Projects ending

[OldChap]

I want to bring your attention to a thread over at WCG relating to the charts and graphs mentioned above

http://www.worldcommunitygrid.org/fo...d_thread,25987

I started this here as I thought it may help a few people but it seems that we (and by that I mean all WCG'ers) have been putting some strain on the source's hosting. So, I may edit out the pic at the top and just leave the address in situ.

Open to thoughts and comment on this.

[Otis11]

Does XS do automatic refreshing or only when we actually look at it?

If it's auto we should definately fix that... maybe just put the link to the normal page and a link directly to this main chart?

And it may just be an increase in traffic - I know I've been coming here to look at that chart b/c the WCG site loads slowly for me... So I've looked at it more, and you brought it to the attention of a bunch of crunchers.

[mreuter80]

I want to bring your attention to a thread over at WCG relating to the charts and graphs mentioned above

http://www.worldcommunitygrid.org/fo...d_thread,25987

I started this here as I thought it may help a few people but it seems that we (and by that I mean all WCG'ers) have been putting some strain on the source's hosting. So, I may edit out the pic at the top and just leave the address in situ.

Open to thoughts and comment on this.

They experienced some bandwidth issues in 2009 - don't think they have it now. But if you concerned just put the link in the message.

[Otis11]

Didn't catch that... Agreed.

[Bobsama]

Anyone want to help me accumulate HCMD time before that project finishes? I've got both Core i7's working on it for 16 threads but it's finished in another 2 weeks--about 224 days of work and short of the 730 needed for Sapphire. I even ordered a PCIe x1 wifi adapter, so I can move my PCI adapter to the Q8300.

Alternatively, I might be able to cheat--run WCG in several Linux virtual machines each simulating 8 threads. I've got RAM space for 5 VMs, so one Core i7 possibly crunching 48 WUs at a time.

[haschioz]

i can help with that. just tell me how switch my machines to your account. 10threads should help a little

[Moo]

I can help you out too, like haschioz said just tell me how to change over to your account. Got 18 threads to lend.

[Rob_B]

I can help too. 12 threads here. I just go my sapphire HCMD badge my self.

[D_A]

Post up your public key so people can help out.

[Otis11]

Good news! HFCC just added more work into their queue - now up to 115 days of work.

So badge hunters can concentrate on finishing HCMD and HCC and get the badge for HFCC in a month when those are done.

Edit: Work not added yet... reduced feeder priority to keep a constant supply since the next set of WUs isn't expected to be ready till May.

[Bobsama]

Yikes! I didn't realize there were responses and we're almost out of time!

449538_91c2c6b9a0dbab8981223f624bff9f3f

I'm up to 132 days of work and there's 4 days remaining according to the projection. I ordered 10 days of cache for the i7-950, i7-920, & Q8300. The Q8300 definitely does not have 10 days cached.

[Otis11]

Cache all ya got!

Sounds like it was just too far to go, but with what you've got you'll get 100 days before work is out, then you should be able to get 250 days or so on the cached WUs... That's 480 days of the 720? So you need 240 days at least.

That means at least 24 more threads figuring a 10 day cache... If these guys jump on and you Cache to the max you should (just) make it.

FYI that's his public key guys.

[Bobsama]

Well I've got both Core i7's with Hyperthreading enabled. I've also got the Q8300 running, but its cache is basically for 1 day's work. I've got your Xeon X5675's for another 12 threads--and it looks like the Xeon has at least a 5-day cache. I'm up to 32 threads right now. Anyone who adds would be great.

(730-132=598 hours)/32 threads=18.69 calendar days

It'd take 150 threads to be ensured the Sapphire badge within 4 days from today. 50 threads may get it within 12 days--the problem will be populating the new hosts caches. Damn it--if only I would have looked at this thread again sooner--we'd have had more time!

[D_A]

Forum went weird on me right after I replied to your message. +18 threads here if you didn't count 'em already. I'm already seeing messages complaining about lack of work though it might just be a feeder slowdown.

[D_A]

Hmmm, threw another two cores on and it can't get work. Also starting to see idle cores amongst the rest. They might release some more overnight, we'll see what happens.

[Bobsama]

Pretty much to be expected. I'm half thinking of "adjusting" my core clocks to get less work done per WU.

Anyways--my main (i7-950) & old main (Q8300) have been getting new WUs but it seems that my server (i7-920), well, isn't. The only due dates there are the 21st & 22nd while the mains have WUs for the 25th.

[bluestang]

Yeah, same here. No new WU for HCMD. I'm down to needing only 90 days. I have 14 cores and just changed my cache from 2 to 5 days. Let's hope for the best and we all get our goals.

[Otis11]

Just push the cache in your preferences to 10 days - that way every WU it returns it will request the max number of units. (but turn off allow other work first if you haven't already)

I think you have a serious chance to make it if you do this.

