Category Archives: Benchmark

GOOD Benchmark Results for PHP Master 2017-07-20

Results for project PHP master, build date 2017-07-20 19:22:06-07:00

  • commit: a6b529c
  • previous commit: 9c73be8
  • revision date: 2017-07-20 22:30:58+02:00
  • environment: Haswell-EP
    • cpu: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz 2×18 cores, stepping 2, LLC 45 MB
    • mem: 128 GB
    • os: CentOS 7.1
    • kernel: Linux 3.10.0-229.4.2.el7.x86_64

Baseline results were generated using release php-7.0.0, with hash 60fffd2 from 2015-12-01 04:16:47+00:00

benchmark relative std_dev* change since last run change since baseline current rev run with PGO
😐 WordPress 4.2.2 cgi -T10000 0.23% 0.10% 3.86% 9.37%
🙂 Drupal 7.36 cgi -T10000 0.16% 1.84% 3.76% 5.75%
😐 MediaWiki 1.23.9 cgi -T5000 0.13% -0.24% 3.49% 4.44%
😐 bench.php cgi -T100 0.00% -0.03% 45.72% 3.33%
😐 micro_bench.php cgi -T10 0.02% 0.01% 28.59% 2.30%
😐 mandelbrot.php cgi -T100 0.01% -0.05% 42.73% 3.43%

* Relative Standard Deviation (Standard Deviation/Average)

Note: Benchmark results for WordPress, Drupal, MediaWiki are measured in fetches/second while all others are measured in seconds.
More details on measurements methodology at:
https://01.org/lp/documentation/php-environment-setup.

Subject Label Legend:
Attributes are determined based on the performance evolution of the workloads compared to the previous measurement iteration.
NEUTRAL: performance did not change by more than 1% for any workload
GOOD: performance improved by more than 1% for at least one workload and there is no regression greater than 1%
BAD: performance dropped by more than 1% for at least one workload and there is no improvement greater than 1%
UGLY: performance improved by more than 1% for at least one workload and also dropped by more than 1% for at least one workload

Our lab does a nightly source pull and build of the PHP project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly.

Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.

[65 flat] Results for Python (master branch) 2017-07-21

Results for project python/master, build date: 2017-07-21 03:04:30-07:00.

  • commit: 7c5798e
  • previous commit: 5bffcf3
  • revision date: 2017-07-21 03:47:01-04:00
  • environment: Broadwell-EP
    • cpu: Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz 2×22 cores, stepping 1, LLC 55 MB
    • mem: 128 GB
    • os: Ubuntu 16.04.2 LTS
    • kernel: 4.4.0-62-generic x86_64 GNU/Linux

Baseline results were generated using release v3.6.0, with hash 5c4568a from 2016-12-22 23:38:47+00:00.

