Boinc 4.45

Jim Baize
Jim Baize
Joined: 22 Jan 05
Posts: 116
Credit: 582144
RAC: 0

As you can you are making

Message 14441 in response to message 14440

As you can you are making progress. you are down from 31 to 28 hours, right?

Why is it moving so slow? I don't know, but it is moving. Just because it isn't running as you expect it to run does not make it goofy.

My observation is that when BOINC is micromanged, it does not do what is expected. If left to its own devices it usually does what it's told.

Quote:
I Started Einstein long after Seti & Climate, but it's long term debt is 31 hrs now. I had trouble for about 10 days so there was probably another 30 hours I've made up for. The shares were set to be equal for all projects, Seti & Climate work fine, Einstein doesn't! AND THAT"S NOT GOOFY!!!!!!!!!!!!!!!!!!!!


Jim

Gordon Hartman
Gordon Hartman
Joined: 19 Feb 05
Posts: 34
Credit: 34812635
RAC: 2777

RE: As you can you are

Message 14442 in response to message 14441

Quote:

As you can you are making progress. you are down from 31 to 28 hours, right?

Why is it moving so slow? I don't know, but it is moving. Just because it isn't running as you expect it to run does not make it goofy.

My observation is that when BOINC is micromanged, it does not do what is expected. If left to its own devices it usually does what it's told.

Quote:
I Started Einstein long after Seti & Climate, but it's long term debt is 31 hrs now. I had trouble for about 10 days so there was probably another 30 hours I've made up for. The shares were set to be equal for all projects, Seti & Climate work fine, Einstein doesn't! AND THAT"S NOT GOOFY!!!!!!!!!!!!!!!!!!!!


See I think everyone is missing the point. Boinc 4.25 worked! No user problems! Now all this negative energy to run 4.45! I haven't touched it after everyone said, "leave it alone"! Now I have "expectations"! Bullshit! I just want it too work, it's not, but I will post my findings and I am allowed to have my opinions. IT"S GOOFY!

Heffed
Heffed
Joined: 18 Jan 05
Posts: 257
Credit: 12368
RAC: 0

RE: I Started Einstein long

Message 14443 in response to message 14440

Quote:
I Started Einstein long after Seti & Climate, but it's long term debt is 31 hrs now. I had trouble for about 10 days so there was probably another 30 hours I've made up for. The shares were set to be equal for all projects, Seti & Climate work fine, Einstein doesn't! AND THAT"S NOT GOOFY!!!!!!!!!!!!!!!!!!!!


Hi Gordon. Interesting that you posted two different answers to the same post, but whatever.

So here are some questions. Is BOINC set to "run always", or "run based on preferences"? Do you run other applications that may be taking CPU time away from BOINC? If not, have you scanned for virii/spyware/adware/malware lately? (check your task manager to see if there is anything else competing for CPU cycles)

And no, it's not goofy. BOINC is determining whether or not you need work. It's nothing to do with Einstein. If S@H and CPDN are fine, BOINC is working as it should.

Gordon Hartman
Gordon Hartman
Joined: 19 Feb 05
Posts: 34
Credit: 34812635
RAC: 2777

RE: RE: RE: RE: Hi

Message 14444 in response to message 14438

Quote:
Quote:
Quote:
Quote:

Hi Gordon,

Still the same problem for me as well. Read our disscussion at

http://einsteinathome.org/node/189616

V7


As of Day 3 no work!

7/28/2005 9:45:10 AM|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
7/28/2005 9:45:10 AM|Einstein@Home|Requesting 0 seconds of work, returning 0 results
7/28/2005 9:45:20 AM|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded

7/29/2005 6:05:08 AM|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
7/29/2005 6:05:08 AM|Einstein@Home|Requesting 0 seconds of work, returning 0 results
7/29/2005 6:05:10 AM|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded

7/30/2005 5:06:43 AM|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
7/30/2005 5:06:43 AM|Einstein@Home|Requesting 0 seconds of work, returning 0 results
7/30/2005 5:06:55 AM|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded

7/31/2005 4:15:35 AM|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
7/31/2005 4:15:35 AM|Einstein@Home|Requesting 0 seconds of work, returning 0 results
7/31/2005 4:15:37 AM|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded

gravywavy
gravywavy
Joined: 22 Jan 05
Posts: 392
Credit: 68962
RAC: 0

RE: RE: 112184seconds is

Message 14445 in response to message 14435

Quote:
Quote:

112184seconds is only 31.xx hours. That's not that big of a deal.

Quote:
Well, this is stupid. I change to 4.45 and in long term Einstein is -112,184s, Seti is +57,570, and Climate is +54,613. It will be years be fore I can use Einstein!!!!!!!!!!!!!!!!!!!!!! sOMEBODY REALY GOOFED!

Yes 31 hours, but everything else is goofy in Einstein's calculations, so....................we'll see. In 7 days I should get some work........right!

7 days, A figure of 112184 sec is ~31 hours. This means the client figures that Einstein has had 31 hours more than its fair share. If I remember corrdctly (and if I am not mixing up your figures with another contributor to this thread), you had a 20% work share for Einstein. If so, as a very crude approximation, Einstein will go positive in 5 * 31 hours, = 155 hours or nearly 7 days.

