Tags:
create new tag
view all tags

Arrow left Go to previous page / next page of Tier3 site log MOVED TO...

30. 07. 2012 Dealing with the fallout from the new memory limits

Analyzing maximal memory usage of the last few days based on data recorded by ARCO:

Comparing crab and non-crab jobs

Histogram of maximal memory usage for CRAB based jobs:
mysql> select count(*),round(maxvmem/1000000000,1) as maxmem_GB,name as example_name from view_accounting
      WHERE start_time >= '2012-07-20' AND name LIKE '%crab%' GROUP BY maxmem_GB ORDER BY maxmem_GB;
+----------+-----------+--------------------------------------+
| count(*) | maxmem_GB | example_name                         |
+----------+-----------+--------------------------------------+
|     1872 |       0.0 | naegelic_crab_0_120719_150727_3x2e6k |
|        1 |       0.3 | naegelic_crab_0_120719_150727_3x2e6k |
|        1 |       0.6 | naegelic_crab_0_120726_225537_b86az1 |
|     2751 |       0.9 | naegelic_crab_0_120724_212404_7c56bu |
|      493 |       1.0 | favaro_crab_0_120725_100244_y1b8m2   |
|        1 |       1.1 | meier_f1_crab_0_120717_073931_54bu8y |
|     1289 |       1.2 | meier_f1_crab_0_120717_073931_54bu8y |
|        9 |       1.3 | meier_f1_crab_0_120720_080552_hs90g1 |
|        2 |       2.2 | meier_f1_crab_0_120717_073931_54bu8y |
|      152 |       3.2 | naegelic_crab_0_120725_210329_2zl6f7 |
|        1 |       3.7 | naegelic_crab_0_120719_150727_3x2e6k |
|      544 |       4.4 | naegelic_crab_0_120719_150727_3x2e6k |
|      126 |       6.3 | naegelic_crab_0_120726_225537_b86az1 |
+----------+-----------+--------------------------------------+

Same for non-crab jobs

mysql> select count(*),round(maxvmem/1000000000,1) as maxmem_GB,name as example_name from view_accounting
    WHERE start_time >= '2012-07-20' AND name NOT LIKE '%crab%' GROUP BY maxmem_GB ORDER BY maxmem_GB;
+----------+-----------+-----------------------------------------------+
| count(*) | maxmem_GB | example_name                                  |
+----------+-----------+-----------------------------------------------+
|     1732 |       0.0 | aCBAF                                         |
|       36 |       0.1 | QLOGIN                                        |
|     2805 |       0.2 | ssdl_9                                        |
|     2425 |       0.3 | ssdl_6                                        |
|      237 |       0.4 | ssdl_7                                        |
|      148 |       0.5 | latino_step3_070_WZFull.root                  |
|       98 |       0.6 | Gemschi                                       |
|       19 |       0.7 | MultiJet-Run2012A-PromptReco-v1-AOD           |
|        7 |       0.8 | upst198609                                    |
|       10 |       0.9 | DoubleElectron-Run2012A-PromptReco-v1-AOD     |
|       76 |       1.0 | QCD_Pt-20_MuEnrichedPt-15_TuneZ2_7TeV_pythia6 |
|       34 |       1.1 | job_scan.sh                                   |
|      890 |       1.2 | KV                                            |
|       35 |       1.3 | aCBAF                                         |
|      244 |       1.4 | aCBAF                                         |
|     1026 |       1.5 | brillante                                     |
|      207 |       1.6 | aCBAF                                         |
|      766 |       1.7 | aCBAF                                         |
|      923 |       1.8 | aCBAF                                         |
|      430 |       1.9 | aCBAF                                         |
|      242 |       2.0 | snoek_H300_A20_whatever_patlv2_v2_01bx3t      |
|      128 |       2.1 | bmm-v12-Run2012C-PromptReco-v1_2-0040.csh     |
|       51 |       2.2 | RMG40electron_pt_5_100_eta_0_1p5              |
|        4 |       2.3 | bmm-v11-Run2011B-PromptReco-v1-0142.csh       |
|       50 |       2.6 | RMG17electron_pt_5_100_eta_0_1p5              |
|       52 |       2.7 | snoek_H300_A20_whatever_patlv2_v2_01bx3t      |
|       68 |       2.8 | snoek_H300_A20_whatever_patlv2_v2_01bx3t      |
|      430 |       3.2 | Schubert                                      |
|        1 |       3.3 | cbmm-v12-Run2012C-PromptReco-v1_2-00.csh      |
|      876 |       6.3 | bmm-v11-Run2011A-PromptReco-v5-0150.csh       |
+----------+-----------+-----------------------------------------------+

