» Search forum


Index :: New projects :: New project Volpex@UH


Pages: [1]
zioriga
BAM!ID: 71771
Joined: 2009-07-09
Posts: 11
Credits: 0
World-rank: 0

2011-10-15 19:48:03

http://129.7.248.104/VCP/
This project is esporting stats
[BOINCstats] Willy
 
Forum moderator - Administrator - Developer - Tester - Translator
BAM!ID: 1
Joined: 2006-01-09
Posts: 9461
Credits: 353,172,950
World-rank: 5,088

2011-10-15 20:28:44

Sending password reminders does not work.
Please do not PM, IM or email me for support (they will go unread/ignored). Use the forum for support.
Guest

2011-10-16 12:35:11

Sun 16 Oct 2011 08:32:10 AM EDT http://volpex.cs.uh.edu/VCP/ Master file download succeeded
Sun 16 Oct 2011 08:32:16 AM EDT http://volpex.cs.uh.edu/VCP/ Sending scheduler request: Project initialization.
Sun 16 Oct 2011 08:32:16 AM EDT http://volpex.cs.uh.edu/VCP/ Requesting new tasks
Sun 16 Oct 2011 08:32:18 AM EDT Server for testing Bolpex Scheduler request completed: got 0 new tasks
Sun 16 Oct 2011 08:32:18 AM EDT Server for testing Bolpex You used the wrong URL for this project
Sun 16 Oct 2011 08:32:18 AM EDT Server for testing Bolpex The correct URL is http://129.7.248.104/VCP/
Sun 16 Oct 2011 08:32:18 AM EDT Server for testing Bolpex Using the wrong URL can cause problems in some cases.
Sun 16 Oct 2011 08:32:18 AM EDT Server for testing Bolpex When convenient, detach this project, then reattach to http://129.7.248.104/VCP/


?????????????
Saenger
 
Tester - Translator
BAM!ID: 5
Joined: 2006-01-10
Posts: 1735
Credits: 231,959,497
World-rank: 6,804

2011-10-16 13:18:24

NATE1 wrote:
?????????????


Even with the right URL you will only get this:
Code:
So 16 Okt 2011 15:17:25 CEST Server for testing Bolpex [sched_op_debug] Starting scheduler request
So 16 Okt 2011 15:17:25 CEST Server for testing Bolpex Sending scheduler request: Requested by project.
So 16 Okt 2011 15:17:25 CEST Server for testing Bolpex Requesting new tasks for CPU and GPU
So 16 Okt 2011 15:17:25 CEST Server for testing Bolpex [sched_op_debug] CPU work request: 1.00 seconds; 0.09 CPUs
So 16 Okt 2011 15:17:25 CEST Server for testing Bolpex [sched_op_debug] NVIDIA GPU work request: 1.00 seconds; 0.01 GPUs
So 16 Okt 2011 15:17:27 CEST Server for testing Bolpex Scheduler request completed: got 0 new tasks
So 16 Okt 2011 15:17:27 CEST Server for testing Bolpex [sched_op_debug] Server version 611
So 16 Okt 2011 15:17:27 CEST Server for testing Bolpex Message from server: No work sent
So 16 Okt 2011 15:17:27 CEST Server for testing Bolpex Message from server: Project has no tasks available
So 16 Okt 2011 15:17:27 CEST Server for testing Bolpex Project requested delay of 7 seconds
So 16 Okt 2011 15:17:27 CEST Server for testing Bolpex [sched_op_debug] Deferring communication for 7 sec
So 16 Okt 2011 15:17:27 CEST Server for testing Bolpex [sched_op_debug] Reason: requested by project
Grüße vom Sänger
Saenger
 
Tester - Translator
BAM!ID: 5
Joined: 2006-01-10
Posts: 1735
Credits: 231,959,497
World-rank: 6,804

2011-10-16 13:22:15

BOINCstats Willy wrote:
Sending password reminders does not work.

I have asked about that.
Grüße vom Sänger
wertperch
 
Tester
BAM!ID: 86245
Joined: 2010-06-08
Posts: 80
Credits: 1,342,967
World-rank: 184,574

2011-11-04 22:07:30

There are a couple of problems with this project, not least of which is that no-one seems to be responding to problems on their discussion fora.
The main issue for me is that the workunits report stupidly long durations (in the tens of days) but absurdly short deadlines (I have had one that appeared already to have passed its deadline). See an example here.


