
Posted 16 Mar 2009 by MontagsMeeting Post: with a setting of <avg_ncpus> greater than 0.02 i loose more and more cpus to mw. at 0.03 one cpu idles, i think i had up to 3 idle cpus but don't know the settings anymore. how many cores do you have? |
|
54)
Message boards :
Number crunching :
3rd.in - optimized apps
(Message 15603)
Posted 16 Mar 2009 by MontagsMeeting Post: Hello i have a problem with the mw 0.19e zslip app, had it with 0.19e too. As you can see in the pic below there are a million 0.19 tasks running. They don't do anything but waiting for the 2 actually done WUs. My System core i7 + Ati 4850 BOINC 6.6.15 + SETI AP v5.03 Optimized + MW 0.19e <avg_ncpus>0.02</avg_ncpus> + <cmdline>f30 n2 w1.2</cmdline> i tried different <max_ncpus> with no effect share SETI/MW is 100/100 Cache is at 1 day, sometimes i get a WU that isn't high priority but active anyway. since i disabled <report_results_immediately> i get much more WUs and so the problem gets worse 
|
|
55)
Message boards :
Number crunching :
zslip 0.19d gpu and seti
(Message 15590)
Posted 16 Mar 2009 by MontagsMeeting Post: read something about 0.19e yesterday but couldn't find it. It works great, now i can run mw all the time. Now there are "only" the "thousands" of processes left as small problem. isn't there a way to limit the number of running processes? They use hundreds of megabytes but should below 50. While updating to 0.19e i noticed a folder "sym" in the project folder. it contains a lot of other folders with a extension of *.pdb. On my both systems were completely different folders. They all contain folders that seem to be named with a hash of something and most of them are empty. example: \sym\lpk.pdb\595ABFF66E154C36972782446C646DF72\ \sym\UxTheme.pdb\957EDB1FC41740F2B701D7E762173EB32\ Hmm seems the actual problems aren't related to seti anymore, so i should "complain" ;) in the zslip thread. thanks for all the help MilkyWay is great Holger |
|
56)
Message boards :
Number crunching :
zslip 0.19d gpu and seti
(Message 15494)
Posted 15 Mar 2009 by MontagsMeeting Post: This is a somewhat problematic solution as seti is my favourite project and the periodic server problems they have, made me set the work buffer higher. When i fill my seti cache, say once a week with ~10 days and then set back work buffer to 1 day, will mw do good with this? And i still have problems with a choppy windows. tried a lot with the <cmdline> settings but couldn't get a acceptable setting. The only thing i noticed that regardless of high priority or not all downloaded WUs were started at once, staying in memory, waiting. I stopped the client from reporting every single task. With this i managed to have always enough WUs but this makes the above situation worse. I'm really confused, every setting i change does good at one end, but things get worse somewhere else and i can't get the correlation. Does nobody have a working set of options - ok, i know it's not that easy :( greets Holger P.S. I have 47 mw processes each using 15MB of physical memory. i think the solution to this problem could be the solution to other problems too. so this is the high priority task for me ;) |
|
57)
Message boards :
Number crunching :
zslip 0.19d gpu and seti
(Message 15201)
Posted 13 Mar 2009 by MontagsMeeting Post: Ok the problem with the many processes comes from high priority WUs, they seem to get started as soon they were downloaded regardless if they could really processed. only the number of WUs set by the <cmdline> n# option were really processed the others stay in memory waiting for processing. |
|
58)
Message boards :
Number crunching :
zslip 0.19d gpu and seti
(Message 15199)
Posted 13 Mar 2009 by MontagsMeeting Post: Ok, i tried various options, but couldn't get it to work as expected. One weird thing i observed is that i have 2-3x more mw-tasks running. I reduced "Number of concurrent WUs" to 1 but still had 2 processes of astronomy_0.19_ATI_x64d.exe running (only one using some cpu). Now i'm back to 2 concurrent WUs and actually have 7 astronomy_0.19_ATI_x64d.exe processes. I could reduce cpu-usage to 1 core by setting <avg_ncpus> in app_info.xml to 0.25 and so i set it to 0.1 and finally all 8 cores do seti again with mw running on the gpu. Now only 1 problem left: My System is very sluggish. Tried the <cmdline> setting w#.##. I set it to w1.1 and w1.1 with the effect of even lower cpu usage (most of the time 0.2%-0.4% instead of 0.5%-1%) but still sluggish setting it to w1.2 made it a little better and with w1.3 i had the feeling it gets worse again. total cpu-time dropped to ~1s - very nice to seti :) Alert On now i have 32 astronomy_0.19_ATI_x64d.exe processes and seti stopped on one core again - setting <avg_ncpus> to 0.01 Alert Off Holger |
|
59)
Message boards :
Number crunching :
zslip 0.19d gpu and seti
(Message 15079)
Posted 12 Mar 2009 by MontagsMeeting Post: Yes, that's not what i'm looking for. I got cpu seti and gpu mw running. My problem is coexistence of seti and mw. RAC isn't my point too. i think i'll wait doing mw until it can coexist with seti, as seti is my thingy :) thanks Holger |
|
60)
Message boards :
Number crunching :
zslip 0.19d gpu and seti
(Message 15048)
Posted 12 Mar 2009 by MontagsMeeting Post: Hello I want to run Milkyway along with seti. My main project is seti, it should run on all cpus. as my PCs have ATI 4850 i thought i could let them do some Milkyway GPU. I tried to install the zslip opti app and nothing worked. i had boinc 6.4.7 running as service on vista controlled with remote desktop - arrgh, worst case i think :) installed vnc, updated boinc to 6.6.15 as normal app. After a few hours i now have a working seti/milkyway setup with only one problem left. while milkyway is running with 2 threads on the gpu 2 of my 8 cores suspend seti. the 2 gpu threads use less than 1% cpu. Is there a chance to use all 8 cores for seti and the gpu for milkyway? And how :) greets Holger |
Previous 20
©2024 Astroinformatics Group