◐ Shell
reader mode source ↗
Skip to content

bpo-40255: Implement Immortal Instances - Optimizations Combined#31491

Closed
eduardo-elizondo wants to merge 1 commit into
python:mainfrom
eduardo-elizondo:immortal-references-opt-combined
Closed

bpo-40255: Implement Immortal Instances - Optimizations Combined#31491
eduardo-elizondo wants to merge 1 commit into
python:mainfrom
eduardo-elizondo:immortal-references-opt-combined

Conversation

@eduardo-elizondo

@eduardo-elizondo eduardo-elizondo commented Feb 22, 2022 •
edited by encukou
Loading

Copy link
Copy Markdown
Contributor

This is an optimization on top of PR19474.

It combines PR31488, PR31489, and PR31490 into a single change to measure the combined performance benefits.

These results do not change too much from what was already achieved independently by these optimizations (as some of the immortalized instances start overlaping with each other). That being said, performance will keep scaling as the application scales as well. The current microbenchmarks do not measure applications that contain hundreds of imports or thousands of interned strings. Nonetheless, it is still worthwhile to consider all of these improvements in conjunction when thinking about larger scale applications.

Benchmark Results

Overall: 0% faster compared to the main branch and the highest number of non-stat significant benchmarks (18)

pyperformance results 
2to3: Mean +- std dev: [cpython_master] 432 ms +- 15 ms -> [immortal_instances_opt_combined_v3] 451 ms +- 16 ms: 1.04x slower
chaos: Mean +- std dev: [cpython_master] 126 ms +- 4 ms -> [immortal_instances_opt_combined_v3] 123 ms +- 4 ms: 1.03x faster
deltablue: Mean +- std dev: [cpython_master] 7.35 ms +- 0.20 ms -> [immortal_instances_opt_combined_v3] 7.74 ms +- 0.41 ms: 1.05x slower
django_template: Mean +- std dev: [cpython_master] 62.2 ms +- 2.0 ms -> [immortal_instances_opt_combined_v3] 63.7 ms +- 2.4 ms: 1.02x slower
fannkuch: Mean +- std dev: [cpython_master] 664 ms +- 15 ms -> [immortal_instances_opt_combined_v3] 677 ms +- 18 ms: 1.02x slower
float: Mean +- std dev: [cpython_master] 128 ms +- 4 ms -> [immortal_instances_opt_combined_v3] 135 ms +- 7 ms: 1.05x slower
go: Mean +- std dev: [cpython_master] 244 ms +- 10 ms -> [immortal_instances_opt_combined_v3] 228 ms +- 14 ms: 1.07x faster
json_dumps: Mean +- std dev: [cpython_master] 19.2 ms +- 0.7 ms -> [immortal_instances_opt_combined_v3] 20.1 ms +- 0.8 ms: 1.04x slower
logging_format: Mean +- std dev: [cpython_master] 10.4 us +- 0.3 us -> [immortal_instances_opt_combined_v3] 11.0 us +- 0.4 us: 1.06x slower
logging_silent: Mean +- std dev: [cpython_master] 201 ns +- 8 ns -> [immortal_instances_opt_combined_v3] 205 ns +- 7 ns: 1.02x slower
logging_simple: Mean +- std dev: [cpython_master] 9.77 us +- 0.32 us -> [immortal_instances_opt_combined_v3] 10.2 us +- 0.4 us: 1.04x slower
nqueens: Mean +- std dev: [cpython_master] 159 ms +- 5 ms -> [immortal_instances_opt_combined_v3] 154 ms +- 3 ms: 1.03x faster
pickle: Mean +- std dev: [cpython_master] 16.0 us +- 0.5 us -> [immortal_instances_opt_combined_v3] 16.6 us +- 0.7 us: 1.04x slower
pickle_dict: Mean +- std dev: [cpython_master] 37.3 us +- 0.6 us -> [immortal_instances_opt_combined_v3] 35.6 us +- 2.1 us: 1.05x faster
pidigits: Mean +- std dev: [cpython_master] 284 ms +- 15 ms -> [immortal_instances_opt_combined_v3] 273 ms +- 9 ms: 1.04x faster
pyflate: Mean +- std dev: [cpython_master] 770 ms +- 28 ms -> [immortal_instances_opt_combined_v3] 746 ms +- 24 ms: 1.03x faster
python_startup: Mean +- std dev: [cpython_master] 12.6 ms +- 0.4 ms -> [immortal_instances_opt_combined_v3] 11.6 ms +- 0.6 ms: 1.08x faster
python_startup_no_site: Mean +- std dev: [cpython_master] 8.89 ms +- 0.39 ms -> [immortal_instances_opt_combined_v3] 8.11 ms +- 0.43 ms: 1.10x faster
raytrace: Mean +- std dev: [cpython_master] 529 ms +- 16 ms -> [immortal_instances_opt_combined_v3] 544 ms +- 17 ms: 1.03x slower
scimark_fft: Mean +- std dev: [cpython_master] 571 ms +- 12 ms -> [immortal_instances_opt_combined_v3] 589 ms +- 13 ms: 1.03x slower
scimark_lu: Mean +- std dev: [cpython_master] 195 ms +- 6 ms -> [immortal_instances_opt_combined_v3] 205 ms +- 7 ms: 1.05x slower
scimark_sor: Mean +- std dev: [cpython_master] 211 ms +- 6 ms -> [immortal_instances_opt_combined_v3] 216 ms +- 6 ms: 1.03x slower
scimark_sparse_mat_mult: Mean +- std dev: [cpython_master] 8.28 ms +- 0.40 ms -> [immortal_instances_opt_combined_v3] 8.66 ms +- 0.30 ms: 1.05x slower
sympy_expand: Mean +- std dev: [cpython_master] 878 ms +- 34 ms -> [immortal_instances_opt_combined_v3] 850 ms +- 27 ms: 1.03x faster
sympy_integrate: Mean +- std dev: [cpython_master] 35.2 ms +- 1.0 ms -> [immortal_instances_opt_combined_v3] 36.6 ms +- 1.5 ms: 1.04x slower
sympy_sum: Mean +- std dev: [cpython_master] 291 ms +- 13 ms -> [immortal_instances_opt_combined_v3] 309 ms +- 7 ms: 1.06x slower
sympy_str: Mean +- std dev: [cpython_master] 514 ms +- 11 ms -> [immortal_instances_opt_combined_v3] 535 ms +- 13 ms: 1.04x slower
telco: Mean +- std dev: [cpython_master] 10.4 ms +- 0.2 ms -> [immortal_instances_opt_combined_v3] 10.7 ms +- 0.5 ms: 1.03x slower
unpack_sequence: Mean +- std dev: [cpython_master] 77.9 ns +- 2.3 ns -> [immortal_instances_opt_combined_v3] 70.6 ns +- 2.0 ns: 1.10x faster
unpickle_list: Mean +- std dev: [cpython_master] 6.77 us +- 0.25 us -> [immortal_instances_opt_combined_v3] 6.95 us +- 0.22 us: 1.03x slower
xml_etree_parse: Mean +- std dev: [cpython_master] 245 ms +- 6 ms -> [immortal_instances_opt_combined_v3] 238 ms +- 5 ms: 1.03x faster
xml_etree_generate: Mean +- std dev: [cpython_master] 146 ms +- 4 ms -> [immortal_instances_opt_combined_v3] 143 ms +- 6 ms: 1.02x faster
xml_etree_process: Mean +- std dev: [cpython_master] 102 ms +- 2 ms -> [immortal_instances_opt_combined_v3] 100 ms +- 3 ms: 1.02x faster