benchmark relative std_dev* change since last run change since baseline current rev run with PGO
😐 2to3 1.105% +0.093% +4.957% +6.344%
😐 call_method 1.894% -0.340% +24.160% +10.873%
😐 call_method_slots 2.888% +0.685% +24.866% +10.339%
😐 call_method_unknown 1.522% -0.653% +22.398% +8.511%
😐 call_simple 3.543% -0.583% +5.524% +11.245%
😐 chameleon 1.241% +0.370% +11.653% +10.342%
😐 chaos 0.790% +0.633% +8.483% +8.052%
😐 crypto_pyaes 0.577% +0.094% +5.476% +4.885%
😐 deltablue 3.893% +0.673% +6.769% +18.738%
😐 django_template 4.199% -0.021% +9.729% +11.983%
😐 dulwich_log 1.031% +0.085% +4.131% +6.259%
😐 fannkuch 1.258% -0.109% +4.867% +4.810%
😐 float 0.751% +0.176% +3.674% +6.077%
😐 genshi_text 2.399% -0.269% +9.317% +11.284%
😐 genshi_xml 2.556% +0.179% +6.819% +10.960%
😐 go 0.831% -0.138% +7.547% +9.425%
😐 hexiom 0.432% +0.224% +10.735% +9.809%
😐 html5lib 3.663% -0.455% +9.457% +7.392%
😐 json_dumps 1.626% -0.209% +3.209% +10.369%
😐 json_loads 2.432% +0.473% +1.920% +9.153%
😐 logging_format 1.929% -1.455% +7.355% +9.759%
😐 logging_silent 2.152% -0.572% +8.931% +15.836%
😐 logging_simple 1.807% -0.467% +8.314% +10.612%
😐 mako 0.608% +0.079% +18.243% +12.377%
😐 mdp 8.832% -0.160% +3.491% +15.690%
😐 meteor_contest 1.880% -0.388% +4.279% +5.522%
😐 nbody 0.630% +0.026% -2.826% +2.329%
😐 nqueens 1.981% -0.400% +2.147% +7.383%
😐 pathlib 1.013% -0.035% +3.116% +12.571%
😐 pickle 1.272% +0.382% +3.552% +20.613%
😐 pickle_dict 0.628% -0.116% +2.171% +15.788%
😐 pickle_list 0.689% -0.180% +5.981% +16.122%
😐 pickle_pure_python 3.676% +0.030% +12.930% +8.008%
😐 pidigits 0.334% -0.024% +0.306% +10.157%
😐 python_startup 0.127% -0.030% +9.760% +4.053%
😐 python_startup_no_site 0.091% -0.011% +1.432% +4.188%
😐 raytrace 1.616% +0.085% +10.208% +12.160%
😐 regex_compile 5.158% -1.833% -9.166% +10.712%
😐 regex_dna 2.044% -0.777% +1.478% +8.012%
😐 regex_effbot 2.121% +0.866% -1.529% +3.373%
😐 regex_v8 1.603% +0.749% +12.189% +3.739%
😐 richards 2.585% -1.262% +9.147% +12.835%
😐 scimark_fft 0.498% +0.154% +1.157% +1.023%
😐 scimark_lu 2.015% +0.256% +26.865% +8.071%
😐 scimark_monte_carlo 1.833% -0.250% +5.732% +5.773%
😐 scimark_sor 0.903% +0.008% +16.219% +7.314%
😐 scimark_sparse_mat_mult 0.989% +0.535% +1.611% -6.971%
😐 spectral_norm 0.579% -0.030% +6.520% +0.972%
😐 sqlalchemy_declarative 1.180% +0.326% +5.718% +5.507%
😐 sqlalchemy_imperative 3.647% -0.423% +4.243% +4.594%
😐 sqlite_synth 3.389% -0.184% +1.841% +8.720%
😐 sympy_expand 2.127% +0.793% +13.398% +6.350%
😐 sympy_integrate 1.794% -0.182% +10.486% +5.589%
😐 sympy_str 3.576% +0.730% +12.924% +7.178%
😐 sympy_sum 4.767% +1.560% +13.352% +8.221%
😐 telco 3.497% +1.783% +22.741% +8.913%
😐 tornado_http 1.120% +0.033% +7.028% +4.761%
😐 unpack_sequence 1.171% +0.081% +2.072% -0.573%
😐 unpickle 6.167% +0.857% +8.545% +20.817%
😐 unpickle_list 0.997% -0.077% -1.102% +18.153%
😐 unpickle_pure_python 2.624% -0.290% +6.746% +8.120%
😐 xml_etree_generate 0.971% -0.178% +5.831% +7.724%
😐 xml_etree_iterparse 2.686% -0.284% +1.227% +7.212%
😐 xml_etree_parse 2.051% +0.119% -5.723% +10.401%
😐 xml_etree_process 1.241% +0.189% +6.838% +7.272%

* Relative Standard Deviation (Standard Deviation/Average)

Our lab does a nightly source pull and build of the Python project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly.

Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.

BAD Benchmark Results for Node.JS Master 2017-07-20

Results for project Node.js master, build date 2017-07-20 23:56:48-07:00

  • commit: 6a5bdfb
  • previous commit: 65eefa0
  • revision date: 2017-07-20 18:16:37-04:00
  • environment: Haswell-EP
    • cpu: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz 2×18 cores, stepping 2, LLC 45MB
    • mem: 128GB
    • os: Ubuntu Server 14.04 LTS
    • kernel: 3.19.0-25-generic

Baseline results were generated using release v4.2.2, with hash bcf6ac1 from 2015-11-03 19:49:06+00:00

benchmark relative std_dev* change since last run change since baseline change with PGO since baseline
😐 acmeair 0.60% -0.30% 46.14% 46.41%
😐 employeedb 0.61% -0.08% 24.31% 24.03%
😐 buffers/dataview-set.js 0.59% 0.10% 314.29% 314.99%
🙁 word-finder 9.87% -2.89% -2.11% 0.77%
😐 nodejs-mud 2.01% 0.51% 0.41% 1.03%

