Difference: CMSLoadTestLHCONEResults (1 vs. 35)

Revision 352011/06/04 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Changed:
<
<
Sites marked by an asterisk are connected to the LHCONE network.
>
>
Sites marked by an asterisk are connected to the LHCONE network when the test took place.
 
Changed:
<
<
New script can calculate the maximum rate by taking a moving hourly average over 12 x 5 minute bins. Note that 1280 MiB/s corresponds to 10Gbps and 128 MiB/s to 1Gbps. The binning may make it appear that the the maximum transfer rate is higher than the line speed in some cases, due to rounding.
>
>
Our script calculates the maximum rate by taking a moving hourly average over 12 x 5 minute bins. Note that 1280 MiB/s corresponds to 10Gbps and 128 MiB/s to 1Gbps. The binning may make it appear that the the maximum transfer rate is higher than the line speed in some cases, due to rounding.
 
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MiB/s) Latency (h) Quality (%) Comment
T2_US_Wisconsin T2_US_MIT 1000 1305328040 859.1 4.9 97.5%  
Line: 33 to 33
 
T2_RU_RRC_KI T2_IT_Pisa 500 1306996437 50.5 17.9 96.7%  
T2_DE_RWTH T2_US_Wisconsin 1000 1307032216 205.0 4.3 94.5%  
T2_US_Purdue T2_US_MIT 2000 1307057388 330.7 6.5 98.2%  
Changed:
<
<
T2_US_MIT T2_IT_Pisa 1000 1307115239       No files transferred in 3.4h
T2_DE_RWTH T2_UK_London_IC 1000 1307114126 236.4   80.4% 46% complete
T2_ES_IFCA T2_DE_DESY 1000 1307113378 69.9   100.0% 12% complete
>
>
T2_US_MIT T2_IT_Pisa 1000 1307115239       No files transferred in first 12h
T2_DE_RWTH T2_UK_London_IC 1000 1307114126 248.6 6.8 83.8%  
T2_ES_IFCA T2_DE_DESY 1000 1307113378 85.7   100.0% 83% complete
T2_US_Wisconsin T2_IT_Legnaro 1000 1307125203 138.8   98.7% 97% complete
 

Observations:

Revision 342011/06/03 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 23 to 23
 
T2_IT_Pisa T2_US_Wisconsin 1000 1306843798 161.2 28.8 60.4%  
T2_IT_Legnaro T2_US_MIT 1000 1306876968 179.4 5.0 100.0%  
T2_US_MIT T2_DE_RWTH 1000 1306898705 438.7 3.4 92.4%  
Deleted:
<
<
T2_IT_Legnaro T2_IT_Pisa 1000          
 
T2_UK_London_IC T2_FR_GRIF_LLR 1000 1306919225 93.9 18.6 99.0%  
T2_IT_Pisa T2_ES_IFCA 1000 1306923212 80.6 20.1 98.6%  
T2_DE_RWTH T2_US_MIT 1000 1306915439 263.3 5.6 98.7%  
T2_US_MIT T2_DE_DESY 1000 1306958130 89.5 10.5 99.9%  
Deleted:
<
<
T2_FR_GRIF_LLR T2_UK_London_IC 500 1306988286       2 files transferred in 6h
 
T2_DE_DESY T2_US_MIT 1000 1306996242 277.8 6.7 100.0%  
T2_US_Wisconsin T2_DE_RWTH 1000 1306999306 472.9 6.5 93.9%  
T2_US_Purdue T2_IT_Legnaro 1000 1306998314 78.2 21.7 76.4%  
Changed:
<
<
T2_RU_RRC_KI T2_IT_Pisa 1000 1306996437 50.5     50% complete, stalled.
>
>
T2_RU_RRC_KI T2_IT_Pisa 500 1306996437 50.5 17.9 96.7%  
 
T2_DE_RWTH T2_US_Wisconsin 1000 1307032216 205.0 4.3 94.5%  
Changed:
<
<
T2_US_Purdue T2_US_MIT 2000 1307057388 330.7 6.5 98.2% Production transfers from CNAF to Purdue but rate/success is low.
>
>
T2_US_Purdue T2_US_MIT 2000 1307057388 330.7 6.5 98.2%  
T2_US_MIT T2_IT_Pisa 1000 1307115239       No files transferred in 3.4h
T2_DE_RWTH T2_UK_London_IC 1000 1307114126 236.4   80.4% 46% complete
T2_ES_IFCA T2_DE_DESY 1000 1307113378 69.9   100.0% 12% complete
 

Observations:

Revision 332011/06/03 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 31 to 31
 
T2_FR_GRIF_LLR T2_UK_London_IC 500 1306988286       2 files transferred in 6h
T2_DE_DESY T2_US_MIT 1000 1306996242 277.8 6.7 100.0%  
T2_US_Wisconsin T2_DE_RWTH 1000 1306999306 472.9 6.5 93.9%  
Changed:
<
<
T2_US_Purdue T2_IT_Legnaro 1000 1306998314 78.2     96% complete
>
>
T2_US_Purdue T2_IT_Legnaro 1000 1306998314 78.2 21.7 76.4%  
 
T2_RU_RRC_KI T2_IT_Pisa 1000 1306996437 50.5     50% complete, stalled.
T2_DE_RWTH T2_US_Wisconsin 1000 1307032216 205.0 4.3 94.5%  
Changed:
<
<
T2_US_Purdue T2_US_MIT 2000 1307057388       Production transfers from CNAF to Purdue but rate/success is low.
>
>
T2_US_Purdue T2_US_MIT 2000 1307057388 330.7 6.5 98.2% Production transfers from CNAF to Purdue but rate/success is low.
 

Observations:

Revision 322011/06/02 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 31 to 31
 
T2_FR_GRIF_LLR T2_UK_London_IC 500 1306988286       2 files transferred in 6h
T2_DE_DESY T2_US_MIT 1000 1306996242 277.8 6.7 100.0%  
T2_US_Wisconsin T2_DE_RWTH 1000 1306999306 472.9 6.5 93.9%  
Changed:
<
<
T2_US_Purdue T2_IT_Legnaro 1000 1306998314 78.2     93% complete
T2_RU_RRC_KI T2_IT_Pisa 1000 1306996437 50.5     50% complete
>
>
T2_US_Purdue T2_IT_Legnaro 1000 1306998314 78.2     96% complete
T2_RU_RRC_KI T2_IT_Pisa 1000 1306996437 50.5     50% complete, stalled.
 
