The Condor JobRouter

1
The Condor JobRouter
2
aka schedd on the side
3
Status

• Its in the current development series Condor
  7.1.0, unix (windows soonish)
• Used heavily by CMS physics experiment for
  simulation on Open Science Grid (millions of jobs
  routed)

4
What is job routing?
routed (grid) job
original (vanilla) job
Universe vanilla Executable sim Arguments
seed345 Output stdout.345 Error
stderr.345 ShouldTransferFiles
True WhenToTransferOutput ON_EXIT
Universe grid GridType gt2 GridResource
\ cmsgrid01.hep.wisc.edu/jobmanager-condor
Executable sim Arguments seed345 Output
stdout Error stderr ShouldTransferFiles
True WhenToTransferOutput ON_EXIT
JobRouter
Routing Table Site 1 Site 2
final status
5
Routing is just site-level matchmaking

• With feedback from job queue
• number of jobs currently routed to site X
• number of idle jobs routed to site X
• rate of recent success/failure at site X
• And with power to modify job ad
• change attribute values (e.g. Universe)
• insert new attributes (e.g. GridResource)
• add a portal grid proxy if desired

6
Configuring the Routing Table

• JOB_ROUTER_ENTRIES
• list site ClassAds in configuration file
• JOB_ROUTER_ENTRIES_FILE
• read site ClassAds periodically from a file
• JOB_ROUTER_ENTRIES_CMD
• read periodically from a script
• example query a collector such as Open Science
  Grid Resource Selection Service

7
Syntax

• Read the 7.1 manual.
• Its in the chapter on Grid Computing
• Name Grid Site 1GridResource gt2
  gatekeeperMaxIdleJobs 10FailureRateThresho
  ld 0.01

8
What Types of Input Jobs?

• Vanilla Universe
• Self Contained(everything needed is in file
  transfer list)
• High Throughput(many more jobs than cpus)

9
What Target Grid Types?

• Globus, Condor-C work well
• others untested, but should be fine
• Why only target the grid universe?
• no reason at all
• 7.1.1 now allows any destination universe

10
Grid Gotchas

• Globus gt2
• no exit status from job (reported as 0)
• must explicitly list desired output files

11
JobRouter vs. Glidein

• Glidein - Condor overlays the grid
• job never waits in remote queue
• job runs in its normal universe
• private networks doable, but add to complexity
• need something to submit glideins on demand
• JobRouter
• some jobs wait in remote queue (MaxIdleJobs)
• job must be compatible with target grid semantics
• simple to set up, fully automatic to run