El Miércoles 22 Jul 2009 21:17:02 Martin escribió:
> >> How can you satisfy all of these?
> >> 1. credits per day in CPU app same for both projects
> >> 2. credits per day in GPU app same for both projects
> >> 3. within each project, same WU should give same credits no matter the
> >> processor us
David Anderson wrote:
> That's an excellent example;
> it shows that no credit system can satisfy all goals simultaneously.
> Any given system will violate some goals -
> the challenge is to balance the levels of violation
> based on the importance of the goals.
>
> Currently this is mostly fuzzy
Unless one was to take the example of successful businesses, and provide
some kind of incentive for fair completion of workunits over a broad system
Simply: bo...@home closed source benchmarking project module.
As some 'major upgrade' like "version 7.0" the bo...@home 'project' would need
to be
That's an excellent example;
it shows that no credit system can satisfy all goals simultaneously.
Any given system will violate some goals -
the challenge is to balance the levels of violation
based on the importance of the goals.
Currently this is mostly fuzzy and subjective;
there's no official
Sounds reasonable to me. Not sure if that is what was intended.
Al Reust wrote:
> Seti Beta - NQueens (running backup project) 6.6.38 Aries
> http://setiathome.berkeley.edu/beta/show_host_detail.php?hostid=22789
> 47 Cuda stuck in "project backoff"
> [s...@home Beta Test] Backing off 2 hr 6 min 3
Seti Beta - NQueens (running backup project) 6.6.38 Aries
http://setiathome.berkeley.edu/beta/show_host_detail.php?hostid=22789
47 Cuda stuck in "project backoff"
[s...@home Beta Test] Backing off 2 hr 6 min 34 sec on upload of
09mr09aa.11273.1299.3.13.117_1_0
I could select one result and click
If there is a bug, it may be in how fast the project-wide delay grows.
I've only had a couple of cycles and it's already 2.7 hours.
If there are unwarranted bug reports, it is because uploads sit at
"upload pending" and there is no indication in the GUI that there is a
project wide delay, or wh
El Miércoles 22 Jul 2009 14:35:37 David Anderson escribió:
> Nicolás Alvarez wrote:
>
> > So it would distribute computing power across projects, which is what we
> > want, isn't it?
>
> Not exactly. We want projects that are doing better or more important
> science (as judged by volunteers) to ge
Nicolás Alvarez wrote:
...
> So it would distribute computing power across projects, which is what we
> want,
> isn't it?
Not exactly. We want projects that are doing better or more important science
(as judged by volunteers) to get more computing power.
The main criterion for choosing a projec
El Miércoles 22 Jul 2009 06:25:11 Oliver Bock escribió:
> "Sending sigkill (ctrl-c) to an application that is currently running a
> kernel on the GPU may not result in a clean shutdown of the process as the
> kernel may continue running for a long time afterwards on the GPU. In such
> cases, a syst
El Miércoles 22 Jul 2009 09:43:52 john.mcl...@sybase.com escribió:
> Rank is rather hopeless as a metric. I dedicate 1% of computing power to
> some projects and I am in the top few. On others I dedicate 33% of
> computing power and I am not in the top thousand.
If you want to get into the top f
The percentage could always be weighted with regard to the number of
participants in the project. So a standing at a larger project would
have more weight than the standing at a smaller project, and that
should take care of most of the issues.
Michael Tughan
Michael's Programming
On 22-Jul-
Benchmarks are the worst, except of course for the alternatives.
Martin wrote:
> The present Boinc run benchmarks are too unrepresentative across
> different hardware to work well enough even within a project. Hence the
> dire inaccuracies when trying to award credit using run time.
___
On Fri, 10 Jul 2009 10:44:33 -0700, you wrote:
>I just checked in the change you describe.
>Actually it was added 4 years ago,
>but was backed out because there were reports that it wasn't working right.
>
>This time I added some -enabled messages,
>so we should be able to track down any problems.
Hi again,
The structure APP_INIT_DATA (still) contains a member called "hostid".
However, the only reference to that member in lib/app_ipc.cpp can be found in
APP_INIT_DATA::copy(), hence it seems that this member is neither written nor
parsed (anymore), right?
As I want to pass a unique host
jitesh dundas wrote:
> Dear Sir/Madam,
>
> I am unable to download the tasks for BIONC projects.
> After I have downloaded the tasks, how do I go ahead and work on them,
> please let me know.
See:
http://boinc.berkeley.edu/download.php
Then "attach" to whatever projects of your choice.
See:
ht
This is the wrong forum for this question. There should be a question and
answer forum on each project. There is some reading that might help at
http://boinc.ssl.berkeley.edu/wiki/BOINC_Help. There are also volunteers
available through email or skype at http://boinc.berkeley.edu/help.php.
jm7
Rank is rather hopeless as a metric. I dedicate 1% of computing power to
some projects and I am in the top few. On others I dedicate 33% of
computing power and I am not in the top thousand.
jm7
zombie67
Dear Sir/Madam,
I am unable to download the tasks for BIONC projects.
After I have downloaded the tasks, how do I go ahead and work on them,
please let me know.
I have another query, but not related to BOINC (Sorry About that.)
I am in process of submitting my extended research paper(area of
zombie67 wrote:
> On Jul 21, 2009, at 1:19 PM, Lynn W. Taylor wrote:
>> I don't see a way out, short of doing exactly what Eric Korpela's
>> script
>> tries to do -- normalize FLOPS credit to that predicted by the
>> (imperfect) benchmarks.
>
> Solution: Give up on cross-project credit parity.
Hi David/Eric,
On Tuesday 21 July 2009 22:42:47 David Anderson wrote:
> > 3) What about signal handling in CUDA apps? How important is it to free
> > allocated device memory when the host app is terminated/killed (SIGKILL
> > five secs. after SIGTERM) by the BOINC client? I found some indications
21 matches
Mail list logo