T2_DE_RWTH T2_US_Wisconsin 1000 1307032216 205.0 4.3 94.5%  
Changed:
<
<
>
>
T2_US_Purdue T2_US_MIT 2000 1307057388       Production transfers from CNAF to Purdue but rate/success is low.
 

Observations:

Revision 312011/06/02 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 31 to 31
 
T2_FR_GRIF_LLR T2_UK_London_IC 500 1306988286       2 files transferred in 6h
T2_DE_DESY T2_US_MIT 1000 1306996242 277.8 6.7 100.0%  
T2_US_Wisconsin T2_DE_RWTH 1000 1306999306 472.9 6.5 93.9%  
Changed:
<
<
T2_US_Purdue T2_IT_Legnaro 1000 1306998314 78.2     55% complete
T2_RU_RRC_KI T2_IT_Pisa 1000 1306996437 50.5     42% complete
T2_DE_RWTH T2_US_Wisconsin 1000 1307032216 160.9     26% complete
>
>
T2_US_Purdue T2_IT_Legnaro 1000 1306998314 78.2     93% complete
T2_RU_RRC_KI T2_IT_Pisa 1000 1306996437 50.5     50% complete
T2_DE_RWTH T2_US_Wisconsin 1000 1307032216 205.0 4.3 94.5%  
 

Observations:

Line: 43 to 43
 
  1. PhEDEx download agent configuration (number of simultaneous file transfers)
  2. Network bandwidth (1-20 Gbps)
Changed:
<
<
When the results above are ordered by destination site, then the limitation becomes evident. For example, MIT can pull data at wire speed from Wisconsin.

Source Destination Max Rate (Gbps)
T2_US_MIT T2_DE_DESY 0.699219
T2_US_MIT T2_DE_RWTH 3.42734
T2_US_Wisconsin T2_DE_RWTH 3.69453
T2_FR_GRIF_LLR T2_DE_RWTH 5.1125
T2_IT_Pisa T2_ES_IFCA 0.629687
T2_UK_London_IC T2_FR_GRIF_LLR 0.733594
T2_DE_RWTH T2_FR_GRIF_LLR 5.00625
T2_US_Purdue T2_IT_Legnaro 0.610938
T2_US_MIT T2_IT_Legnaro 1.06797
T2_IT_Pisa T2_IT_Legnaro 2.36953
T2_RU_RRC_KI T2_IT_Pisa 0.394531
T2_ES_IFCA T2_IT_Pisa 0.910156
T2_FR_GRIF_LLR T2_UK_London_IC 0.015625
T2_FR_GRIF_LLR T2_UK_London_IC 2.49922
T2_IT_Legnaro T2_US_MIT 1.40156
T2_US_Wisconsin T2_US_MIT 10.7766
T2_US_Wisconsin T2_US_MIT 11.368
T2_US_Wisconsin T2_US_MIT 11.368
T2_DE_RWTH T2_US_MIT 2.05703
T2_DE_DESY T2_US_MIT 2.17031
T2_DE_RWTH T2_US_MIT 4.7
T2_US_Wisconsin T2_US_MIT 6.71172
T2_US_Wisconsin T2_US_MIT 8.20625
T2_UK_London_IC T2_US_Purdue 4.11641
T2_IT_Pisa T2_US_Wisconsin 1.25937
>
>
When the results above are ordered by destination site, then the limitation may become evident.
 
Added:
>
>
For example, MIT can pull data at wire speed from Wisconsin but not from Europe, where network speed may be the limiting factor.
 

Action Item List:

Revision 302011/06/02 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 33 to 33
 
T2_US_Wisconsin T2_DE_RWTH 1000 1306999306 472.9 6.5 93.9%  
T2_US_Purdue T2_IT_Legnaro 1000 1306998314 78.2     55% complete
T2_RU_RRC_KI T2_IT_Pisa 1000 1306996437 50.5     42% complete
Changed:
<
<
T2_DE_RWTH T2_US_Wisconsin 1000 1307032216        
>
>
T2_DE_RWTH T2_US_Wisconsin 1000 1307032216 160.9     26% complete
 

Observations:

Changed:
<
<
There are two fundamental limitations to PhEDEx transfer rates, either of which may be the limiting factor:
  1. FTS channel configuration
  2. Network bandwidth
>
>
There are three fundamental limitations to PhEDEx transfer rates, any of which may be the limiting factor in the maximum hourly transfer rate:
  1. FTS channel configuration (number of simultaneous file transfers, streams per transfer, queue length)
  2. PhEDEx download agent configuration (number of simultaneous file transfers)
  3. Network bandwidth (1-20 Gbps)
  When the results above are ordered by destination site, then the limitation becomes evident. For example, MIT can pull data at wire speed from Wisconsin.

Revision 292011/06/02 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 25 to 25
 
T2_US_MIT T2_DE_RWTH 1000 1306898705 438.7 3.4 92.4%  
T2_IT_Legnaro T2_IT_Pisa 1000          
T2_UK_London_IC T2_FR_GRIF_LLR 1000 1306919225 93.9 18.6 99.0%  
Changed:
<
<
T2_IT_Pisa T2_ES_IFCA 1000 1306923212 80.6   98.6% 1 file remaining
>
>
T2_IT_Pisa T2_ES_IFCA 1000 1306923212 80.6 20.1 98.6%  
 
T2_DE_RWTH T2_US_MIT 1000 1306915439 263.3 5.6 98.7%  
T2_US_MIT T2_DE_DESY 1000 1306958130 89.5 10.5 99.9%  
Changed:
<
<
T2_FR_GRIF_LLR T2_UK_London_IC 500 1306988286        
T2_DE_DESY T2_US_MIT 1000 1306996242        
>
>
T2_FR_GRIF_LLR T2_UK_London_IC 500 1306988286       2 files transferred in 6h
T2_DE_DESY T2_US_MIT 1000 1306996242 277.8 6.7 100.0%  
T2_US_Wisconsin T2_DE_RWTH 1000 1306999306 472.9 6.5 93.9%  
T2_US_Purdue T2_IT_Legnaro 1000 1306998314 78.2     55% complete
T2_RU_RRC_KI T2_IT_Pisa 1000 1306996437 50.5     42% complete
T2_DE_RWTH T2_US_Wisconsin 1000 1307032216        
 

Observations:

Changed:
<
<
  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).
>
>
There are two fundamental limitations to PhEDEx transfer rates, either of which may be the limiting factor:
  1. FTS channel configuration
  2. Network bandwidth

When the results above are ordered by destination site, then the limitation becomes evident. For example, MIT can pull data at wire speed from Wisconsin.