And are you guys really having difficulty getting work??? Just looked at mine and it's got a full queue.

[Bobsama]

And thank you everyone--I'm counting 7 computers and I don't know who to thank for each.

Compute-0-0.local -- 12 cores -- Otis
Slack-Boxen --
Sausageman --
Grumblebum --
Girlz-Desktop --
Boyz-Desktop --
Bollocks --

On a side-note, I was investigating virtual machines but it BOINC was reporting something about an invalid platform. I was going to run an 8-core VM or four but I can't get it working. (Ubuntu 10.04 LTS x64, 4GB partition, 1GB RAM, 8 cores, Intel VT-x.)

[haschioz]

how do i assign my machines to your key?

[Otis11]

On linux (command line) you can simply attach to his key just like you would with yours, but on windows I would look here at post 180.

I haven't done this in windows before, but it seems to work for those guys.

[Bobsama]

Adjusted for mine...

cd C:\Program Files\BOINC
boinccmd --project http://www.worldcommunitygrid.org detach
boinccmd --project_attach http://www.worldcommunitygrid.org 449538_91c2c6b9a0dbab8981223f624bff9f3f
cd C:\Program Files (x86)\BOINC
boinccmd --project http://www.worldcommunitygrid.org detach
boinccmd --project_attach http://www.worldcommunitygrid.org 449538_91c2c6b9a0dbab8981223f624bff9f3f

Stick that in a *.cmd or *.bat and execute it. That should work for any version of Windows.

boinccmd --project_attach www.worldcommunitygrid.com 449538_91c2c6b9a0dbab8981223f624bff9f3f

I think that's the command for Linux but you'll first need to navigate to the BOINC folder.

To verify in Windows at least...

boinccmd --get_project_status

And you'll get something like...

======== Projects ========
1) -----------
   name: World Community Grid
   master URL: http://www.worldcommunitygrid.org/
   user_name: Bobsama
   team_name: XtremeSystems
   resource share: 100.000000
   user_total_credit: 1205135.850283
   user_expavg_credit: 6788.963658
   host_total_credit: 29453.873749
   host_expavg_credit: 2109.108449
   nrpc_failures: 0
   master_fetch_failures: 0
   master fetch pending: no
   scheduler RPC pending: no
   trickle upload pending: no
   attached via Account Manager: no
   ended: no
   suspended via GUI: no
   don't request more work: no
   disk usage: 0.000000
   last RPC: 1329324001.409542
   project files downloaded: 0.000000
GUI URL:
   name: Global Statistics
   description: Summary statistics for World Community Grid
   URL: http://www.worldcommunitygrid.org/stat/viewGlobal.do
GUI URL:
   name: My Grid
   description: Your statistics and settings
   URL: http://www.worldcommunitygrid.org/ms/viewMyMemberPage.do
GUI URL:
   name: Results Status
   description: View the status of your assigned work
   URL: http://www.worldcommunitygrid.org/ms/viewBoincResults.do
GUI URL:
   name: Device Profiles
   description: Update your device settings
   URL: http://www.worldcommunitygrid.org/ms/device/viewProfiles.do
GUI URL:
   name: Research
   description: Learn about the projects hosted at World Community Grid
   URL: http://www.worldcommunitygrid.org/projects_showcase/viewResearch.do
GUI URL:
   name: Forums
   description: Visit the World Community Grid forums
   URL: http://www.worldcommunitygrid.org/forumLogin.do
GUI URL:
   name: Help
   description: Search for help in our help system
   URL: http://www.worldcommunitygrid.org/help/viewHelp.do

C:\Program Files\BOINC>

[Rob_B]

I have my HTPC on your account now. But it looks like there is no more work. Only got 19 tasks. And now see these messages.

World Community Grid	15-2-2012 21:18:42	Scheduler request completed: got 0 new tasks	
World Community Grid	15-2-2012 21:18:42	You used the wrong URL for this project.  When convenient, remove this project, then add http://www.worldcommunitygrid.org/	
World Community Grid	15-2-2012 21:18:42	No work sent	
World Community Grid	15-2-2012 21:18:42	No work is available for Help Cure Muscular Dystrophy - Phase 2	
World Community Grid	15-2-2012 21:18:42	No work available for the applications you have selected.  Please check your settings on the website.

[Bobsama]

Every WU counts--give it a bit of time and hopefully some more tasks will be sent out. I have some "A" WU's in my cache so there'll be "B" workunits and then a tertiary group to follow.

All three of my systems appear to have full caches with Feb 25 deadline WUs and a lot of "high priority" tasks. All three are running at stock turbo clocks.

The alternative is I could start allowing other projects to run (if no work is available, crunch other projects) but then we'd have a lower scheduler priority with HCMD.

[D_A]

I picked up some more overnight, though I still have idle cores. It should get some more when the next lot of units drop.