Difference between revisions of "ARC HTCondor Accounting"
From GridPP Wiki
Line 6: | Line 6: | ||
Here ''RalScaling'' is a scaling factor used to normalize the CPU benchmarks. Quattor is configured to do this automatically. | Here ''RalScaling'' is a scaling factor used to normalize the CPU benchmarks. Quattor is configured to do this automatically. | ||
− | The HTCondor configuration on the ARC CEs contains | + | The HTCondor configuration on the ARC CEs contains: |
# CPU benchmark scaling factor | # CPU benchmark scaling factor | ||
MachineRalScaling = "$$([ifThenElse(isUndefined(RalScaling), 1.00, RalScaling)])" | MachineRalScaling = "$$([ifThenElse(isUndefined(RalScaling), 1.00, RalScaling)])" | ||
SUBMIT_EXPRS = $(SUBMIT_EXPRS) MachineRalScaling | SUBMIT_EXPRS = $(SUBMIT_EXPRS) MachineRalScaling | ||
+ | This causes the attribute ''MachineRalScaling'' to be added to job ClassAds, and is available as ''MATCH_EXP_MachineRalScaling''. | ||
We include the following line in the ''[grid-manager]'' section of /etc/arc.conf: | We include the following line in the ''[grid-manager]'' section of /etc/arc.conf: | ||
authplugin="FINISHING timeout=60,onfailure=pass,onsuccess=pass /usr/local/bin/scaling_factors_plugin.py %S %C %I" | authplugin="FINISHING timeout=60,onfailure=pass,onsuccess=pass /usr/local/bin/scaling_factors_plugin.py %S %C %I" | ||
where scaling_factors_plugin.py is available here: https://gist.github.com/alahiff/3a427620e73b42d9282d#file-scaling_factors_plugin-py. | where scaling_factors_plugin.py is available here: https://gist.github.com/alahiff/3a427620e73b42d9282d#file-scaling_factors_plugin-py. |
Revision as of 11:03, 5 June 2014
This page explains how APEL accounting is done at the RAL Tier-1 in order to ensure that CPU benchmarks are taken into account properly.
The HTCondor configuration on the worker nodes includes the following:
RalScaling = 2.14 STARTD_ATTRS = $(STARTD_ATTRS) RalScaling
Here RalScaling is a scaling factor used to normalize the CPU benchmarks. Quattor is configured to do this automatically.
The HTCondor configuration on the ARC CEs contains:
# CPU benchmark scaling factor MachineRalScaling = "$$([ifThenElse(isUndefined(RalScaling), 1.00, RalScaling)])" SUBMIT_EXPRS = $(SUBMIT_EXPRS) MachineRalScaling
This causes the attribute MachineRalScaling to be added to job ClassAds, and is available as MATCH_EXP_MachineRalScaling.
We include the following line in the [grid-manager] section of /etc/arc.conf:
authplugin="FINISHING timeout=60,onfailure=pass,onsuccess=pass /usr/local/bin/scaling_factors_plugin.py %S %C %I"
where scaling_factors_plugin.py is available here: https://gist.github.com/alahiff/3a427620e73b42d9282d#file-scaling_factors_plugin-py.