Looking at the number of crab jobs below and above a 3GB threshold

Crab jobs below the 3 GB threshold

mysql> select count(*),round(maxvmem/1000000000,1) as maxmem_GB,username,name from view_accounting
   WHERE start_time >= '2012-07-01' AND maxvmem <= 3000000000 AND name LIKE '%crab%' GROUP BY maxmem_GB,username
   ORDER BY maxmem_GB;
+----------+-----------+----------+-----------------------------------------------------------------------------------------------------------------------+
| count(*) | maxmem_GB | username | name                                                                                                                  |
+----------+-----------+----------+-----------------------------------------------------------------------------------------------------------------------+
|     1087 |       0.0 | haweber  | haweber_crab_PYTHIA6_SUSY_Stop400_to_bChi300_to_WLSP100-8TeV-GEN_FASTSIM-PU2012_Summer_inTimeOnly_s53nf7              |
|      244 |       0.0 | leo      | leo_crab_PYTHIA6_SUSY_LM10_sftsht_8TeV_cff_py_GEN_FASTSIM_HLT_PU-V3_wc80j4                                            |
|     1000 |       0.0 | meier_f1 | meier_f1_crab_0_120717_073931_54bu8y                                                                                  |
|      292 |       0.0 | mirena   | mirena_crab_0_120703_180906_98xp6c                                                                                    |
|     1162 |       0.0 | naegelic | naegelic_crab_0_120719_150727_3x2e6k                                                                                  |
|        1 |       0.0 | peruzzi  | qspymc_V02-06-12_multicrab.cfg                                                                                        |
|      982 |       0.0 | snoek    | snoek_crab_step2_20_8ps3j1                                                                                            |
|        1 |       0.1 | haweber  | haweber_crab_PYTHIA6_SUSY_Stop400_to_bChi300_to_WLSP100-8TeV-GEN_FASTSIM-PU2012_Summer_inTimeOnly_cq721a              |
|       17 |       0.3 | naegelic | naegelic_crab_0_120719_140732_4km6e7                                                                                  |
|        3 |       0.4 | naegelic | naegelic_crab_0_120719_140732_4km6e7                                                                                  |
|      183 |       0.5 | snoek    | snoek_crab_cpy_20_v389ih                                                                                              |
|       93 |       0.6 | favaro   | favaro_crab_0_120710_142546_32fjn0                                                                                    |
|        1 |       0.6 | naegelic | naegelic_crab_0_120726_225537_b86az1                                                                                  |
|       39 |       0.6 | snoek    | snoek_crab_cpy_20_v389ih                                                                                              |
|      367 |       0.7 | snoek    | snoek_crab_cpy_20_6ohx37                                                                                              |
|        1 |       0.8 | peruzzi  | qspyGJet_Pt_20_doubleEMEnriched_TuneZ2_7TeV_pythia6_Fall11_PU_S6_START42_V14B_v1_AODSIM_mc_V02-06-12_multicrab.cfg    |
|        7 |       0.9 | favaro   | favaro_crab_0_120725_100244_y1b8m2                                                                                    |
|       29 |       0.9 | mirena   | mirena_crab_0_120703_164846_0i2h4a                                                                                    |
|     2746 |       0.9 | naegelic | naegelic_crab_0_120719_140732_4km6e7                                                                                  |
|       14 |       0.9 | snoek    | snoek_crab_cpy_20_v389ih                                                                                              |
|     1362 |       1.0 | favaro   | favaro_crab_0_120705_124438_4jb8t7                                                                                    |
|      392 |       1.0 | haweber  | haweber_crab_PYTHIA6_SUSY_Stop400_to_bChi300_to_WLSP100-8TeV-GEN_FASTSIM-PU2012_Summer_inTimeOnly_cq721a              |
|      541 |       1.0 | mirena   | mirena_crab_0_120703_164846_0i2h4a                                                                                    |
|        3 |       1.0 | naegelic | naegelic_crab_0_120719_140732_4km6e7                                                                                  |
|        1 |       1.1 | meier_f1 | meier_f1_crab_0_120717_073931_54bu8y                                                                                  |
|       52 |       1.1 | mirena   | mirena_crab_0_120703_164846_0i2h4a                                                                                    |
|        2 |       1.1 | naegelic | naegelic_crab_0_120719_140732_4km6e7                                                                                  |
|     1440 |       1.2 | meier_f1 | meier_f1_crab_0_120717_073931_54bu8y                                                                                  |
