Difference between revisions of "Staff Priorities - Scott"

From SCECpedia
Jump to navigationJump to search
Line 1: Line 1:
 
== Yesterday ==
 
== Yesterday ==
*Retried s351 post-processing on Kraken, ran successfully, matched
+
*Got ALP run with AWP to work, 75% faster than RWG
*Retried ALP run with AWP on Kraken, continuing to get MPI errors
+
*Looked at memcached errors, added additional logging
*Submitted SGT run with fast velocity, low Q
+
*Submitted run to test memcached logging and stage-out with no zip jobs
*Changed memcached invocation on Kraken to pipe output to log file
+
*Installed current version of ioprof, tried also upgrading Pegasus
  
 
== Today ==
 
== Today ==
*Keep testing MPI issues with ALP run
 
*Look at small (~900) memcached errors
 
 
*Check low Q runs
 
*Check low Q runs
*Queue run with ioprof to get file I/O numbers with new executables
+
*Queue run with ioprof on hpc-scec
 +
*Queue RWG and AWP runs from Rob's specifications from yesterday
 +
*Check on s351 run to test memcached, stage-out
  
 
== Blocked ==
 
== Blocked ==
Line 15: Line 15:
  
 
== Follow-ups ==
 
== Follow-ups ==
*CyberShake calls
+
*No
  
  

Revision as of 17:09, 7 June 2012

Yesterday

  • Got ALP run with AWP to work, 75% faster than RWG
  • Looked at memcached errors, added additional logging
  • Submitted run to test memcached logging and stage-out with no zip jobs
  • Installed current version of ioprof, tried also upgrading Pegasus

Today

  • Check low Q runs
  • Queue run with ioprof on hpc-scec
  • Queue RWG and AWP runs from Rob's specifications from yesterday
  • Check on s351 run to test memcached, stage-out

Blocked

  • No

Follow-ups

  • No


  • Compare En-Jui's SGT results to AWP


Areas of Responsibilities

  • CyberShake 1.4
  • Broadband CyberShake
  • Code migration
  • Create bio, work summary, abstract for EU-US summer school