Difference between revisions of "Staff Priorities - Scott"

From SCECpedia
Jump to navigationJump to search
Line 1: Line 1:
 
== Yesterday ==
 
== Yesterday ==
*Finished verification 0.5 Hz workflow on Titan, results match
+
*Modified Titan pilots to abort if not enough walltime remaining
*Queued full-scale 0.5 Hz workflow with SeisPSA code with RotD options on Blue Waters; ran into some small issues, resolved
+
*Modified Titan workflow jobs to only match if enough walltime left
*Queued test of random vs uniform hypocenters
+
*Full-scale 0.5Hz workflow with RotD test on Blue Waters verified (few small issues with workflow)
*Queue up 1 Hz SGTs on Titan for PAS, SBSM
+
*Bug in random vs uniform hypocenter test, fixed, resubmitted
 +
*Monitor Titan 1 Hz SGTs for PAS, SBSM
 +
*Generated results from 100m mesh at 0.5 Hz, seismograms suggest a scaling issue
 +
*Found hard-coded values in SGT header generation, emailed Rob
  
 
== Today ==
 
== Today ==
*Modify Titan workflows to abort if not enough walltime remaining
 
 
*Generate estimate of single pilot job size for Titan workflow
 
*Generate estimate of single pilot job size for Titan workflow
*Monitor full-scale 0.5Hz workflow test on Blue Waters
+
*Monitor uniform vs random hypocenter test
*Monitor random vs uniform hypocenter test
 
*Monitor Titan 1 Hz SGTs for PAS, SBSM
 
 
*Start on presentation for HPCC booth
 
*Start on presentation for HPCC booth
 +
*Set up meeting time at SC14 with NCSA staff
 +
*Email Matt Norman to see if he'll be at SC14
 +
*Set up meeting at SC14 with ISI group
 +
*Get feedback from Rob and Kim about SGT header hard-coded values
  
 
== Blocked ==
 
== Blocked ==
Line 17: Line 21:
  
 
== Follow-ups ==
 
== Follow-ups ==
 +
*HighF
 +
*CyberShake
 
*Shakeout Discussion
 
*Shakeout Discussion
  

Revision as of 18:01, 11 November 2014

Yesterday

  • Modified Titan pilots to abort if not enough walltime remaining
  • Modified Titan workflow jobs to only match if enough walltime left
  • Full-scale 0.5Hz workflow with RotD test on Blue Waters verified (few small issues with workflow)
  • Bug in random vs uniform hypocenter test, fixed, resubmitted
  • Monitor Titan 1 Hz SGTs for PAS, SBSM
  • Generated results from 100m mesh at 0.5 Hz, seismograms suggest a scaling issue
  • Found hard-coded values in SGT header generation, emailed Rob

Today

  • Generate estimate of single pilot job size for Titan workflow
  • Monitor uniform vs random hypocenter test
  • Start on presentation for HPCC booth
  • Set up meeting time at SC14 with NCSA staff
  • Email Matt Norman to see if he'll be at SC14
  • Set up meeting at SC14 with ISI group
  • Get feedback from Rob and Kim about SGT header hard-coded values

Blocked

  • No

Follow-ups

  • HighF
  • CyberShake
  • Shakeout Discussion

Areas of Responsibilities

  • CyberShake
  • Broadband CyberShake
  • Code migration