|      754 |       1.2 | mirena   | mirena_crab_0_120703_164846_0i2h4a                                                                                    |
|     3070 |       1.3 | haweber  | haweber_crab_PYTHIA6_SUSY_Stop400_to_bChi300_to_WLSP100-8TeV-GEN_FASTSIM-PU2012_Summer_inTimeOnly_cq721a              |
|       11 |       1.3 | meier_f1 | meier_f1_crab_0_120717_073931_54bu8y                                                                                  |
|       87 |       1.3 | mirena   | mirena_crab_0_120703_164846_0i2h4a                                                                                    |
|      142 |       1.4 | haweber  | haweber_crab_PYTHIA6_SUSY_Stop400To_bChi300To_Sl200Nu_LSnu200_LSP100-8TeV-GEN_FASTSIM-PU2012_Summer_inTimeOnly_56y9ik |
|       43 |       1.4 | mirena   | mirena_crab_0_120703_164846_0i2h4a                                                                                    |
|       66 |       1.5 | haweber  | haweber_crab_PYTHIA6_SUSY_Stop400_to_bChi300_to_WLSP100-8TeV-GEN_FASTSIM-PU2012_Summer_inTimeOnly_cq721a              |
|       71 |       1.5 | mirena   | mirena_crab_0_120703_164846_0i2h4a                                                                                    |
|      103 |       1.6 | haweber  | haweber_crab_PYTHIA6_SUSY_Stop400_to_bChi300_to_WLSP100-8TeV-GEN_FASTSIM-PU2012_Summer_inTimeOnly_cq721a              |
|       15 |       1.6 | mirena   | mirena_crab_0_120705_095950_6df30j                                                                                    |
|        1 |       1.6 | snoek    | snoek_crab_step3_50_topup_u97ep4                                                                                      |
|        2 |       1.7 | mirena   | mirena_crab_0_120705_095950_6df30j                                                                                    |
|        1 |       1.8 | naegelic | naegelic_crab_0_120719_140732_4km6e7                                                                                  |
|        2 |       1.8 | snoek    | snoek_crab_step3_50_14y2rg                                                                                            |
|        3 |       1.9 | naegelic | naegelic_crab_0_120719_140732_4km6e7                                                                                  |
|      119 |       1.9 | snoek    | snoek_crab_step3_80_394moh                                                                                            |
|        7 |       2.0 | naegelic | naegelic_crab_0_120719_140732_4km6e7                                                                                  |
|       82 |       2.0 | snoek    | snoek_crab_step3_80_394moh                                                                                            |
|        2 |       2.1 | naegelic | naegelic_crab_0_120719_140732_4km6e7                                                                                  |
|        9 |       2.1 | snoek    | snoek_crab_step3_20_sd1v78                                                                                            |
|        3 |       2.2 | meier_f1 | meier_f1_crab_0_120717_073931_54bu8y                                                                                  |
|       19 |       2.2 | snoek    | snoek_crab_step3_50_14y2rg                                                                                            |
|       16 |       2.3 | snoek    | snoek_crab_step3_50_14y2rg                                                                                            |
|       31 |       2.4 | snoek    | snoek_crab_step3_50_14y2rg                                                                                            |
|        3 |       2.5 | snoek    | snoek_crab_step3_50_14y2rg                                                                                            |
|        4 |       2.6 | snoek    | snoek_crab_step2_v2_5mz7q0                                                                                            |
|      118 |       2.7 | snoek    | snoek_crab_step2_v2_5mz7q0                                                                                            |
|      398 |       2.8 | snoek    | snoek_crab_step2_20_8ps3j1                                                                                            |
|      117 |       2.9 | leo      | leo_crab_PYTHIA6_SUSY_LM7_sftsht_8TeV_cff_py_GEN_FASTSIM_HLT_PU-V3_4r90jh                                             |
|      281 |       2.9 | snoek    | snoek_crab_step2_20_8ps3j1                                                                                            |
|      316 |       3.0 | leo      | leo_crab_PYTHIA6_SUSY_LM7_sftsht_8TeV_cff_py_GEN_FASTSIM_HLT_PU-V3_4r90jh                                             |
|        3 |       3.0 | snoek    | snoek_crab_step2_20_8ps3j1                                                                                            |
+----------+-----------+----------+-----------------------------------------------------------------------------------------------------------------------+
60 rows in set (2.30 sec)

Crab jobs above the 3 GB threshold

mysql> select count(*),round(maxvmem/1000000000,1) as maxmem_GB,username,name from view_accounting
   WHERE start_time >= '2012-07-01' AND maxvmem > 3000000000 AND name LIKE '%crab%' GROUP BY maxmem_GB,username
   ORDER BY maxmem_GB;
+----------+-----------+----------+---------------------------------------------------------------------------+
| count(*) | maxmem_GB | username | name                                                                      |
+----------+-----------+----------+---------------------------------------------------------------------------+
|       44 |       3.0 | leo      | leo_crab_PYTHIA6_SUSY_LM7_sftsht_8TeV_cff_py_GEN_FASTSIM_HLT_PU-V3_4r90jh |
|        1 |       3.1 | leo      | leo_crab_PYTHIA6_SUSY_LM7_sftsht_8TeV_cff_py_GEN_FASTSIM_HLT_PU-V3_4r90jh |
|      152 |       3.2 | naegelic | naegelic_crab_0_120725_210329_2zl6f7                                      |
|        1 |       3.2 | snoek    | snoek_crab_step2_20_second_58p3tb                                         |
|        2 |       3.3 | snoek    | snoek_crab_step2_20_second_58p3tb                                         |
|        1 |       3.7 | naegelic | naegelic_crab_0_120719_150727_3x2e6k                                      |
|        1 |       4.0 | snoek    | snoek_crab_step3_50_14y2rg                                                |
|      616 |       4.4 | naegelic | naegelic_crab_0_120719_150727_3x2e6k                                      |
|      126 |       6.3 | naegelic | naegelic_crab_0_120726_225537_b86az1                                      |
+----------+-----------+----------+---------------------------------------------------------------------------+
9 rows in set (1.79 sec)

-- DerekFeichtinger - 2012-07-30


Arrow left Go to previous page / next page of Tier3 site log MOVED TO...

Topic revision: r1 - 2012-07-30 - DerekFeichtinger
 
This site is powered by the TWiki collaboration platform Powered by Perl This site is powered by the TWiki collaboration platformCopyright © 2008-2024 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback