My Linux BOINC client is doing 4 hours for SETI and 1 hour for CPDN, despite being set up for 200:100 between the two project.

It is behaving like it was 400:100.

Anyone else seen this? I have it happening on two systems. consistently.

I think just about everyone has seen this!

I have 8:1:1 set up for CPDN, SETI and LHC. I get 1:1:1 (or I was when there was some SETI work to do).

There are numerous threads about this here and on LHC and CPDN boards.

I think the BOINC team need to fix this before unleashing yet more projects upon us!



Giskard - the first telepathic robot.

win xp box
seti > 100 , lch> 200
Timing is NOW.It just changed between seti to LHC
It's a bit hard to follow
notice the sequence.

LHC@home - 2004-09-03 23:31:31 - Resuming computation for result v64bbe644s6_8615_1_sixvf_4511_0 using sixtrack version 4.43

LHC@home - 2004-09-04 01:31:31 - Preempting result v64bbe644s6_8615_1_sixvf_4511_0 (quit)

SETI@home - 2004-09-04 01:31:31 - Starting result 04my04aa.2277.10688.390916.110_5 using setiathome version 4.03

LHC@home - 2004-09-04 02:31:31 - Restarting result v64bbe644s6_8615_1_sixvf_4511_0 using sixtrack version 4.43

SETI@home - 2004-09-04 02:31:31 - Preempting result 04my04aa.2277.10688.390916.110_5 (quit)

LHC@home - 2004-09-04 04:31:31 - Preempting result v64bbe644s6_8615_1_sixvf_4511_0 (quit)

SETI@home - 2004-09-04 04:31:31 - Restarting result 04my04aa.2277.10688.390916.110_5 using setiathome version 4.03

LHC@home - 2004-09-04 05:31:31 - Restarting result v64bbe644s6_8615_1_sixvf_4511_0 using sixtrack version 4.43

SETI@home - 2004-09-04 05:31:31 - Preempting result 04my04aa.2277.10688.390916.110_5 (quit)

LHC@home - 2004-09-04 07:31:31 - Preempting result v64bbe644s6_8615_1_sixvf_4511_0 (quit)

SETI@home - 2004-09-04 07:31:31 - Restarting result 04my04aa.2277.10688.390916.110_5 using setiathome version 4.03

LHC@home - 2004-09-04 08:31:31 - Restarting result v64bbe644s6_8615_1_sixvf_4511_0 using sixtrack version 4.43

SETI@home - 2004-09-04 08:31:31 - Preempting result 04my04aa.2277.10688.390916.110_5 (quit)

> win xp box
> seti > 100 , lch> 200

Follow-up

LHC@home - 2004-09-04 10:31:31 - Preempting result v64bbe644s6_8615_1_sixvf_4511_0 (quit)
SETI@home - 2004-09-04 10:31:31 - Restarting result 04my04aa.2277.10688.390916.110_5 using setiathome version 4.03

LHC@home - 2004-09-04 11:31:31 - Restarting result v64bbe644s6_8615_1_sixvf_4511_0 using sixtrack version 4.43
SETI@home - 2004-09-04 11:31:31 - Preempting result 04my04aa.2277.10688.390916.110_5 (quit)

LHC@home - 2004-09-04 13:31:31 - Preempting result v64bbe644s6_8615_1_sixvf_4511_0 (quit)
SETI@home - 2004-09-04 13:31:31 - Restarting result 04my04aa.2277.10688.390916.110_5 using setiathome version 4.03

xxxxxxxxxxxx - 12 min. system maintance, shutdown boinc,reboot system etc.

LHC@home - 2004-09-04 13:43:11 - Deferring computation for result v64bbe644s6_8615_1_sixvf_4511_0
SETI@home - 2004-09-04 13:43:11 - Resuming computation for result 04my04aa.2277.10688.390916.110_5 using setiathome version 4.03

SETI@home - 2004-09-04 14:06:17 - Computation for result 04my04aa.2277.10688.390916.110 finished
LHC@home - 2004-09-04 14:06:17 - Restarting result v64bbe644s6_8615_1_sixvf_4511_0 using sixtrack version 4.43

LHC@home - 2004-09-04 14:31:15 - Computation for result v64bbe644s6_8615_1_sixvf_4511 finished
SETI@home - 2004-09-04 14:31:15 - Starting result 25ap04aa.23553.11410.909646.113_6 using setiathome version 4.03

SETI@home - 2004-09-04 15:31:15 - Preempting result 25ap04aa.23553.11410.909646.113_6 (quit)
LHC@home - 2004-09-04 15:31:15 - Starting result v64bbe6ib017s4_6660_1_sixvf_6564_2 using sixtrack version 4.43

Update on this - BOINC Devs - please take note!

I have one machine set up 8:2:1 CPDN:LHC:SETI. It actually runs closer to 1:1:1 (with SETI having been down so much it's difficult to tell accurately)

I have another machine set up to run 2:1 LHC:SETI (No CPDN) and that runs about 2:1, as you'd expect.

The problem seems to be the big difference in deadlines between SETI and LHC (about two weeks) and CPDN (12 months). The despatcher gives a much higher priority to the shorter deadline WUs, even though they can be completed within their deadline with ease at the specified rate.

My cache is set low so as not to fill up with work that can't be done - I have at most two or three WUs from each project.

BOINC Devs - can the priority be adjusted in line with actual computation performance and deadlines so that the long projects like CPDN don't miss out?



Giskard - the first telepathic robot.