Now I hate wasting time, so I'm curious as to (i) whether others have experienced the same problem, and (ii) whether the workunits finish in reasonable time (and grant reasonable credit)!

I also shoot cancer with a 686 Magnum
Bruce Kennedy
 
Tester - BOINCstats SOFA member
BAM!ID: 64959
Joined: 2009-01-31
Posts: 731
Credits: 1,889,182,306
World-rank: 1,483

2011-11-05 00:27:52

I set this project to no new work a several days ago after trying to give it a fair shot.

My experience was roughly 80 to 90 percent of the tasks end very quickly because of computational errors or did not need.

Of the 10 or 20 percent that completed as valid, the run times were 1 to 3 hours, so the reported long durations are very wrong.

The credits were failry low. Running 6 or 8 cores 24 x 7 the best I could get was 800 credits a day, most days were below 200.

Butteroven
 
BAM!ID: 656
Joined: 2006-05-24
Posts: 23
Credits: 12,410,656
World-rank: 48,468

2011-11-05 06:46:10

From what I've gleaned so far, it seems the app spends the great majority of its time waiting for something to do - which may or may not happen. It happily hogs an entire core to do that, though. Non-CPU-intensive has been suggested on the boards, but there's no telling if the admins have decided it doesn't work that way, or they're thinking about it, or they don't know what they're doing.

I gather they're trying to use BOINC in some kind of massively-parallel way. Offhand, BOINC doesn't seem well-suited. It could turn out to be very interesting yet, but so far they're just wasting absolute scads of processor time for no credit. I'm going to keep testing for a bit (curiosity and all that) but they've got lots to figure out before I'll take it seriously.

STE\/E
Tester
BAM!ID: 57534
Joined: 2008-08-27
Posts: 2078
Credits: 1,915,713,450,792
World-rank: 2

2011-11-05 09:28:50

R Bruce Kennedy wrote:
Running 6 or 8 cores 24 x 7 the best I could get was 800 credits a day, most days were below 200


You will only get 1 Wu Per Computer no matter how many Cores you run. They have talked about sending more than 1 later on but to me that would even be a bigger waste of time by the Project ...
Bruce Kennedy
 
Tester - BOINCstats SOFA member
BAM!ID: 64959
Joined: 2009-01-31
Posts: 731
Credits: 1,889,182,306
World-rank: 1,483

2011-11-05 16:12:59

STE/E wrote:
R Bruce Kennedy wrote:
Running 6 or 8 cores 24 x 7 the best I could get was 800 credits a day, most days were below 200


You will only get 1 Wu Per Computer no matter how many Cores you run. They have talked about sending more than 1 later on but to me that would even be a bigger waste of time by the Project ...


I agree Ste/e about 1 WU per machine, the 6 to 8 cores I had running were on seperate computers. I should have been more clear.
RvP_LaN
 
BAM!ID: 25148
Joined: 2007-05-01
Posts: 25
Credits: 34,842,768
World-rank: 24,016

2011-11-23 14:07:26

wertperch wrote:
There are a couple of problems with this project, not least of which is that no-one seems to be responding to problems on their discussion forum. The main issue for me is that the workunits report stupidly long durations (in the tens of days) but absurdly short deadlines

+1 for your post!
I always feel a little bit alone when it's time to play "the grouchy grumpy" with these "new projects".

These guys are Doctors from Houston University, but they launch a Boinc Project as if it was their first BASIC coding??????
Allo? Being responsible for the project you launch and what that impact on user's computers??? No?!

There are so many successfull Boinc projects now, so many project initiated by universities, so why these guys don't ask ADVICES to their colleagues BEFORE wasting the cpu time of others projects.

And therefore wasting the time of generous donators which also give their valuable and costly electric power to these researches...

I'm sure than the Boinc Team itself has valuable tutorials to help launching a project...

I don't recommand this project until they will be able to respond to "customers" (us!), and until they stop sending stupid formated WUs.
Zdenko Janov
BAM!ID: 112409
Joined: 2011-12-04
Posts: 1
Credits: 1,898,179
World-rank: 153,116

2012-01-23 18:20:25

Idem už spať, čaute
Pages: [1]

Index :: New projects :: New project Volpex@UH
Reason: