10-26-2021: As we migrating to SDF, decommissioning old hardware and RHEL6, we will no longer actively update the fair shares in this page - some of the major stakeholders no longer use LSF batch system in large scale.
The Shared (General) Farm consists of several physical clusters that are available to all SLAC users. The cluster hardware was purchased incrementally over several years by various stakeholder groups. Each physical cluster is based on a specific hardware model but all hosts in the farm run 64bit RHEL/CentOS6. The LSF "general" queues feed user's jobs to the shared farm. The stakeholders have associated LSF user groups with a fairshare (scheduling priority) that reflects their cluster investment. This ensures that stakeholders will always get some runtime on the cluster when utilization is high. Users that are not members of stakeholder groups still have the ability to run jobs "for free". There is a superset fairshare group "AllUsers" that includes all SLAC users. A non-stakeholder must compete for priority with all other users running jobs on the shared farm. This may be acceptable for some, but production environments may demand priority scheduling. The free "AllUsers" fairshare is actually subsided by the paying stakeholders. A stakeholder's fairshare value is derived from the compute power they have purchased. A HS06 CPU benchmark is calculated for each cluster server model.
|Cluster||# of hosts||# cores/host||# of cores||# Batch slots||HS06/slot||HS06||Owner/Group||Purchased date||Notes|
|fell||-||-||-||-||-||-||-||2007||IGNORE - obsolete hardware in run-to-fail mode|
|kiso(i)||68||24||1632||1360||9.97||13559.2||ATLAS||9/23/2011||RUN-to-FAIL. CentOS7 HT enabled/ 20 slots per host|
|bullet||77.5||16||1240||1240||16.05||19902||PPA||12/5/2012||RUN-to-FAIL. RHEL6 for MPI use - do not map to fairshare|
|bullet||32.5||16||520||520||16.05||8346||Beamphysics||10/2/2014||RUN-to-FAIL. RHEL6. for MPI use - do not map to fairshare|
|deft(i)||22||24||528||528||14.97||7904||ATLAS||1/1/2016||CentOS7 HT disabled, two purchases at 11/2015 and 1/2016|
|8||36||288||288||21.64||6232||Fermi||3/2018||CentOS7. HT disabled|
|bubble(i)||12||36||432||432||21.64||9348.48||Beamphysics||9/2018||CentOS7.Priority for Beamphysics MPI use - do not map to fairshare|
i) CentOS 7 with Singularity
bullet MPI users
The bulletmpi queues are not associated with these fairshare policies. MPI parallel jobs reserve cores across multiple hosts. This does not work well with fairshares because the fairshare formula takes reserved cores into consideration. Large parallel jobs would never start because a job's dynamic scheduling priority would drop before it reserves all the cores it needs to get started. The bulletmpi queues have a higher queue-level priority than all of the fairshare general queues. For this reason we do not assign general queue fairshares for MPI investments.
The stakeholders subside the AllUsers ("free") fairshare by paying a 15% tax on their shares:
|Group||HS06||HS06 (after tax)|
The 15% tax results in 16086 shares for AllUsers
LSF fairshare user groups
The stakeholders are responsible for distributing their shares among their respective LSF groups. The table below should match the production LSF config:
|All Users||16086||15.00%||Everyone (15% tax)|
Fairshares should be subtracted from stakeholder groups and the associated AllUsers tax when hardware is declared obsolete and put into run-to-fail mode. The fairshare distribution should always reflect stakeholder investment in current production cluster resources.
Chargeback model for sustainable hardware lifecycle
This effort paves the way for a possible chargeback model where Computing Division could lease cluster hardware and charge stakeholders a rate for fairshares.