Source Destination Max Rate (Gbps)
T2_US_MIT T2_DE_DESY 0.699219
T2_US_MIT T2_DE_RWTH 3.42734
T2_US_Wisconsin T2_DE_RWTH 3.69453
T2_FR_GRIF_LLR T2_DE_RWTH 5.1125
T2_IT_Pisa T2_ES_IFCA 0.629687
T2_UK_London_IC T2_FR_GRIF_LLR 0.733594
T2_DE_RWTH T2_FR_GRIF_LLR 5.00625
T2_US_Purdue T2_IT_Legnaro 0.610938
T2_US_MIT T2_IT_Legnaro 1.06797
T2_IT_Pisa T2_IT_Legnaro 2.36953
T2_RU_RRC_KI T2_IT_Pisa 0.394531
T2_ES_IFCA T2_IT_Pisa 0.910156
T2_FR_GRIF_LLR T2_UK_London_IC 0.015625
T2_FR_GRIF_LLR T2_UK_London_IC 2.49922
T2_IT_Legnaro T2_US_MIT 1.40156
T2_US_Wisconsin T2_US_MIT 10.7766
T2_US_Wisconsin T2_US_MIT 11.368
T2_US_Wisconsin T2_US_MIT 11.368
T2_DE_RWTH T2_US_MIT 2.05703
T2_DE_DESY T2_US_MIT 2.17031
T2_DE_RWTH T2_US_MIT 4.7
T2_US_Wisconsin T2_US_MIT 6.71172
T2_US_Wisconsin T2_US_MIT 8.20625
T2_UK_London_IC T2_US_Purdue 4.11641
T2_IT_Pisa T2_US_Wisconsin 1.25937
 

Action Item List:

  • Developed a PhEDEx data service-based tool to get the metric numbers, especially the transfer latency. Script can be found here.

Revision 282011/06/02 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 16 to 16
 
T2_FR_GRIF_LLR T2_DE_RWTH 1000 1305921458 654.4 3.0 96.7%  
T2_DE_RWTH T2_FR_GRIF_LLR 1000 1305937471 640.8 6.6 100.0%  
T2_DE_RWTH T2_US_MIT 1000 1306351647 601.6 90.5 88.3% After 989 files, all transfers from RWTH fail with [GENERAL_FAILURE] AsyncWait and it appears the proxy at MIT has expired.
Changed:
<
<
T2_IT_Pisa T2_IT_Legnaro 500 1306447372 303.3 80% completed 95.5% Halted.
>
>
T2_IT_Pisa T2_IT_Legnaro 500 1306447372 303.3   95.5% Halted after 80% completed
 
T2_FR_GRIF_LLR T2_UK_London_IC 1000 1306758701 319.9 7.0 84.0%  
T2_ES_IFCA T2_IT_Pisa 1000 1306762972 116.5 19.5 97.6%  
T2_US_MIT T2_IT_Legnaro 1000 1306840432 136.7 9.4 97.9%  
Line: 25 to 25
 
T2_US_MIT T2_DE_RWTH 1000 1306898705 438.7 3.4 92.4%  
T2_IT_Legnaro T2_IT_Pisa 1000          
T2_UK_London_IC T2_FR_GRIF_LLR 1000 1306919225 93.9 18.6 99.0%  
Changed:
<
<
T2_IT_Pisa T2_ES_IFCA 1000 1306923212 80.6     1 file remaining
>
>
T2_IT_Pisa T2_ES_IFCA 1000 1306923212 80.6   98.6% 1 file remaining
 
T2_DE_RWTH T2_US_MIT 1000 1306915439 263.3 5.6 98.7%  
Changed:
<
<
T2_US_MIT T2_DE_DESY 1000 1306958130 82.2     71% completed
>
>
T2_US_MIT T2_DE_DESY 1000 1306958130 89.5 10.5 99.9%  
 
T2_FR_GRIF_LLR T2_UK_London_IC 500 1306988286        
Added:
>
>
T2_DE_DESY T2_US_MIT 1000 1306996242        
 

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).

Revision 272011/06/02 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 24 to 24
 
T2_IT_Legnaro T2_US_MIT 1000 1306876968 179.4 5.0 100.0%  
T2_US_MIT T2_DE_RWTH 1000 1306898705 438.7 3.4 92.4%  
T2_IT_Legnaro T2_IT_Pisa 1000          
Changed:
<
<
T2_UK_London_IC T2_FR_GRIF_LLR 1000 1306919225        
T2_IT_Pisa T2_ES_IFCA 1000 1306923212        
>
>
T2_UK_London_IC T2_FR_GRIF_LLR 1000 1306919225 93.9 18.6 99.0%  
T2_IT_Pisa T2_ES_IFCA 1000 1306923212 80.6     1 file remaining
 
T2_DE_RWTH T2_US_MIT 1000 1306915439 263.3 5.6 98.7%  
Changed:
<
<
T2_US_MIT T2_DE_DESY 1000 1306958130        
>
>
T2_US_MIT T2_DE_DESY 1000 1306958130 82.2     71% completed
T2_FR_GRIF_LLR T2_UK_London_IC 500 1306988286        
 

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).

Revision 262011/06/01 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 27 to 27
 
T2_UK_London_IC T2_FR_GRIF_LLR 1000 1306919225        
T2_IT_Pisa T2_ES_IFCA 1000 1306923212        
T2_DE_RWTH T2_US_MIT 1000 1306915439 263.3 5.6 98.7%  
Changed:
<
<
>
>
T2_US_MIT T2_DE_DESY 1000 1306958130        
 

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).

Revision 252011/06/01 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 20 to 20
 
T2_FR_GRIF_LLR T2_UK_London_IC 1000 1306758701 319.9 7.0 84.0%  
T2_ES_IFCA T2_IT_Pisa 1000 1306762972 116.5 19.5 97.6%  
T2_US_MIT T2_IT_Legnaro 1000 1306840432 136.7 9.4 97.9%  
Changed:
<
<
T2_IT_Pisa T2_US_Wisconsin 1000 1306843798 161.2 96% completed 61.4%  
>
>
T2_IT_Pisa T2_US_Wisconsin 1000 1306843798 161.2 28.8 60.4%  
 
T2_IT_Legnaro T2_US_MIT 1000 1306876968 179.4 5.0 100.0%  
Changed:
<
<
T2_US_MIT T2_DE_RWTH 1000 1306898705        
>
>
T2_US_MIT T2_DE_RWTH 1000 1306898705 438.7 3.4 92.4%  
 
T2_IT_Legnaro T2_IT_Pisa 1000          
Changed:
<
<
T2_UK_London_IC T2_FR_GRIF_LLR 1000          
T2_IT_Pisa T2_ES_IFCA 1000          

>
>
T2_UK_London_IC T2_FR_GRIF_LLR 1000 1306919225        
T2_IT_Pisa T2_ES_IFCA 1000 1306923212        
T2_DE_RWTH T2_US_MIT 1000 1306915439 263.3 5.6 98.7%  
 

Observations:

Revision 242011/06/01 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 16 to 16
 
T2_FR_GRIF_LLR T2_DE_RWTH 1000 1305921458 654.4 3.0 96.7%  
T2_DE_RWTH T2_FR_GRIF_LLR 1000 1305937471 640.8 6.6 100.0%  
T2_DE_RWTH T2_US_MIT 1000 1306351647 601.6 90.5 88.3% After 989 files, all transfers from RWTH fail with [GENERAL_FAILURE] AsyncWait and it appears the proxy at MIT has expired.
Changed:
<
<
T2_IT_Pisa T2_IT_Legnaro 500 1306447372 303.3 80% completed 95.5%  
>
>
T2_IT_Pisa T2_IT_Legnaro 500 1306447372 303.3 80% completed 95.5% Halted.
 
T2_FR_GRIF_LLR T2_UK_London_IC 1000 1306758701 319.9 7.0 84.0%  
T2_ES_IFCA T2_IT_Pisa 1000 1306762972 116.5 19.5 97.6%  
T2_US_MIT T2_IT_Legnaro 1000 1306840432 136.7 9.4 97.9%  
Changed:
<
<
T2_IT_Pisa T2_US_Wisconsin 1000 1306843798 161.2 81% completed 63.4%  
T2_IT_Legnaro T2_US_MIT 1000 1306876968        
>
>
T2_IT_Pisa T2_US_Wisconsin 1000 1306843798 161.2 96% completed 61.4%  
T2_IT_Legnaro T2_US_MIT 1000 1306876968 179.4 5.0 100.0%  
T2_US_MIT T2_DE_RWTH 1000 1306898705        
 
T2_IT_Legnaro T2_IT_Pisa 1000          
T2_UK_London_IC T2_FR_GRIF_LLR 1000          
T2_IT_Pisa T2_ES_IFCA 1000          

Revision 232011/05/31 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Sites marked by an asterisk are connected to the LHCONE network.

Changed:
<
<
New script can calculate the maximum rate by taking a moving hourly average over 12 x 5 minute bins. Note that 1280 MiB/s corresponds to 10Gbps and 128 MiB/s to 1Gbps. The binning may make is appear that the the maximum transfer rate is higher than the line speed in some cases, due to rounding.
>
>
New script can calculate the maximum rate by taking a moving hourly average over 12 x 5 minute bins. Note that 1280 MiB/s corresponds to 10Gbps and 128 MiB/s to 1Gbps. The binning may make it appear that the the maximum transfer rate is higher than the line speed in some cases, due to rounding.
 
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MiB/s) Latency (h) Quality (%) Comment
T2_US_Wisconsin T2_US_MIT 1000 1305328040 859.1 4.9 97.5%  
Line: 16 to 16
 
T2_FR_GRIF_LLR T2_DE_RWTH 1000 1305921458 654.4 3.0 96.7%  
T2_DE_RWTH T2_FR_GRIF_LLR 1000 1305937471 640.8 6.6 100.0%  
T2_DE_RWTH T2_US_MIT 1000 1306351647 601.6 90.5 88.3% After 989 files, all transfers from RWTH fail with [GENERAL_FAILURE] AsyncWait and it appears the proxy at MIT has expired.
Changed:
<
<
T2_IT_Pisa T2_IT_Legnaro 500 1306447372 303.3 Incomplete 95.5%  
>
>
T2_IT_Pisa T2_IT_Legnaro 500 1306447372 303.3 80% completed 95.5%  
 
T2_FR_GRIF_LLR T2_UK_London_IC 1000 1306758701 319.9 7.0 84.0%  
T2_ES_IFCA T2_IT_Pisa 1000 1306762972 116.5 19.5 97.6%  
Changed:
<
<
T2_US_MIT T2_IT_Legnaro 1000 1306840432 133.2 Incomplete 97.6%  
T2_IT_Pisa T2_US_Wisconsin 1000 1306843798 161.2 Incomplete 66.5%  
>
>
T2_US_MIT T2_IT_Legnaro 1000 1306840432 136.7 9.4 97.9%  
T2_IT_Pisa T2_US_Wisconsin 1000 1306843798 161.2 81% completed 63.4%  
T2_IT_Legnaro T2_US_MIT 1000 1306876968        
 
T2_IT_Legnaro T2_IT_Pisa 1000          
T2_UK_London_IC T2_FR_GRIF_LLR 1000          
T2_IT_Pisa T2_ES_IFCA 1000          

Revision 222011/05/31 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Sites marked by an asterisk are connected to the LHCONE network.

Added:
>
>
New script can calculate the maximum rate by taking a moving hourly average over 12 x 5 minute bins. Note that 1280 MiB/s corresponds to 10Gbps and 128 MiB/s to 1Gbps. The binning may make is appear that the the maximum transfer rate is higher than the line speed in some cases, due to rounding.
 
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MiB/s) Latency (h) Quality (%) Comment
Changed:
<
<
T2_US_Wisconsin T2_US_MIT 1000 May 13 23:07:20 UTC 2011 329.6 4.6 97.9%  
T2_US_Wisconsin T2_US_MIT 2000 May 14 03:47:58 UTC 2011 328.9 ~8 96.2%  
T2_US_Wisconsin T2_US_MIT 2000 May 14 18:10:50 UTC 2011 286.5 7.5 ~91% Did not suspend other debug transfers. Throughput lower.
T2_US_Wisconsin T2_US_MIT 2000 May 15 06:27:30 UTC 2011 293.2 ~7 91.9% Did not suspend other debug transfers.
T2_US_Wisconsin T2_US_MIT 2000 1305614014 152.8 Incomplete 53.4% Did not suspend other debug transfers.
T2_UK_London_IC T2_US_Purdue 1000 1305655527 91.7 Incomplete 100.0% Did not suspend significant production transfers. Purdue site went down.
T2_UK_London_IC T2_US_Purdue 561 1305845178 115.9 Incomplete 100.0% 10TB production transfer queue, impossible to suspend.
T2_FR_GRIF_LLR T2_DE_RWTH 1000 1305921458 448.9 3.8 96.7%  
T2_DE_RWTH T2_FR_GRIF_LLR 1000 1305937471 108.3 6.9 100.0%  
T2_DE_RWTH T2_US_MIT 1000 1306351647 261.4 Incomplete 95.7% After 989 files, all transfers from RWTH fail with [GENERAL_FAILURE] AsyncWait and it appears the proxy at MIT has expired.
T2_IT_Pisa T2_IT_Legnaro 500 1306447372 136.9 5.2 96.3%  
T2_FR_GRIF_LLR T2_UK_London_IC 1000 1306758701        
>
>
T2_US_Wisconsin T2_US_MIT 1000 1305328040 859.1 4.9 97.5%  
T2_US_Wisconsin T2_US_MIT 2000 1305344878 1455.1 7.2 96.8%  
T2_US_Wisconsin T2_US_MIT 2000 1305396650 1379.4 12.8 88.8% Did not suspend other debug transfers.
T2_US_Wisconsin T2_US_MIT 2000 1305440850 1455.1 6.5 96.3% Did not suspend other debug transfers.
T2_US_Wisconsin T2_US_MIT 2000 1305614014 1050.4 217.4 62.3% Did not suspend other debug transfers.
T2_UK_London_IC T2_US_Purdue 1000 1305655527 526.9 287.9 94.9% Did not suspend significant production transfers. Purdue site went down.
T2_FR_GRIF_LLR T2_DE_RWTH 1000 1305921458 654.4 3.0 96.7%  
T2_DE_RWTH T2_FR_GRIF_LLR 1000 1305937471 640.8 6.6 100.0%  
T2_DE_RWTH T2_US_MIT 1000 1306351647 601.6 90.5 88.3% After 989 files, all transfers from RWTH fail with [GENERAL_FAILURE] AsyncWait and it appears the proxy at MIT has expired.
T2_IT_Pisa T2_IT_Legnaro 500 1306447372 303.3 Incomplete 95.5%  
T2_FR_GRIF_LLR T2_UK_London_IC 1000 1306758701 319.9 7.0 84.0%  
T2_ES_IFCA T2_IT_Pisa 1000 1306762972 116.5 19.5 97.6%  
T2_US_MIT T2_IT_Legnaro 1000 1306840432 133.2 Incomplete 97.6%  
T2_IT_Pisa T2_US_Wisconsin 1000 1306843798 161.2 Incomplete 66.5%  
T2_IT_Legnaro T2_IT_Pisa 1000          
 
T2_UK_London_IC T2_FR_GRIF_LLR 1000          
Deleted:
<
<
T2_ES_IFCA T2_IT_Pisa 1000 1306762972        
 
T2_IT_Pisa T2_ES_IFCA 1000          
Changed:
<
<
T2_US_MIT T2_IT_Legnaro 1000 1306840432        
T2_IT_Pisa T2_US_Wisconsin 1000 1306843798        
T2_IT_Pisa T2_IT_Legnaro 1000 1306447372        
T2_IT_Legnaro T2_IT_Pisa 1000          
>
>
 

Revision 212011/05/31 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 16 to 16
 
T2_DE_RWTH T2_FR_GRIF_LLR 1000 1305937471 108.3 6.9 100.0%  
T2_DE_RWTH T2_US_MIT 1000 1306351647 261.4 Incomplete 95.7% After 989 files, all transfers from RWTH fail with [GENERAL_FAILURE] AsyncWait and it appears the proxy at MIT has expired.
T2_IT_Pisa T2_IT_Legnaro 500 1306447372 136.9 5.2 96.3%  
Added:
>
>
T2_FR_GRIF_LLR T2_UK_London_IC 1000 1306758701        
T2_UK_London_IC T2_FR_GRIF_LLR 1000          
T2_ES_IFCA T2_IT_Pisa 1000 1306762972        
T2_IT_Pisa T2_ES_IFCA 1000          
T2_US_MIT T2_IT_Legnaro 1000 1306840432        
T2_IT_Pisa T2_US_Wisconsin 1000 1306843798        
T2_IT_Pisa T2_IT_Legnaro 1000 1306447372        
T2_IT_Legnaro T2_IT_Pisa 1000          
 

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).

Revision 202011/05/27 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 15 to 15
 
T2_FR_GRIF_LLR T2_DE_RWTH 1000 1305921458 448.9 3.8 96.7%  
T2_DE_RWTH T2_FR_GRIF_LLR 1000 1305937471 108.3 6.9 100.0%  
T2_DE_RWTH T2_US_MIT 1000 1306351647 261.4 Incomplete 95.7% After 989 files, all transfers from RWTH fail with [GENERAL_FAILURE] AsyncWait and it appears the proxy at MIT has expired.
Added:
>
>
T2_IT_Pisa T2_IT_Legnaro 500 1306447372 136.9 5.2 96.3%  
 

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).

Revision 192011/05/26 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 14 to 14
 
T2_UK_London_IC T2_US_Purdue 561 1305845178 115.9 Incomplete 100.0% 10TB production transfer queue, impossible to suspend.
T2_FR_GRIF_LLR T2_DE_RWTH 1000 1305921458 448.9 3.8 96.7%  
T2_DE_RWTH T2_FR_GRIF_LLR 1000 1305937471 108.3 6.9 100.0%  
Changed:
<
<
T2_DE_RWTH T2_US_MIT 1000 1306351647        
>
>
T2_DE_RWTH T2_US_MIT 1000 1306351647 261.4 Incomplete 95.7% After 989 files, all transfers from RWTH fail with [GENERAL_FAILURE] AsyncWait and it appears the proxy at MIT has expired.
 

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).
Line: 22 to 22
 

Action Item List:

  • Developed a PhEDEx data service-based tool to get the metric numbers, especially the transfer latency. Script can be found here.
Deleted:
<
<

Log

  • 2011/05/17

Wisconsin has storage issues, MIT has download problems, RRC_KI is dead.

Testing IC to Purdue. After 2 hours, the link stopped transferring data.