* Relative Standard Deviation (Standard Deviation/Average)

Subject Label Legend:
Attributes are determined based on the performance evolution of the workloads compared to the previous measurement iteration.
NEUTRAL: performance did not change by more than 1% for any workload
GOOD: performance improved by more than 1% for at least one workload and there is no regression greater than 1%
BAD: performance dropped by more than 1% for at least one workload and there is no improvement greater than 1%
UGLY: performance improved by more than 1% for at least one workload and also dropped by more than 1% for at least one workload

Our lab does a nightly source pull and build of the Node project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly.

Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.

UGLY Benchmark Results for PHP Master 2017-07-19

Results for project PHP master, build date 2017-07-19 19:22:28-07:00

  • commit: 9c73be8
  • previous commit: 21d5636
  • revision date: 2017-07-19 23:59:42+02:00
  • environment: Haswell-EP
    • cpu: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz 2×18 cores, stepping 2, LLC 45 MB
    • mem: 128 GB
    • os: CentOS 7.1
    • kernel: Linux 3.10.0-229.4.2.el7.x86_64

Baseline results were generated using release php-7.0.0, with hash 60fffd2 from 2015-12-01 04:16:47+00:00

benchmark relative std_dev* change since last run change since baseline current rev run with PGO
😐 WordPress 4.2.2 cgi -T10000 0.23% 0.25% 3.75% 9.40%
😐 Drupal 7.36 cgi -T10000 0.22% 0.50% 1.89% 6.37%
😐 MediaWiki 1.23.9 cgi -T5000 0.08% -0.30% 3.74% 3.52%
😐 bench.php cgi -T100 0.01% 0.19% 45.73% -0.01%
🙂 micro_bench.php cgi -T10 0.01% 2.06% 28.58% 2.56%
🙁 mandelbrot.php cgi -T100 0.02% -5.06% 42.76% 4.54%

* Relative Standard Deviation (Standard Deviation/Average)

Note: Benchmark results for WordPress, Drupal, MediaWiki are measured in fetches/second while all others are measured in seconds.
More details on measurements methodology at:
https://01.org/lp/documentation/php-environment-setup.

Subject Label Legend:
Attributes are determined based on the performance evolution of the workloads compared to the previous measurement iteration.
NEUTRAL: performance did not change by more than 1% for any workload
GOOD: performance improved by more than 1% for at least one workload and there is no regression greater than 1%
BAD: performance dropped by more than 1% for at least one workload and there is no improvement greater than 1%
UGLY: performance improved by more than 1% for at least one workload and also dropped by more than 1% for at least one workload

Our lab does a nightly source pull and build of the PHP project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly.

Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.

[65 flat] Results for Python (master branch) 2017-07-20

Results for project python/master, build date: 2017-07-20 03:04:27-07:00.

  • commit: 5bffcf3
  • previous commit: d81bea6
  • revision date: 2017-07-20 03:58:46-04:00
  • environment: Broadwell-EP
    • cpu: Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz 2×22 cores, stepping 1, LLC 55 MB
    • mem: 128 GB
    • os: Ubuntu 16.04.2 LTS
    • kernel: 4.4.0-62-generic x86_64 GNU/Linux

Baseline results were generated using release v3.6.0, with hash 5c4568a from 2016-12-22 23:38:47+00:00.

benchmark relative std_dev* change since last run change since baseline current rev run with PGO
😐 2to3 1.159% +0.305% +4.868% +6.856%
😐 call_method 1.958% -0.252% +24.417% +4.080%
😐 call_method_slots 3.238% -1.009% +24.348% +3.634%
😐 call_method_unknown 1.183% +0.395% +22.902% +0.553%
😐 call_simple 3.364% -0.030% +6.072% +9.548%
😐 chameleon 1.966% +0.070% +11.325% +10.065%
😐 chaos 2.719% +0.386% +7.900% +8.285%
😐 crypto_pyaes 0.474% +0.108% +5.387% +4.275%
😐 deltablue 4.412% -0.553% +6.138% +18.141%
😐 django_template 5.167% +0.972% +9.748% +11.836%
😐 dulwich_log 1.160% +0.256% +4.050% +5.974%
😐 fannkuch 0.803% +0.103% +4.971% +6.294%
😐 float 1.251% -0.036% +3.504% +4.483%
😐 genshi_text 1.448% +0.307% +9.561% +9.871%
😐 genshi_xml 1.687% -0.232% +6.652% +9.343%
😐 go 0.972% +0.205% +7.674% +7.751%
😐 hexiom 0.480% +0.075% +10.534% +10.079%
😐 html5lib 3.610% +0.776% +9.867% +8.056%
😐 json_dumps 1.456% +0.434% +3.411% +9.379%
😐 json_loads 5.346% -0.449% +1.454% +12.337%
😐 logging_format 1.376% +0.978% +8.683% +7.856%
😐 logging_silent 2.071% +0.213% +9.449% +13.508%
😐 logging_simple 1.961% +0.237% +8.740% +9.817%
😐 mako 0.654% -0.254% +18.178% +12.316%
😐 mdp 8.935% -1.378% +3.645% +15.423%
😐 meteor_contest 2.384% -0.199% +4.649% +4.650%
😐 nbody 1.316% -0.067% -2.853% +3.565%
😐 nqueens 0.728% -0.148% +2.537% +7.513%
😐 pathlib 1.402% +0.549% +3.150% +10.591%
😐 pickle 2.517% +0.142% +3.182% +20.995%
😐 pickle_dict 0.343% -0.086% +2.284% +20.655%
😐 pickle_list 0.833% +0.392% +6.150% +18.829%
😐 pickle_pure_python 2.536% +0.173% +12.904% +8.112%
😐 pidigits 0.317% -0.033% +0.329% +10.040%
😐 python_startup 0.121% +0.080% +9.788% +4.112%
😐 python_startup_no_site 0.089% +0.043% +1.444% +3.987%
😐 raytrace 1.407% +0.136% +10.131% +10.343%
😐 regex_compile 5.015% +1.018% -7.201% +10.202%
😐 regex_dna 0.298% +0.308% +2.237% +9.062%
😐 regex_effbot 2.104% +0.041% -2.417% +4.632%
😐 regex_v8 1.443% -0.048% +11.526% +3.911%
😐 richards 1.282% +0.502% +10.279% +10.210%
😐 scimark_fft 1.369% -0.397% +1.004% +2.240%
😐 scimark_lu 3.214% -0.843% +26.677% +8.925%
😐 scimark_monte_carlo 1.509% -0.048% +5.968% +3.319%
😐 scimark_sor 0.644% +0.106% +16.212% +5.857%
😐 scimark_sparse_mat_mult 1.804% -0.170% +1.081% -4.536%
😐 spectral_norm 0.689% -0.033% +6.548% +1.937%
😐 sqlalchemy_declarative 0.997% +0.020% +5.409% +6.930%
😐 sqlalchemy_imperative 3.041% +0.337% +4.647% +4.741%
😐 sqlite_synth 4.078% -0.692% +2.022% +8.717%
😐 sympy_expand 2.383% +0.025% +12.706% +8.152%
😐 sympy_integrate 1.775% +0.366% +10.649% +5.645%
😐 sympy_str 3.409% +0.478% +12.284% +8.982%
😐 sympy_sum 4.967% -0.492% +11.979% +11.226%
😐 telco 5.436% -0.792% +21.339% +11.365%
😐 tornado_http 1.039% +0.100% +6.997% +2.357%
😐 unpack_sequence 1.225% +0.078% +1.993% -1.736%
😐 unpickle 7.163% -1.684% +7.754% +21.907%
😐 unpickle_list 1.006% -0.294% -1.024% +19.277%
😐 unpickle_pure_python 2.618% +0.025% +7.015% +7.375%
😐 xml_etree_generate 0.881% +0.191% +5.998% +7.530%
😐 xml_etree_iterparse 2.582% -0.270% +1.507% +6.443%
😐 xml_etree_parse 1.938% +0.506% -5.849% +10.122%
😐 xml_etree_process 1.395% +0.205% +6.662% +7.456%

* Relative Standard Deviation (Standard Deviation/Average)

Our lab does a nightly source pull and build of the Python project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly.

Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.

