Development for LHC@home

Thank you, good job from you and the team!

Is there a way to get this app working more efficiently? It uses a lot of non-CPU time, which seems inefficient.

Thanks!

I have a task running one day, with 5:45 min CPU time.
Tullio

Computer-ID 1165 finished second LHCb task
AMD-A10-7850 Win 10pro(x64),
Virtualbox 5.0.16 with ExtensionPack 5.0.18. Boinc 7.6.22

saw this in protocol of the second task:

2016-04-25 17:02:50 (331796): Status Report: CPU Time: '10186.859375'
2016-04-25 17:19:46 (331796): Guest Log: VBoxGuest: VBoxGuestCommonGuestCapsAcquire: pSession(0xffff880074d93610), OR(0x0), NOT(0xffffffff), flags(0x0)
2016-04-25 18:42:40 (331796): Status Report: Job Duration: '129600.000000'
2016-04-25 18:42:40 (331796): Status Report: Elapsed Time: '30023.769165'

Second task finished correct.

Computer-ID 1165 running now 3.5 hours third LHCb task
AMD-A10-7850 Win 10pro(x64)-now 10.586,
Virtualbox 5.0.16 with ExtensionPack 5.0.20. Boinc 7.6.22

Info RDP under F3:
Mem 2.050.972k total, 1.976.500k used
Swap 1.040.572k total, 160k used

Startup time from boinc-task start to dominant process starting up: about 9 min.

There is no job progress indicated anywhere.

2016-05-06 00:08:04 UTC INFO [LHCbConfigureArchitecture] Setting variable CMTCONFIG=x86_64-slc6
2016-05-06 00:08:04 UTC INFO [Pilot] Command LHCbConfigureCPURequirements instantiated from LHCbPilotCommands
2016-05-06 00:08:04 UTC INFO [LHCbConfigureCPURequirements] Executing command dirac-wms-cpu-normalization -U -o /DIRAC/Security/UseServerCertificate=yes -R pilot.cfg pilot.cfg
Estimated CPU power is 4.3 HS06
MJF not available on this node

2016-05-06 00:10:19 UTC INFO [LHCbConfigureCPURequirements] Current normalized

LHCb is now running Task Number six.

Up to now - no problems. Get Cobblestones :-)

This are infos from running.log. Task six

CPU as determined by 'dirac-wms-cpu-normalization' is 4.300000
2016-05-06 00:10:19 UTC INFO [LHCbConfigureCPURequirements] Executing command dirac-wms-get-queue-cpu-time -o /DIRAC/Security/UseServerCertificate=yes pilot.cfg
23255

2016-05-06 00:10:23 UTC INFO [LHCbConfigureCPURequirements] CPUTime left (in seconds) is 23255
2016-05-06 00:10:23 UTC INFO [LHCbConfigureCPURequirements] Queue length (which is also set as CPUTimeLeft) is 99996.500000
2016-05-06 00:10:23 UTC INFO [LHCbConfigureCPURequirements] Executing command dirac-configure -FDMH -o /DIRAC/Security/UseServerCertificate=yes -O pilot.cfg pilot.cfg -o /LocalSite/CPUTimeLeft=99996
Executing: /cvmfs/lhcb.cern.ch/lib/lhcb/DIRAC/DIRAC_v6r14p28/scripts/dirac-configure -FDMH -o /DIRAC/Security/UseServerCertificate=yes -O pilot.cfg pilot.cfg -o /LocalSite/CPUTimeLeft=99996
Checking DIRAC installation at "/cvmfs/lhcb.cern.ch/lib/lhcb/DIRAC/DIRAC_v6r14p28"
Will update the output file pilot.cfg

There is no job progress indicated anywhere.

Look at the left side of the mainpage:

LHCb jobs

There are Infos about work of LHCb.

That is not, what i was talking about.
While the job is running, there is no way to see, if it is making any progress.
Nothing in the running.log after the initial setup.
Something like:
event 1 processed
event 2 processed
.
.
.
Eventually, th job just finishes.

FYI

An LHCb task downloads about 300MB at the beginning.
Near the end of the task, it downloads about another 530MB.???

Upload at the end: abou 7.5MB.

It is unclear to me, why the majority of the data is downloaded at the end.
One would assume, the data to download at the beginning to be processed.

The only thing, i can think of is, that some sort of comparison is taking place near the end.

It seems, that only one job is processed per task (contrary to the other Sub-projects here)

Still no task/job progress displayed (logs or Console)