 -- JamesLetts - 2011/05/13 \ No newline at end of file

Revision 182011/05/25 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 12 to 12
 
T2_US_Wisconsin T2_US_MIT 2000 1305614014 152.8 Incomplete 53.4% Did not suspend other debug transfers.
T2_UK_London_IC T2_US_Purdue 1000 1305655527 91.7 Incomplete 100.0% Did not suspend significant production transfers. Purdue site went down.
T2_UK_London_IC T2_US_Purdue 561 1305845178 115.9 Incomplete 100.0% 10TB production transfer queue, impossible to suspend.
Deleted:
<
<
T2_BE_IIHE T2_DE_DESY 2984 1305920134   In Progress   This is a production transfer, tracking it now. FAIL
 
T2_FR_GRIF_LLR T2_DE_RWTH 1000 1305921458 448.9 3.8 96.7%  
T2_DE_RWTH T2_FR_GRIF_LLR 1000 1305937471 108.3 6.9 100.0%  
Added:
>
>
T2_DE_RWTH T2_US_MIT 1000 1306351647        
 

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).

Revision 172011/05/21 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 14 to 14
 
T2_UK_London_IC T2_US_Purdue 561 1305845178 115.9 Incomplete 100.0% 10TB production transfer queue, impossible to suspend.
T2_BE_IIHE T2_DE_DESY 2984 1305920134   In Progress   This is a production transfer, tracking it now. FAIL
T2_FR_GRIF_LLR T2_DE_RWTH 1000 1305921458 448.9 3.8 96.7%  
Changed:
<
<
T2_DE_RWTH T2_FR_GRIF_LLR 1000 1305937471 In Progress      
>
>
T2_DE_RWTH T2_FR_GRIF_LLR 1000 1305937471 108.3 6.9 100.0%  
 

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).

Revision 162011/05/21 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Sites marked by an asterisk are connected to the LHCONE network.

Changed:
<
<
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MiB/s) Latency (h) Quality (%) End Time Comment
T2_US_Wisconsin T2_US_MIT 1000 May 13 23:07:20 UTC 2011 329.6 4.6 97.9% May 14 03:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 May 14 03:47:58 UTC 2011 328.9 ~8 96.2% May 14 10:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 May 14 18:10:50 UTC 2011 286.5 7.5 ~91% May 15 01:44:11 UTC 2011 Did not suspend other debug transfers. Throughput lower.
T2_US_Wisconsin T2_US_MIT 2000 May 15 06:27:30 UTC 2011 293.2 ~7 91.9% May 15 13:xx:xx UTC 2011 Did not suspend other debug transfers.
T2_US_Wisconsin T2_US_MIT 2000 1305614014 152.8 Incomplete 53.4% Abandoned due to errors. Did not suspend other debug transfers.
T2_UK_London_IC T2_US_Purdue 1000 1305655527 91.7 Incomplete 100.0% Abandoned due to errors. Did not suspend significant production transfers. Purdue went down.
T2_UK_London_IC T2_US_Purdue 561 1305845178 115.9 Incomplete 100.0% Stalled after 321 files. 10TB production transfer queue, impossible to suspend.
T2_BE_IIHE T2_DE_DESY 2984 1305920134   In Progress     This is a production transfer, tracking it now.
T2_FR_GRIF_LLR T2_DE_RWTH 1000 1305921458   In Progress      
>
>
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MiB/s) Latency (h) Quality (%) Comment
T2_US_Wisconsin T2_US_MIT 1000 May 13 23:07:20 UTC 2011 329.6 4.6 97.9%  
T2_US_Wisconsin T2_US_MIT 2000 May 14 03:47:58 UTC 2011 328.9 ~8 96.2%  
T2_US_Wisconsin T2_US_MIT 2000 May 14 18:10:50 UTC 2011 286.5 7.5 ~91% Did not suspend other debug transfers. Throughput lower.
T2_US_Wisconsin T2_US_MIT 2000 May 15 06:27:30 UTC 2011 293.2 ~7 91.9% Did not suspend other debug transfers.
T2_US_Wisconsin T2_US_MIT 2000 1305614014 152.8 Incomplete 53.4% Did not suspend other debug transfers.
T2_UK_London_IC T2_US_Purdue 1000 1305655527 91.7 Incomplete 100.0% Did not suspend significant production transfers. Purdue site went down.
T2_UK_London_IC T2_US_Purdue 561 1305845178 115.9 Incomplete 100.0% 10TB production transfer queue, impossible to suspend.
T2_BE_IIHE T2_DE_DESY 2984 1305920134   In Progress   This is a production transfer, tracking it now. FAIL
T2_FR_GRIF_LLR T2_DE_RWTH 1000 1305921458 448.9 3.8 96.7%  
T2_DE_RWTH T2_FR_GRIF_LLR 1000 1305937471 In Progress      
 

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).

Revision 152011/05/20 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 11 to 11
 
T2_US_Wisconsin T2_US_MIT 2000 May 15 06:27:30 UTC 2011 293.2 ~7 91.9% May 15 13:xx:xx UTC 2011 Did not suspend other debug transfers.
T2_US_Wisconsin T2_US_MIT 2000 1305614014 152.8 Incomplete 53.4% Abandoned due to errors. Did not suspend other debug transfers.
T2_UK_London_IC T2_US_Purdue 1000 1305655527 91.7 Incomplete 100.0% Abandoned due to errors. Did not suspend significant production transfers. Purdue went down.
Changed:
<
<
T2_UK_London_IC T2_US_Purdue 561 1305845178         10TB production transfer queue, impossible to suspend.
>
>
T2_UK_London_IC T2_US_Purdue 561 1305845178 115.9 Incomplete 100.0% Stalled after 321 files. 10TB production transfer queue, impossible to suspend.
T2_BE_IIHE T2_DE_DESY 2984 1305920134   In Progress     This is a production transfer, tracking it now.
T2_FR_GRIF_LLR T2_DE_RWTH 1000 1305921458   In Progress      
 

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).

Revision 142011/05/19 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 11 to 11
 
T2_US_Wisconsin T2_US_MIT 2000 May 15 06:27:30 UTC 2011 293.2 ~7 91.9% May 15 13:xx:xx UTC 2011 Did not suspend other debug transfers.
T2_US_Wisconsin T2_US_MIT 2000 1305614014 152.8 Incomplete 53.4% Abandoned due to errors. Did not suspend other debug transfers.
T2_UK_London_IC T2_US_Purdue 1000 1305655527 91.7 Incomplete 100.0% Abandoned due to errors. Did not suspend significant production transfers. Purdue went down.
Changed:
<
<
T2_UK_London_IC T2_US_Purdue 561 May 19th ... preparing         Did suspend significant production transfers.
>
>
T2_UK_London_IC T2_US_Purdue 561 1305845178         10TB production transfer queue, impossible to suspend.
 

Observations:

Revision 132011/05/19 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Sites marked by an asterisk are connected to the LHCONE network.

