- fisyak's home page
- Posts
- 2024
- December (2)
- November (1)
- October (4)
- September (3)
- August (6)
- July (3)
- June (3)
- March (1)
- February (4)
- January (2)
- 2023
- December (2)
- November (1)
- October (2)
- September (3)
- August (3)
- July (2)
- June (4)
- April (2)
- March (2)
- February (1)
- January (2)
- 2022
- 2021
- December (2)
- October (2)
- September (1)
- August (2)
- June (3)
- May (1)
- April (2)
- March (3)
- February (5)
- January (4)
- 2020
- 2019
- 2018
- 2017
- December (3)
- October (2)
- September (1)
- August (7)
- July (3)
- May (2)
- April (1)
- February (1)
- January (2)
- 2016
- 2015
- November (1)
- October (2)
- September (4)
- August (1)
- July (1)
- June (2)
- May (4)
- April (1)
- March (1)
- February (3)
- January (1)
- 2014
- 2013
- December (4)
- My blog
- Post new blog entry
- All blogs
Comparison cpu usage for reconstruction jobs (2013-2016) with -m32 and -m64 options
Updated on Thu, 2016-05-19 15:40. Originally created by fisyak on 2016-05-19 14:00.
In the table below it is presented comparison cpu usage for reconstruction jobs on RCF nodes with 40 Intel(R) Xeon(R) CPU E5-2660 v2 @ 2.20GHz. The numbers are seconds per event <averaged N events>. The compared compiler option are the following:
- G32 : -m32 -g
- O32: -m32 -O2
- G64: -m64 -g
- O64: -m64 -O2
In average
- O32 is 2.73 times faster than G32,
- G64 is 1.12 times faster than G32,
- O64 is 3.00 time faster that G32, and
- O64 is 1.12 time faster than O32.
Thus option "-m64" gives us 12% speed up with respect to "-m32".
year | chain/data | -g -m32 G32 | -O2 -m32 O32 | -g -m64 G64 | -O2 -m64 O64 | G32/O32 | G32/G64 | G32/O64 | O32/O64 | |
Sti | 2013 | RC.pp500 | 96.12 <100> | 39.28 <100> | 85.38 <100> | 36.17 <100> | 2.447 | 1.126 | 2.657 | 1.086 |
Sti | 2014 | RC.AuAu15 | 11.45 <100> | 4.79 <100> | 10.50 <100> | 4.26 <100> | 2.390 | 1.090 | 2.688 | 1.124 |
Sti | 2014 | RC.AuAu200 | 89.87 <100> | 34.97 <100> | 80.53 <100> | 30.57 <100> | 2.570 | 1.116 | 2.940 | 1.144 |
Sti | 2014 | RC.AuAu200.low | 110.70 <100> | 39.34 <100> | 98.23 <100> | 39.13 <100> | 2.814 | 1.127 | 2.829 | 1.005 |
Sti | 2014 | RC.AuAu200.mid | 134.00 <100> | 52.52 <100> | 124.57 <100> | 50.96 <100> | 2.551 | 1.076 | 2.630 | 1.031 |
Sti | 2014 | RC.He3Au200 | 36.71 <100> | 13.58 <100> | 30.00 <100> | 12.37 <100> | 2.703 | 1.224 | 2.968 | 1.098 |
Sti | 2015 | RC.pp200long | 29.62 <100> | 11.34 <100> | 26.80 <100> | 10.51 <100> | 2.612 | 1.105 | 2.818 | 1.079 |
Sti | 2015 | RC.pp200long.NoHFT | 31.02 <100> | 10.67 <100> | 25.78 <100> | 9.75 <100> | 2.907 | 1.203 | 3.182 | 1.094 |
Sti | 2016 | RC.AuAu200.y2016 | 158.96 <100> | 58.93 <100> | 136.58 <100> | 51.49 <100> | 2.697 | 1.164 | 3.087 | 1.144 |
StiCA | 2013 | RC.pp500 | 96.69 <100> | 37.11 <100> | 81.93 <100> | 32.70 <100> | 2.605 | 1.180 | 2.957 | 1.135 |
StiCA | 2014 | RC.AuAu15 | 10.90 <100> | 4.37 <100> | 10.22 <100> | 4.10 <100> | 2.494 | 1.067 | 2.659 | 1.066 |
StiCA | 2014 | RC.AuAu200 | 87.14 <100> | 33.20 <100> | 77.74 <100> | 26.59 <100> | 2.625 | 1.121 | 3.277 | 1.249 |
StiCA | 2014 | RC.AuAu200.low | 113.68 <100> | 40.03 <100> | 98.91 <100> | 36.66 <100> | 2.840 | 1.149 | 3.101 | 1.092 |
StiCA | 2014 | RC.AuAu200.mid | 133.18 <100> | 48.77 <100> | 123.84 <100> | 46.58 <100> | 2.731 | 1.075 | 2.859 | 1.047 |
StiCA | 2014 | RC.He3Au200 | 31.12 <100> | 12.55 <100> | 29.59 <100> | 11.16 <100> | 2.480 | 1.052 | 2.789 | 1.125 |
StiCA | 2015 | RC.pp200long | 27.97 <100> | 10.67 <100> | 24.98 <100> | 9.34 <100> | 2.621 | 1.120 | 2.995 | 1.142 |
StiCA | 2015 | RC.pp200long.NoHFT | 25.71 <100> | 9.77 <100> | 24.33 <100> | 9.24 <100> | 2.632 | 1.057 | 2.782 | 1.057 |
StiCA | 2016 | RC.AuAu200.y2016 | 162.59 <100> | 55.58 <100> | 140.46 <100> | 51.34 <100> | 2.925 | 1.158 | 3.167 | 1.083 |
StiCAKFV | 2013 | RC.pp500 | 128.69 <100> | 49.19 <100> | 119.19 <100> | 40.78 <100> | 2.616 | 1.080 | 3.156 | 1.206 |
StiCAKFV | 2014 | RC.AuAu15 | 12.72 <100> | 4.67 <100> | 11.00 <100> | 4.31 <100> | 2.724 | 1.156 | 2.951 | 1.084 |
StiCAKFV | 2014 | RC.AuAu200 | 106.48 <100> | 39.64 <100> | 89.54 <100> | 32.84 <100> | 2.686 | 1.189 | 3.242 | 1.207 |
StiCAKFV | 2014 | RC.AuAu200.low | 115.41 <100> | 41.37 <100> | 105.02 <100> | 34.16 <100> | 2.790 | 1.099 | 3.379 | 1.211 |
StiCAKFV | 2014 | RC.AuAu200.mid | 169.99 <100> | 62.03 <100> | 2.740 | |||||
StiCAKFV | 2014 | RC.He3Au200 | 34.42 <100> | 13.56 <100> | 32.99 <100> | 11.61 <100> | 2.538 | 1.043 | 2.965 | 1.168 |
StiCAKFV | 2015 | RC.pp200long | 31.33 <100> | 11.37 <100> | 29.53 <100> | 9.95 <100> | 2.755 | 1.061 | 3.149 | 1.143 |
StiCAKFV | 2015 | RC.pp200long.NoHFT | 30.57 <100> | 11.52 <100> | 28.78 <100> | 10.04 <100> | 2.654 | 1.062 | 3.045 | 1.147 |
StiCAKFV | 2016 | RC.AuAu200.y2016 | 181.20 < 99> | 59.98 <100> | 150.34 <100> | 48.74 <100> | 3.021 | 1.205 | 3.718 | 1.231 |
Average | 2.673 | 1.119 | 3.000 | 1.123 |
»
- fisyak's blog
- Login or register to post comments