BAD Benchmark Results for Node.JS Master 2017-07-19

Results for project Node.js master, build date 2017-07-19 23:56:58-07:00

  • commit: 65eefa0
  • previous commit: a3365c6
  • revision date: 2017-07-19 18:40:42-04:00
  • environment: Haswell-EP
    • cpu: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz 2×18 cores, stepping 2, LLC 45MB
    • mem: 128GB
    • os: Ubuntu Server 14.04 LTS
    • kernel: 3.19.0-25-generic

Baseline results were generated using release v4.2.2, with hash bcf6ac1 from 2015-11-03 19:49:06+00:00

benchmark relative std_dev* change since last run change since baseline change with PGO since baseline
😐 acmeair 0.24% 0.07% 46.59% 46.76%
😐 employeedb 0.50% 0.24% 24.40% 24.20%
😐 buffers/dataview-set.js 0.66% -0.10% 313.88% 315.70%
😐 word-finder 0.40% -0.12% 0.81% 0.78%
🙁 nodejs-mud 2.24% -1.15% -0.11% -0.48%

* Relative Standard Deviation (Standard Deviation/Average)

Subject Label Legend:
Attributes are determined based on the performance evolution of the workloads compared to the previous measurement iteration.
NEUTRAL: performance did not change by more than 1% for any workload
GOOD: performance improved by more than 1% for at least one workload and there is no regression greater than 1%
BAD: performance dropped by more than 1% for at least one workload and there is no improvement greater than 1%
UGLY: performance improved by more than 1% for at least one workload and also dropped by more than 1% for at least one workload

Our lab does a nightly source pull and build of the Node project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly.

Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.

GOOD Benchmark Results for PHP Master 2017-07-18

Results for project PHP master, build date 2017-07-18 19:22:37-07:00

  • commit: 21d5636
  • previous commit: a39fd99
  • revision date: 2017-07-19 01:06:01+02:00
  • environment: Haswell-EP
    • cpu: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz 2×18 cores, stepping 2, LLC 45 MB
    • mem: 128 GB
    • os: CentOS 7.1
    • kernel: Linux 3.10.0-229.4.2.el7.x86_64

Baseline results were generated using release php-7.0.0, with hash 60fffd2 from 2015-12-01 04:16:47+00:00

benchmark relative std_dev* change since last run change since baseline current rev run with PGO
😐 WordPress 4.2.2 cgi -T10000 0.14% -0.45% 3.49% 9.06%
😐 Drupal 7.36 cgi -T10000 0.18% -0.55% 1.39% 5.48%
😐 MediaWiki 1.23.9 cgi -T5000 0.17% 0.18% 4.06% 3.90%
🙂 bench.php cgi -T100 0.02% 1.52% 45.63% 2.58%
🙂 micro_bench.php cgi -T10 0.23% 3.54% 27.08% 4.50%
🙂 mandelbrot.php cgi -T100 0.01% 4.35% 45.51% -0.74%

* Relative Standard Deviation (Standard Deviation/Average)

Note: Benchmark results for WordPress, Drupal, MediaWiki are measured in fetches/second while all others are measured in seconds.
More details on measurements methodology at:
https://01.org/lp/documentation/php-environment-setup.

Subject Label Legend:
Attributes are determined based on the performance evolution of the workloads compared to the previous measurement iteration.
NEUTRAL: performance did not change by more than 1% for any workload
GOOD: performance improved by more than 1% for at least one workload and there is no regression greater than 1%
BAD: performance dropped by more than 1% for at least one workload and there is no improvement greater than 1%
UGLY: performance improved by more than 1% for at least one workload and also dropped by more than 1% for at least one workload

Our lab does a nightly source pull and build of the PHP project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly.

Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.

[1 up, 64 flat] Results for Python (master branch) 2017-07-19

Results for project python/master, build date: 2017-07-19 03:04:28-07:00.

  • commit: d81bea6
  • previous commit: efa26bc
  • revision date: 2017-07-19 05:27:24+03:00
  • environment: Broadwell-EP
    • cpu: Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz 2×22 cores, stepping 1, LLC 55 MB
    • mem: 128 GB
    • os: Ubuntu 16.04.2 LTS
    • kernel: 4.4.0-62-generic x86_64 GNU/Linux

Baseline results were generated using release v3.6.0, with hash 5c4568a from 2016-12-22 23:38:47+00:00.

benchmark relative std_dev* change since last run change since baseline current rev run with PGO
😐 2to3 1.101% -0.071% +4.578% +7.153%
😐 call_method 1.450% +2.074% +24.607% +8.853%
😐 call_method_slots 3.255% +0.745% +25.103% +8.755%
😐 call_method_unknown 1.626% +0.305% +22.596% +8.101%
😐 call_simple 4.230% +2.404% +6.100% +11.979%
😐 chameleon 2.007% -0.692% +11.263% +11.416%
😐 chaos 1.963% +0.559% +7.543% +8.358%
😐 crypto_pyaes 0.572% +1.444% +5.285% +4.672%
😐 deltablue 3.417% -1.074% +6.655% +17.765%
😐 django_template 4.951% -0.998% +8.862% +13.567%
😐 dulwich_log 1.221% +0.181% +3.804% +6.183%
😐 fannkuch 0.968% -1.423% +4.872% +5.825%
😐 float 1.447% -0.793% +3.539% +5.225%
😐 genshi_text 1.343% -0.246% +9.282% +11.037%
😐 genshi_xml 1.726% -1.071% +6.868% +9.632%
😐 go 0.979% +0.375% +7.484% +9.700%
😐 hexiom 0.653% +0.968% +10.467% +9.446%
😐 html5lib 3.396% +1.157% +9.162% +8.197%
😐 json_dumps 3.204% -0.882% +2.990% +9.837%
😐 json_loads 2.218% -2.376% +1.895% +11.646%
😐 logging_format 1.594% +0.237% +7.781% +10.044%
😐 logging_silent 1.781% -3.911% +9.255% +15.524%
😐 logging_simple 1.408% +0.290% +8.524% +11.848%
😐 mako 0.561% +0.634% +18.385% +12.917%
😐 mdp 6.774% -0.350% +4.955% +13.701%
😐 meteor_contest 2.809% +0.759% +4.838% +4.774%
😐 nbody 0.592% +1.840% -2.784% +5.398%
😐 nqueens 0.721% +0.757% +2.680% +6.756%
😐 pathlib 1.348% -0.445% +2.615% +11.185%
😐 pickle 2.799% +1.123% +3.044% +22.213%
😐 pickle_dict 0.323% -0.574% +2.369% +21.142%
😐 pickle_list 1.046% +0.196% +5.781% +19.350%
😐 pickle_pure_python 3.930% +2.951% +12.753% +10.006%
😐 pidigits 0.148% -0.042% +0.363% +9.775%
😐 python_startup 0.129% +0.042% +9.716% +4.131%
😐 python_startup_no_site 0.093% -0.113% +1.401% +4.321%
😐 raytrace 1.237% +0.831% +10.009% +11.659%
😐 regex_compile 4.854% -0.101% -8.303% +11.265%
🙂 regex_dna 1.418% +3.751% +1.935% +7.670%
😐 regex_effbot 2.077% +2.410% -2.458% +7.147%
😐 regex_v8 1.268% +0.965% +11.568% +3.992%
😐 richards 1.491% +0.584% +9.827% +12.318%
😐 scimark_fft 0.410% +0.343% +1.396% +3.809%
😐 scimark_lu 1.439% +0.384% +27.290% +7.700%
😐 scimark_monte_carlo 1.225% +0.382% +6.013% +4.506%
😐 scimark_sor 0.941% +1.087% +16.123% +7.117%
😐 scimark_sparse_mat_mult 1.646% -1.651% +1.249% +0.354%
😐 spectral_norm 0.692% +1.672% +6.578% +1.769%
😐 sqlalchemy_declarative 1.194% +0.112% +5.391% +6.411%
😐 sqlalchemy_imperative 3.385% +1.131% +4.324% +4.581%
😐 sqlite_synth 3.294% +1.172% +2.695% +8.652%
😐 sympy_expand 2.359% +0.314% +12.684% +7.727%
😐 sympy_integrate 1.682% -0.504% +10.321% +6.560%
😐 sympy_str 3.897% +0.171% +11.863% +9.327%
😐 sympy_sum 4.874% -0.262% +12.410% +10.415%
😐 telco 4.918% -0.310% +21.957% +9.688%
😐 tornado_http 1.138% +0.759% +6.904% +5.598%
😐 unpack_sequence 2.173% -0.406% +1.916% -1.407%
😐 unpickle 4.357% -1.189% +9.282% +17.486%
😐 unpickle_list 0.687% +0.189% -0.728% +15.917%
😐 unpickle_pure_python 1.674% -1.231% +6.992% +8.214%
😐 xml_etree_generate 1.197% +0.166% +5.818% +7.962%
😐 xml_etree_iterparse 2.975% -1.894% +1.772% +6.938%
😐 xml_etree_parse 2.396% -0.745% -6.387% +11.747%
😐 xml_etree_process 2.230% -0.109% +6.470% +8.414%

* Relative Standard Deviation (Standard Deviation/Average)

Our lab does a nightly source pull and build of the Python project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly.

Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.

GOOD Benchmark Results for Node.JS Master 2017-07-19

Results for project Node.js master, build date 2017-07-19 02:10:44-07:00

  • commit: a3365c6
  • previous commit: aa1c8c0
  • revision date: 2017-07-18 20:42:42+02:00
  • environment: Haswell-EP
    • cpu: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz 2×18 cores, stepping 2, LLC 45MB
    • mem: 128GB
    • os: Ubuntu Server 14.04 LTS
    • kernel: 3.19.0-25-generic

Baseline results were generated using release v4.2.2, with hash bcf6ac1 from 2015-11-03 19:49:06+00:00

benchmark relative std_dev* change since last run change since baseline change with PGO since baseline
🙂 acmeair 0.58% 14.35% 46.49% 46.42%
😐 employeedb 0.84% -0.11% 24.10% 23.84%
😐 buffers/dataview-set.js 0.68% -0.24% 314.28% 314.75%
😐 word-finder 0.14% 0.02% 0.92% -0.99%
😐 nodejs-mud 1.57% 0.83% 1.06% 1.66%

* Relative Standard Deviation (Standard Deviation/Average)

Subject Label Legend:
Attributes are determined based on the performance evolution of the workloads compared to the previous measurement iteration.
NEUTRAL: performance did not change by more than 1% for any workload
GOOD: performance improved by more than 1% for at least one workload and there is no regression greater than 1%
BAD: performance dropped by more than 1% for at least one workload and there is no improvement greater than 1%
UGLY: performance improved by more than 1% for at least one workload and also dropped by more than 1% for at least one workload

Our lab does a nightly source pull and build of the Node project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly.

Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.

UGLY Benchmark Results for PHP Master 2017-07-17

Results for project PHP master, build date 2017-07-17 19:21:54-07:00

  • commit: a39fd99
  • previous commit: 09d3b73
  • revision date: 2017-07-18 00:35:01+02:00
  • environment: Haswell-EP
    • cpu: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz 2×18 cores, stepping 2, LLC 45 MB
    • mem: 128 GB
    • os: CentOS 7.1
    • kernel: Linux 3.10.0-229.4.2.el7.x86_64

Baseline results were generated using release php-7.0.0, with hash 60fffd2 from 2015-12-01 04:16:47+00:00

benchmark relative std_dev* change since last run change since baseline current rev run with PGO
😐 WordPress 4.2.2 cgi -T10000 0.30% 0.67% 3.96% 8.71%
😐 Drupal 7.36 cgi -T10000 0.19% 0.17% 1.95% 5.11%
😐 MediaWiki 1.23.9 cgi -T5000 0.07% 0.64% 3.87% 3.51%
🙁 bench.php cgi -T100 0.10% -3.76% 44.79% 4.02%
🙂 micro_bench.php cgi -T10 0.01% 3.25% 24.40% 7.46%
😐 mandelbrot.php cgi -T100 0.05% -0.88% 43.04% 3.42%

* Relative Standard Deviation (Standard Deviation/Average)

Note: Benchmark results for WordPress, Drupal, MediaWiki are measured in fetches/second while all others are measured in seconds.
More details on measurements methodology at:
https://01.org/lp/documentation/php-environment-setup.

Subject Label Legend:
Attributes are determined based on the performance evolution of the workloads compared to the previous measurement iteration.
NEUTRAL: performance did not change by more than 1% for any workload
GOOD: performance improved by more than 1% for at least one workload and there is no regression greater than 1%
BAD: performance dropped by more than 1% for at least one workload and there is no improvement greater than 1%
UGLY: performance improved by more than 1% for at least one workload and also dropped by more than 1% for at least one workload

Our lab does a nightly source pull and build of the PHP project and measures performance changes against the previous stable version and the previous nightly measurement. This is provided as a service to the community so that quality issues with current hardware can be identified quickly.

Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.