Message boards :
Number crunching :
Why is it?
Message board moderation
Author | Message |
---|---|
Send message Joined: 24 Oct 04 Posts: 1129 Credit: 49,762,040 RAC: 5,592 |
When I try to load more days worth of LHC by changing the preferences (as you are supposed to) and then tell it to update my LHC it always goes right to Einstein first and instantly loads a bunch of those and then downloads the normal way the LHC that I was wanting to get?? (long sentence eh?) Good thing I don't run other BOINC projects or it would probably load all of them at once even though all I asked for is LHC. On one machine I do have the version where you can suspend one project but that really doesn't do me any good with this. And of course it takes a lot longer to do Einstein wu's than it does to do LHC no matter how fast your processor is (most of us test them daily) Volunteer Mad Scientist For Life |
Send message Joined: 18 Sep 04 Posts: 59 Credit: 317,857 RAC: 0 |
If I understand correctly, what you're doing is increasing the 'Connect to network about every' x days in your general preference. This tells the scheduler how many days of work to keep in your cache. Unless you change the ratio of shares between your projects, what you will get is an even number of WU-hour from each project to fill your cash. That is to say, with every thing else being equal and a 50-50 share, if you ask for two more days of work and have two projects you will receive one days worth of WU from each projects. If your intention is to load more WU in a specific project, you have to skew the shares in favor of that project as well as increasing your connect time. Questions? Answers are in the BOINC Wiki. Boinc V6.10.56 Recommended WinXP C2D 2.1G 3GB |
Send message Joined: 27 Sep 04 Posts: 6 Credit: 2,790 RAC: 0 |
Under General Preferences, it states:
But LHC seems to be out of work for a while now, so it seems likely that you'll just be downloading Einstein WU's anyway. :) |
Send message Joined: 24 Oct 04 Posts: 1129 Credit: 49,762,040 RAC: 5,592 |
Well I have mine set up (since day one)66.67% LHC and 33.33% Einstein on all of my computers. And when I do this I change the preferences just before I go to get more LHC wu's and do so by clicking LHC and telling it to update and yet it instantly has a batch of Einstein loaded (even via my worlds slowest dialup)and then I watch as the LHC wu's d/l And then I go back a switch the preferences to one day. It seems that it would be better if one of the projects didn't give you work unless you ask for an update or are at least sending some and asking to receive. Btw Red Squirrel......all of mine are loaded and busy working on LHC wu's. Which was my purpose to begin with. Oh and I have a new pet that stops by outside my window eating some of the hundreds of pounds of sunflower seeds I buy each year for all my wild critters. Yes indeed.......it is a little Red Squirrel! In fact his picture is the wallpaper on 2 of my pc's Volunteer Mad Scientist For Life |
Send message Joined: 1 Sep 04 Posts: 275 Credit: 2,652,452 RAC: 0 |
What happens is that the connection to LHC is still using the old preferences. Once that connection is made you download the new preferences. Since you just contacted LHC you can not get work from that project immediately but you have to have work to meet the new queue level so you get it from somewhere else. My advice, assuming you want to provide more support to LHC, is not to raise your queue levels, but raise the resource share. In fact the LHC project wants fast turnaround so you should use the lowest queue setting that you can stand so that you get the LHC work processed and returned as quickly as possible. BOINC WIKI BOINCing since 2002/12/8 |
Send message Joined: 2 Sep 04 Posts: 309 Credit: 715,258 RAC: 0 |
The other way to ensure only LHC gets work is to suspend all other projects then do the updates. Just ensure you keep your connect preference set to be less than half the earliest deadline (which is normally LHC with deadlines of 7 days typical - so preference should be less than 3.5 days). Live long and crunch. Paul (S@H1 8888) BOINC/SAH BETA |
©2024 CERN