Benchmark hidden because not significant (18): hexiom, html5lib, json_loads, meteor_contest, nbody, pathlib, pickle_list, pickle_pure_python, regex_compile, regex_dna, regex_effbot, regex_v8, richards, scimark_monte_carlo, spectral_norm, unpickle, unpickle_pure_python, xml_etree_iterparse

Geometric mean: 1.00x faster

https://bugs.python.org/issue40255

@corona10 corona10 requested a review from vstinner February 22, 2022 05:55
@eduardo-elizondo eduardo-elizondo mentioned this pull request Feb 22, 2022
Merged
@kumaraditya303

kumaraditya303 commented Feb 22, 2022

Copy link
Copy Markdown
Contributor

You can use this in deepfrozen modules to get this even faster see

cpython/Tools/scripts/deepfreeze.py

Line 140 in 74127b8

self.write(f".ob_refcnt = 999999999,")

@ericsnowcurrently

ericsnowcurrently commented Feb 22, 2022

Copy link
Copy Markdown
Member

Overall: 0% faster compared to the main branch

That's great news! I'm going to update PEP 683 with the outcome and some of the details.

@brettcannon brettcannon removed their request for review February 23, 2022 21:49
gvanrossum
gvanrossum reviewed Mar 1, 2022
View reviewed changes

@gvanrossum gvanrossum left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hide comment

Nice work! Here are some random review comments. Hopefully they're helpful. I decided to only review the final PR (with all optimizations). I skipped the .py files and a few others for now.

1 hidden conversation Load more…
@eduardo-elizondo eduardo-elizondo mannequin mentioned this pull request Jun 6, 2022
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@methane methane methane left review comments

@gvanrossum gvanrossum gvanrossum left review comments

@pablogsal pablogsal Awaiting requested review from pablogsal pablogsal is a code owner

@encukou encukou Awaiting requested review from encukou

@ericsnowcurrently ericsnowcurrently Awaiting requested review from ericsnowcurrently ericsnowcurrently is a code owner

@ncoghlan ncoghlan Awaiting requested review from ncoghlan ncoghlan is a code owner

@warsaw warsaw Awaiting requested review from warsaw warsaw is a code owner

@vsajip vsajip Awaiting requested review from vsajip vsajip is a code owner

@vstinner vstinner Awaiting requested review from vstinner

Assignees

No one assigned

Labels

awaiting review

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

8 participants

@eduardo-elizondo @kumaraditya303 @ericsnowcurrently @methane @gvanrossum @the-knights-who-say-ni @ezio-melotti @bedevere-bot