Changed:
<
<
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MiB? /s) Latency (h) Quality (%) End Time Comment
>
>
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MiB/s) Latency (h) Quality (%) End Time Comment
 
T2_US_Wisconsin T2_US_MIT 1000 May 13 23:07:20 UTC 2011 329.6 4.6 97.9% May 14 03:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 May 14 03:47:58 UTC 2011 328.9 ~8 96.2% May 14 10:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 May 14 18:10:50 UTC 2011 286.5 7.5 ~91% May 15 01:44:11 UTC 2011 Did not suspend other debug transfers. Throughput lower.
T2_US_Wisconsin T2_US_MIT 2000 May 15 06:27:30 UTC 2011 293.2 ~7 91.9% May 15 13:xx:xx UTC 2011 Did not suspend other debug transfers.
T2_US_Wisconsin T2_US_MIT 2000 1305614014 152.8 Incomplete 53.4% Abandoned due to errors. Did not suspend other debug transfers.
Changed:
<
<
T2_UK_London_IC T2_US_Purdue 1000 1305655527         Did not suspend significant production transfers.
>
>
T2_UK_London_IC T2_US_Purdue 1000 1305655527 91.7 Incomplete 100.0% Abandoned due to errors. Did not suspend significant production transfers. Purdue went down.
T2_UK_London_IC T2_US_Purdue 561 May 19th ... preparing         Did suspend significant production transfers.
 

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).

Revision 122011/05/17 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 19 to 19
 
  • Developed a PhEDEx data service-based tool to get the metric numbers, especially the transfer latency. Script can be found here.

Log

Changed:
<
<
  • 2011/05/17 Wisconsin has storage issues, MIT has download problems, RRC_KI is dead. Testing IC to Purdue
>
>
  • 2011/05/17

Wisconsin has storage issues, MIT has download problems, RRC_KI is dead.

Testing IC to Purdue. After 2 hours, the link stopped transferring data.

 

Revision 112011/05/17 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Sites marked by an asterisk are connected to the LHCONE network.

Changed:
<
<
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MB/s) Latency (h) Quality (%) End Time Comment
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 23:07:20 UTC 2011 329.61 MB/s 4.58 h 97.9% Sat May 14 03:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 03:47:58 UTC 2011 328.87 MB/s ~8h (approx.) 96.2% Sat May 14 10:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 18:10:50 UTC 2011 286.46 MB/s 7.5h ~91% Sun May 15 01:44:11 PDT 2011 Tried without suspending other transfers, results were lower for max rate.
T2_US_Wisconsin T2_US_MIT 2000 Sun May 15 06:27:30 UTC 2011 293.24 MB/s ~7h 91.9% Sun May 15 13:xx:xx UTC 2011 Without suspending other transfers.
T2_US_Wisconsin T2_US_MIT 2000 1305614014         Without suspending other transfers.
>
>
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MiB? /s) Latency (h) Quality (%) End Time Comment
T2_US_Wisconsin T2_US_MIT 1000 May 13 23:07:20 UTC 2011 329.6 4.6 97.9% May 14 03:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 May 14 03:47:58 UTC 2011 328.9 ~8 96.2% May 14 10:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 May 14 18:10:50 UTC 2011 286.5 7.5 ~91% May 15 01:44:11 UTC 2011 Did not suspend other debug transfers. Throughput lower.
T2_US_Wisconsin T2_US_MIT 2000 May 15 06:27:30 UTC 2011 293.2 ~7 91.9% May 15 13:xx:xx UTC 2011 Did not suspend other debug transfers.
T2_US_Wisconsin T2_US_MIT 2000 1305614014 152.8 Incomplete 53.4% Abandoned due to errors. Did not suspend other debug transfers.
T2_UK_London_IC T2_US_Purdue 1000 1305655527         Did not suspend significant production transfers.
 

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).
Line: 18 to 18
 

Action Item List:

  • Developed a PhEDEx data service-based tool to get the metric numbers, especially the transfer latency. Script can be found here.
Added:
>
>

Log

  • 2011/05/17 Wisconsin has storage issues, MIT has download problems, RRC_KI is dead. Testing IC to Purdue

 -- JamesLetts - 2011/05/13

Revision 102011/05/17 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 9 to 9
 
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 03:47:58 UTC 2011 328.87 MB/s ~8h (approx.) 96.2% Sat May 14 10:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 18:10:50 UTC 2011 286.46 MB/s 7.5h ~91% Sun May 15 01:44:11 PDT 2011 Tried without suspending other transfers, results were lower for max rate.
T2_US_Wisconsin T2_US_MIT 2000 Sun May 15 06:27:30 UTC 2011 293.24 MB/s ~7h 91.9% Sun May 15 13:xx:xx UTC 2011 Without suspending other transfers.
Added:
>
>
T2_US_Wisconsin T2_US_MIT 2000 1305614014         Without suspending other transfers.
 

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).

Action Item List:

Changed:
<
<
  • Need a PhEDEx data service-based tool to get the metric numbers, especially the transfer latency.
>
>
  • Developed a PhEDEx data service-based tool to get the metric numbers, especially the transfer latency. Script can be found here.
  -- JamesLetts - 2011/05/13 \ No newline at end of file

Revision 92011/05/17 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Sites marked by an asterisk are connected to the LHCONE network.

SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MB/s) Latency (h) Quality (%) End Time Comment
Changed:
<
<
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 23:07:20 UTC 2011 329.61 MB/s ~4h (approx.) 97.9% Sat May 14 03:xx:xx UTC 2011  
>
>
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 23:07:20 UTC 2011 329.61 MB/s 4.58 h 97.9% Sat May 14 03:xx:xx UTC 2011  
 
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 03:47:58 UTC 2011 328.87 MB/s ~8h (approx.) 96.2% Sat May 14 10:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 18:10:50 UTC 2011 286.46 MB/s 7.5h ~91% Sun May 15 01:44:11 PDT 2011 Tried without suspending other transfers, results were lower for max rate.
T2_US_Wisconsin T2_US_MIT 2000 Sun May 15 06:27:30 UTC 2011 293.24 MB/s ~7h 91.9% Sun May 15 13:xx:xx UTC 2011 Without suspending other transfers.

