Seems to run with a much lower memory load on XP and Vista than openMalariaB 6.47 (70MB vs 500MB or more), but that might be an unrealistic comparison because the test tasks I've run so far have all completed within an hour and some 6.47 tasks were taking close to 5 hours.
____________
"The ultimate test of a moral society is the kind of world that it leaves to its children." - Dietrich Bonhoeffer
/ 
Anyone want my 6.47s? They're taking up 745MB physical memory per task... and I have 6 of them. Run them 4 at a time and see how slow your computer reacts.
Can't these test apps come with a warning? ;-)
____________
Jord.
BOINC FAQ Service
/ 
Hi Thyme
Basically openMalariaBeta, openMalariaA and openMalariaB are using the same science application which has been updated on openMalariaBeta and openMalariaA to the latest version 6.49. This update will follow on openMalariaB as well.
On openMalariaBeta/openMalariaA, we are currently creating a new set of runs using other model components than the ones used on openMalariaB. The currently used components are less demanding in terms of memory usage and computing time.
Speaking about the components used on openMalariaB, the update to the version 6.49 should reduce the memory usage by roughly a half.
Guillaume
____________
Guillaume Gnaegi
Swiss Tropical and Public Health Institute
http://www.swisstph.ch
/ Starting today, 2010 November 11, I am getting one 6.49 after another that run for about one second and fail with "Computation Error." My machine (runs Linux on a dual 550MHz Pentium III) is concurrently running other BOINC projects, mainly World Community Grid, without problems.
wu_892_515_1_0_128943920 is one that just failed. If I copied it correctly. I cannot copy and paste it.
/ You're missing a library that Malaria used to make the new application with. All your malariaB (6.46)tasks err on:
<stderr_txt>
openMalariaB_6.46_i686-pc-linux-gnu: /lib/tls/libc.so.6: version `GLIBC_2.4' not found (required by openMalariaB_6.46_i686-pc-linux-gnu)
</stderr_txt>
____________
Jord.
BOINC FAQ Service
/ That is too bad. That just started today?
I run CentOS4 on that machine and will not be able to upgrade it for quite a while. Should I sign off for malaria control for six months or so, until CentOS 6 comes out?
/ Volunteer moderator
Project administrator
Project developer
Send message
Joined: Feb 18 09
Posts: 142
Credit: 56,936
RAC: 6
That is too bad. That just started today?
I run CentOS4 on that machine and will not be able to upgrade it for quite a while. Should I sign off for malaria control for six months or so, until CentOS 6 comes out?
Hmm, unfortunately supporting both 5-year-old and recent linux distributions becomes somewhat tricky. I'm not sure why this changed from 6.47, but unfortunately can't promise we'll be able to correct this either.
/ Today I noticed that 9 work units from openmalariaA 6.49 failed within 57 seconds because the elapsed time limit was exceeded. The messages for one of these failed units are below. I use Debian Squeeze on an amd64 machine and I wonder what was going on.
Version 6.46 seems to run just fine, albeit with large overestimates of the remaining time.
za 20 nov 2010 07:39:55 CET malariacontrol.net Starting task wu_971_525_255449_0_1290219598_2 using openMalariaA version 649
za 20 nov 2010 07:40:53 CET malariacontrol.net Aborting task wu_971_525_255449_0_1290219598_2: exceeded elapsed time limit 56.528331
za 20 nov 2010 07:40:54 CET malariacontrol.net Computation for task wu_971_525_255449_0_1290219598_2 finished
za 20 nov 2010 07:40:54 CET malariacontrol.net Output file wu_971_525_255449_0_1290219598_2_0 for task wu_971_525_255449_0_1290219598_2 absent
za 20 nov 2010 07:40:54 CET malariacontrol.net Output file wu_971_525_255449_0_1290219598_2_1 for task wu_971_525_255449_0_1290219598_2 absent
/ I received two 6.49 workunits with an estimated time to completion of 00:00:00 and six with an estimated time to completion of 00:03:57. So far two have ended with a computation error after 00:08:56.
Fri Nov 19 23:48:20 2010 malariacontrol.net Aborting task wu_998_527_252798_0_1289920182_1: exceeded elapsed time limit 556.866995
Fri Nov 19 23:48:21 2010 malariacontrol.net Computation for task wu_998_527_252798_0_1289920182_1 finished
Fri Nov 19 23:48:21 2010 malariacontrol.net Output file wu_998_527_252798_0_1289920182_1_0 for task wu_998_527_252798_0_1289920182_1 absent
Fri Nov 19 23:48:21 2010 malariacontrol.net Output file wu_998_527_252798_0_1289920182_1_1 for task wu_998_527_252798_0_1289920182_1 absent
Same messages received for wu_992_402_252797_0_1289920151_2
edited to add this is on a Mac, Darwin 9.8.0
Best,
Snags
/ 
I just got 300 tasks, each with a 3 day deadline, on a laptop that only uses 1 of its dual cores for crunching because the units each have an estimated zero amount of time to crunch! I will turn on the 2nd core and have set it to no new work but something is DEFINITELY WRONG!!!!
/ something is DEFINITELY WRONG!!!!
Sure is. See my post here.
____________
"The ultimate test of a moral society is the kind of world that it leaves to its children." - Dietrich Bonhoeffer
/ That is OK. I just told that computer to get no new tasks, but I can run on the one that runs RHEL 5.
/ Too much credit?
I just got a huge amount of credit for Workunit 24603264.
62521521 88969 15 Nov 2010 11:51:36 UTC 15 Nov 2010 16:35:57 UTC Completed and validated 9,130.01 8,218.70 39,165.21 openMalaria: A simulator of malaria epidemology and control (Branch A) v6.49
Is 6.49 really that generous, or is something wrong?
/ Just returned the rest of that group of WUs with the same "exceeded elapsed time limit" error except one:
wu_908_31_252792_0_1289920033_2
This one completed successfully in 175.83 cpu seconds (205.13 elapsed time) and received 92.06 credits.
Best,
Snags
/ I got one too:
http://www.malariacontrol.net/workunit.php?wuid=24824700
/ 
I've restricted my crunching to openMalaria: A simulator of malaria epidemology and control (Branch A) v6.49 at the moment. The time limit seems to be 6 minutes, that's quite short, and nearly all of my WUs wanted to take more time and thus didn't make it.
Edith asks:
There is still this disgusting message
Aufgaben in Arbeit suppressed pending completion
in the WU. Why is this? I can't see any reason to restrict this and thus restrict us in looking at possible faults in our WUs in comparison with fellow crunchers.
____________
Grüße vom Sänger
/ 
The validation of Branch A is still totally incorrect.
http://www.malariacontrol.net/result.php?resultid=63004740
Nearly 1h and 6.13cr.
http://www.malariacontrol.net/result.php?resultid=63004726
around 35min and 161.93cr.
I also hope to correct the validation of the first serie of WUs which was 30k per hour or 1/2 hour.
/