That is not the end of it though. Einstein will still not get work till it is the most positive of the Long term debts. This may or may not take a little longer than the 7 days.

If one of the other projects also goes into EDF mode, Einstein may even be delayed again.

Alternatively, if one of the other projects run out of work, Einstein will get in earlier than 7 days.

Whichever happens, in the long run it doesn't matter: over a timescale of a dozen weeks the load should balance out to your resource share.

And that did not happen with 4.25, that is why the upgrade to 4.45 has been welcomed by many users. Not by everyone, true. There are disadvantages as well as advantages to the changes, but please don't suggest the chenges are daft: they were produced in response to user feedback.

If you don't like the changes of course you must say so: but please, say so in a way that says "For me the new behaviour is unwelcome because..." rather than "ITS GOOFY"

And please don't say that nobody complained at v 4.25. Actually many people *did* complain and the new release shows just how carefully the BOINC devs listened to those complaints. OK: I accept it worked for you - but sorry, you are not everybody. You will get heard sooner if you acknowledge that your view is one amongst many than if you make out it is the only opinion going. In my opinion, that is, of course ;-)

4.45 did not completely fix all previous issues, and it has cetainly raised new issues as well. There are things I don't like about the new version, especially on 2-cpu machines - I've made critical postings on other threads. But I haven't (I hope) thrown insults at the code or about the decision to try to respond to other user's concerns.

~~gravywavy

eberndl
eberndl
Joined: 18 Jan 05
Posts: 43
Credit: 98691
RAC: 0

RE: If so, as a very crude

Message 14446 in response to message 14445

Quote:

If so, as a very crude approximation, Einstein will go positive in 5 * 31 hours, = 155 hours or nearly 7 days.

That is not the end of it though. Einstein will still not get work till it is the most positive of the Long term debts. This may or may not take a little longer than the 7 days.

From what I've seen this certainly is not the case. I have a debt to Predictor of +471K, but have no problems whatsoever in getting work from SETI, Einstein or LHC (The reason for this huge debt is CP: -460K). From what I've seen, once the debt is about -4000, you can start downloading work again from a project.

Since this is what I have seen over the course of about 3 months (87 days on my stats chart), I'm pretty sure it's correct. Perhaps it downloads from the most positive with no work first (once it's above a certain number)... I'm not exactly sure.

John McLeod VII
John McLeod VII
Moderator
Joined: 10 Nov 04
Posts: 547
Credit: 632255
RAC: 0

If all of the projects can

If all of the projects can accept work and be contacted, then work requests will happen in highest LT debt order until either all projects get work or refuse to deliver work, or there is some deadline problem that causes BOINC to stop asking for work.

If there is a project that is in deferral, that project will not have a work request sent no matter what its LT debt is (unless the user presses update which cancels the deferral).

Gordon Hartman
Gordon Hartman
Joined: 19 Feb 05
Posts: 34
Credit: 34812635
RAC: 2777

RE: RE: RE: RE: Quote

Message 14448 in response to message 14444

Quote:
Quote:
Quote:
Quote:
Quote:

Hi Gordon,

Still the same problem for me as well. Read our disscussion at

http://einsteinathome.org/node/189616

V7


As of Day 4 no work!

7/28/2005 9:45:10 AM|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
7/28/2005 9:45:10 AM|Einstein@Home|Requesting 0 seconds of work, returning 0 results
7/28/2005 9:45:20 AM|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded

7/29/2005 6:05:08 AM|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
7/29/2005 6:05:08 AM|Einstein@Home|Requesting 0 seconds of work, returning 0 results
7/29/2005 6:05:10 AM|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded

7/30/2005 5:06:43 AM|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
7/30/2005 5:06:43 AM|Einstein@Home|Requesting 0 seconds of work, returning 0 results
7/30/2005 5:06:55 AM|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded

7/31/2005 4:15:35 AM|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
7/31/2005 4:15:35 AM|Einstein@Home|Requesting 0 seconds of work, returning 0 results
7/31/2005 4:15:37 AM|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded

8/1/2005 6:17:19 AM|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
8/1/2005 6:17:19 AM|Einstein@Home|Requesting 0 seconds of work, returning 0 results
8/1/2005 6:17:28 AM|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded

John McLeod VII
John McLeod VII
Moderator
Joined: 10 Nov 04
Posts: 547
Credit: 632255
RAC: 0

RE: As of Day 4 no

Message 14449 in response to message 14448

Quote:
As of Day 4 no work!


Is the LT debt becoming less negative?

Gordon Hartman
Gordon Hartman
Joined: 19 Feb 05
Posts: 34
Credit: 34812635
RAC: 2777

RE: RE: As of Day 4 no

Message 14450 in response to message 14449

Quote:
Quote:
As of Day 4 no work!

Is the LT debt becoming less negative?

On the XP it is -48,315 or ~13.5 hrs & the W2k is 40,845 or ~11.5 hrs. Plus I complicated things. I added LHC & Protein, so Climate, LHC, Protein, & Einstein are a 50 share (16.67%) and Seti is 100 (33.3%) I wanted to add these project and felt waiting did not matter at this point!

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.