Revision 82011/05/15 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 7 to 7
 
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MB/s) Latency (h) Quality (%) End Time Comment
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 23:07:20 UTC 2011 329.61 MB/s ~4h (approx.) 97.9% Sat May 14 03:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 03:47:58 UTC 2011 328.87 MB/s ~8h (approx.) 96.2% Sat May 14 10:xx:xx UTC 2011  
Changed:
<
<
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 18:10:50 UTC 2011 286.46 MB/s 7.5h   Sun May 15 01:44:11 PDT 2011 Tried without suspending other transfers, results were lower for max rate.
T2_US_Wisconsin T2_US_MIT 2000 Sun May 15 06:27:30 UTC 2011         Without suspending other transfers.
>
>
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 18:10:50 UTC 2011 286.46 MB/s 7.5h ~91% Sun May 15 01:44:11 PDT 2011 Tried without suspending other transfers, results were lower for max rate.
T2_US_Wisconsin T2_US_MIT 2000 Sun May 15 06:27:30 UTC 2011 293.24 MB/s ~7h 91.9% Sun May 15 13:xx:xx UTC 2011 Without suspending other transfers.
 
Added:
>
>

Observations:

  • There is a measurable difference when suspending other transfers to and from the sites being measured, about 10% of the total rate, some degradation of the transfer quality. Latency seems unaffected. Results are however consistent when comparing multiple runs for the same conditions (other transfers on or off).
 

Action Item List:

Deleted:
<
<
  • Compare transfers with and without suspending other activity on the endpoints. If there is no difference, then we could presumably run these tests without the extra work of suspending injections and subscriptions. The limiting factor may not be the load on the SRM or the PhEDEx box.
 
  • Need a PhEDEx data service-based tool to get the metric numbers, especially the transfer latency.

-- JamesLetts - 2011/05/13 \ No newline at end of file

Revision 72011/05/15 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 8 to 8
 
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 23:07:20 UTC 2011 329.61 MB/s ~4h (approx.) 97.9% Sat May 14 03:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 03:47:58 UTC 2011 328.87 MB/s ~8h (approx.) 96.2% Sat May 14 10:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 18:10:50 UTC 2011 286.46 MB/s 7.5h   Sun May 15 01:44:11 PDT 2011 Tried without suspending other transfers, results were lower for max rate.
Added:
>
>
T2_US_Wisconsin T2_US_MIT 2000 Sun May 15 06:27:30 UTC 2011         Without suspending other transfers.

 
Deleted:
<
<
Sat May 14 18:10:50 UTC 2011
 

Action Item List:

  • Compare transfers with and without suspending other activity on the endpoints. If there is no difference, then we could presumably run these tests without the extra work of suspending injections and subscriptions. The limiting factor may not be the load on the SRM or the PhEDEx box.

Revision 62011/05/15 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 7 to 7
 
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MB/s) Latency (h) Quality (%) End Time Comment
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 23:07:20 UTC 2011 329.61 MB/s ~4h (approx.) 97.9% Sat May 14 03:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 03:47:58 UTC 2011 328.87 MB/s ~8h (approx.) 96.2% Sat May 14 10:xx:xx UTC 2011  
Changed:
<
<
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 18:10:50 UTC 2011 MB/s       Trying without suspending other transfers.
>
>
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 18:10:50 UTC 2011 286.46 MB/s 7.5h   Sun May 15 01:44:11 PDT 2011 Tried without suspending other transfers, results were lower for max rate.
  Sat May 14 18:10:50 UTC 2011

Revision 52011/05/14 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Sites marked by an asterisk are connected to the LHCONE network.

Changed:
<
<
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MB/s) Latency (h) Quality (%) End Time
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 23:07:20 UTC 2011 329.61 MB/s ~4h (approx.) 97.9% Sat May 14 03:xx:xx UTC 2011
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 03:47:58 UTC 2011 328.87 MB/s ~8h (approx.) 96.2% Sat May 14 10:xx:xx UTC 2011
>
>
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MB/s) Latency (h) Quality (%) End Time Comment
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 23:07:20 UTC 2011 329.61 MB/s ~4h (approx.) 97.9% Sat May 14 03:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 03:47:58 UTC 2011 328.87 MB/s ~8h (approx.) 96.2% Sat May 14 10:xx:xx UTC 2011  
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 18:10:50 UTC 2011 MB/s       Trying without suspending other transfers.

Sat May 14 18:10:50 UTC 2011

Action Item List:

  • Compare transfers with and without suspending other activity on the endpoints. If there is no difference, then we could presumably run these tests without the extra work of suspending injections and subscriptions. The limiting factor may not be the load on the SRM or the PhEDEx box.
  • Need a PhEDEx data service-based tool to get the metric numbers, especially the transfer latency.
  -- JamesLetts - 2011/05/13 \ No newline at end of file

Revision 42011/05/14 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Line: 6 to 6
 
SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MB/s) Latency (h) Quality (%) End Time
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 23:07:20 UTC 2011 329.61 MB/s ~4h (approx.) 97.9% Sat May 14 03:xx:xx UTC 2011
Changed:
<
<
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 03:47:58 UTC 2011        
>
>
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 03:47:58 UTC 2011 328.87 MB/s ~8h (approx.) 96.2% Sat May 14 10:xx:xx UTC 2011
  -- JamesLetts - 2011/05/13

Revision 32011/05/14 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Sites marked by an asterisk are connected to the LHCONE network.

SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MB/s) Latency (h) Quality (%) End Time
Changed:
<
<
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 16:07:20 PDT 2011 In Progress      
>
>
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 23:07:20 UTC 2011 329.61 MB/s ~4h (approx.) 97.9% Sat May 14 03:xx:xx UTC 2011
T2_US_Wisconsin T2_US_MIT 2000 Sat May 14 03:47:58 UTC 2011        
  -- JamesLetts - 2011/05/13

Revision 22011/05/13 - Main.JamesLetts

Line: 1 to 1
 
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Sites marked by an asterisk are connected to the LHCONE network.

SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MB/s) Latency (h) Quality (%) End Time
Changed:
<
<
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 13:59:23 PDT 2011        
>
>
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 16:07:20 PDT 2011 In Progress      
  -- JamesLetts - 2011/05/13

Revision 12011/05/13 - Main.JamesLetts

Line: 1 to 1
Added:
>
>
META TOPICPARENT name="CMSLoadTestLHCONE"

Test Results

Sites marked by an asterisk are connected to the LHCONE network.

SOURCE SITE DESTINATION SITE Files Injected Start Time Max. Rate in 1h (MB/s) Latency (h) Quality (%) End Time
T2_US_Wisconsin T2_US_MIT 1000 Fri May 13 13:59:23 PDT 2011        

-- JamesLetts - 2011/05/13

 
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback