Difference between revisions of "Atlas"

Un article de lcgwiki.
Jump to: navigation, search
Ligne 89: Ligne 89:
 
* Current conclusion
 
* Current conclusion
 
** Enabling direct access creates much more network usage -> Is it usefull for ATLAS ? (processing urgent request faster ?)
 
** Enabling direct access creates much more network usage -> Is it usefull for ATLAS ? (processing urgent request faster ?)
** Production job brokering prefers to send input data (Production_input) to IN2P3-CC_DATADISK instead of IN2P3-LAPP_DATADISK because single file connected is (expected to be better). Can we check if there are more temporary files in IN2P3-CC_DATADISK ?
+
** Production job brokering prefers to send input data (Production_input) to IN2P3-CC_DATADISK instead of IN2P3-LAPP_DATADISK because single file connected is (expected to be better). Can we check if there are more temporary files in IN2P3-CC_DATADISK ?  
 +
*** Request to Panda team to request FTS transfer to read_lan0 instead of the network closest/dastest site
 
** 10 Gb/s connection LAPP-CC (used for all LAPP WAN transfers to any site) can get saturated if huge amount of jobs starting at same time (No smoothing by Panda)
 
** 10 Gb/s connection LAPP-CC (used for all LAPP WAN transfers to any site) can get saturated if huge amount of jobs starting at same time (No smoothing by Panda)
 
** Production_output can only be done to 1 site
 
** Production_output can only be done to 1 site
 +
 +
* Issues
 +
  
 
* Next steps
 
* Next steps

Version du 20:00, 21 septembre 2018

Bienvenue sur la page Atlas LCG-France
Welcome to the LCG-France Atlas page

DOMA_FR project

DOMA_FR tests

  • Global transfers from/ to (same site dest/source site excluded)
Header text LAPP LPSC CC
From

Link

Link

Link

To

Link

Link

Link

  • Data access monitorings as seen by site WN
Destination of access LAPP LPSC CC
Production download

Link

Link

Link

Production upload

Link

Link

Link

Production input

Link

Link

Link

Production output

Link

Link

Link

Analysis Download

Link

Link

Link

Analysis Direct Access

link

Link

Link

  • Current conclusion
    • Enabling direct access creates much more network usage -> Is it usefull for ATLAS ? (processing urgent request faster ?)
    • Production job brokering prefers to send input data (Production_input) to IN2P3-CC_DATADISK instead of IN2P3-LAPP_DATADISK because single file connected is (expected to be better). Can we check if there are more temporary files in IN2P3-CC_DATADISK ?
      • Request to Panda team to request FTS transfer to read_lan0 instead of the network closest/dastest site
    • 10 Gb/s connection LAPP-CC (used for all LAPP WAN transfers to any site) can get saturated if huge amount of jobs starting at same time (No smoothing by Panda)
    • Production_output can only be done to 1 site
  • Issues


  • Next steps
    • Deploy Rucio 1.17 to use protocol priority defined in AGIS (bug in 1.16). Solving bug : use most trusted protocol and would help to control srm decreasing usage
    • Monitor job efficiency vs RTT between SE and WN -> Identify when cache does not have impact (assuming no network bandwidth limitation)
    • Understand the ATLAS job brokering algorithm
    • Get the typical transfer rate per job type