Intro

The benchmarks were done using two test files (for mdf version 3 and 4) of around 170MB. The files contain 183 data groups and a total of 36424 channels.

asamdf 2.5.0 was compared against mdfreader 0.2.5 (latest versions from PyPI). mdfreader seems to be the most used Python package to handle MDF files, and it also supports both version 3 and 4 of the standard.

The three benchmark cathegories are file open, file save and extracting the data for all channels inside the file(36424 calls). For each cathegory two aspect were noted: elapsed time and peak RAM usage.

Dependencies

You will need the following packages to be able to run the benchmark script

  • psutil
  • mdfreader

x64 Python results

The test environment used for 64 bit tests had:

  • 3.6.2 (v3.6.2:5fd33b5, Jul 8 2017, 04:57:36) [MSC v.1900 64 bit (AMD64)]
  • Windows-10-10.0.14393-SP0
  • Intel64 Family 6 Model 94 Stepping 3, GenuineIntel
  • 16GB installed RAM

Notations used in the results:

  • nodata = asammdf MDF object created with load_measured_data=False (raw channel data not loaded into RAM)

Files used for benchmark:

  • 183 groups
  • 36424 channels

Raw data

Open file Time [ms] RAM [MB]
asammdf 2.5.0 mdfv3 821 371
asammdf 2.5.0 nodata mdfv3 653 191
mdfreader 0.2.5 mdfv3 2909 537
asammdf 2.5.0 mdfv4 1694 455
asammdf 2.5.0 nodata mdfv4 1297 260
mdfreader 0.2.5 mdfv4 31074 748
Save file Time [ms] RAM [MB]
asammdf 2.5.0 mdfv3 393 373
asammdf 2.5.0 nodata mdfv3 383 198
mdfreader 0.2.5 mdfv3 21464 1997
asammdf 2.5.0 mdfv4 586 465
asammdf 2.5.0 nodata mdfv4 550 275
mdfreader 0.2.5 mdfv4 19036 2795
Get all channels (36424 calls) Time [ms] RAM [MB]
asammdf 2.5.0 mdfv3 613 381
asammdf 2.5.0 nodata mdfv3 9161 207
mdfreader 0.2.5 mdfv3 28 536
asammdf 2.5.0 mdfv4 606 464
asammdf 2.5.0 nodata mdfv4 12403 275
mdfreader 0.2.5 mdfv4 40 748
Convert file Time [ms] RAM [MB]
asammdf 2.5.0 v3 to v4 4773 885
asammdf 2.5.0 v3 to v4 nodata 21903 605
asammdf 2.5.0 v4 to v3 4823 882
asammdf 2.5.0 v4 to v3 nodata 26090 740

x86 Python results

The test environment used for 32 bit tests had:

  • 3.6.1 (v3.6.1:69c0db5, Mar 21 2017, 17:54:52) [MSC v.1900 32 bit (Intel)]
  • Windows-10-10.0.14393-SP0
  • Intel64 Family 6 Model 94 Stepping 3, GenuineIntel
  • 16GB installed RAM

Notations used in the results:

  • nodata = asammdf MDF object created with load_measured_data=False (raw channel data not loaded into RAM)

Files used for benchmark:

  • 183 groups
  • 36424 channels

Raw data

Open file Time [ms] RAM [MB]
asammdf 2.5.0 mdfv3 1009 289
asammdf 2.5.0 nodata mdfv3 663 118
mdfreader 0.2.5 mdfv3 3705 454
asammdf 2.5.0 mdfv4 2031 343
asammdf 2.5.0 nodata mdfv4 1690 161
mdfreader 0.2.5 mdfv4 42315 576
Save file Time [ms] RAM [MB]
asammdf 2.5.0 mdfv3 439 293
asammdf 2.5.0 nodata mdfv3 462 126
mdfreader 0.2.5 mdfv3 19759 1224
asammdf 2.5.0 mdfv4 691 354
asammdf 2.5.0 nodata mdfv4 712 174
mdfreader 0.2.5 mdfv4 17415 1686
Get all channels (36424 calls) Time [ms] RAM [MB]
asammdf 2.5.0 mdfv3 807 298
asammdf 2.5.0 nodata mdfv3 18500 132
mdfreader 0.2.5 mdfv3 36 454
asammdf 2.5.0 mdfv4 804 349
asammdf 2.5.0 nodata mdfv4 21315 171
mdfreader 0.2.5 mdfv4 49 577
Convert file Time [ms] RAM [MB]
asammdf 2.5.0 v3 to v4 5834 709
asammdf 2.5.0 v3 to v4 nodata 28427 494
asammdf 2.5.0 v4 to v3 5474 710
asammdf 2.5.0 v4 to v3 nodata 30423 638