search

lambdaclass / cairo-vm

cairo-vm is a Rust implementation of the Cairo VM. Cairo (CPU Algebraic Intermediate Representation) is a programming language for writing provable programs, where one party can prove to another that a certain computation was executed correctly without the need for this party to re-execute the same program.
https://lambdaclass.github.io/cairo-vm
Apache License 2.0
485 stars 133 forks source link

Implement `CairoPie::read_zip_file` #1729

Closed fmoletta closed 2 months ago

fmoletta commented 2 months ago

Implements custom deserialization for structs using custom serialization. Implement method to deserialize a CairoPie from a zip file Move custom serialization functions into sub-modules along with their added deserialization counterparts to make use of the serde(with) attribute and make code cleaner

github-actions[bot] commented 2 months ago 
**Hyper Thereading Benchmark results**

hyperfine -r 2 -n "hyper_threading_main threads: 1" 'RAYON_NUM_THREADS=1 ./hyper_threading_main' -n "hyper_threading_pr threads: 1" 'RAYON_NUM_THREADS=1 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 1
  Time (mean ± σ):     23.767 s ±  0.001 s    [User: 23.054 s, System: 0.712 s]
  Range (min … max):   23.766 s … 23.768 s    2 runs

Benchmark 2: hyper_threading_pr threads: 1
  Time (mean ± σ):     23.613 s ±  0.023 s    [User: 22.903 s, System: 0.709 s]
  Range (min … max):   23.597 s … 23.629 s    2 runs

Summary
  'hyper_threading_pr threads: 1' ran
    1.01 ± 0.00 times faster than 'hyper_threading_main threads: 1'

hyperfine -r 2 -n "hyper_threading_main threads: 2" 'RAYON_NUM_THREADS=2 ./hyper_threading_main' -n "hyper_threading_pr threads: 2" 'RAYON_NUM_THREADS=2 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 2
  Time (mean ± σ):     12.396 s ±  0.005 s    [User: 23.600 s, System: 0.652 s]
  Range (min … max):   12.393 s … 12.400 s    2 runs

Benchmark 2: hyper_threading_pr threads: 2
  Time (mean ± σ):     12.286 s ±  0.022 s    [User: 23.188 s, System: 0.650 s]
  Range (min … max):   12.270 s … 12.302 s    2 runs

Summary
  'hyper_threading_pr threads: 2' ran
    1.01 ± 0.00 times faster than 'hyper_threading_main threads: 2'

hyperfine -r 2 -n "hyper_threading_main threads: 4" 'RAYON_NUM_THREADS=4 ./hyper_threading_main' -n "hyper_threading_pr threads: 4" 'RAYON_NUM_THREADS=4 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 4
  Time (mean ± σ):     10.099 s ±  0.279 s    [User: 34.968 s, System: 0.909 s]
  Range (min … max):    9.902 s … 10.297 s    2 runs

Benchmark 2: hyper_threading_pr threads: 4
  Time (mean ± σ):      9.691 s ±  0.168 s    [User: 35.073 s, System: 0.910 s]
  Range (min … max):    9.572 s …  9.809 s    2 runs

Summary
  'hyper_threading_pr threads: 4' ran
    1.04 ± 0.03 times faster than 'hyper_threading_main threads: 4'

hyperfine -r 2 -n "hyper_threading_main threads: 6" 'RAYON_NUM_THREADS=6 ./hyper_threading_main' -n "hyper_threading_pr threads: 6" 'RAYON_NUM_THREADS=6 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 6
  Time (mean ± σ):      9.725 s ±  0.305 s    [User: 34.944 s, System: 0.945 s]
  Range (min … max):    9.509 s …  9.940 s    2 runs

Benchmark 2: hyper_threading_pr threads: 6
  Time (mean ± σ):      9.847 s ±  0.108 s    [User: 35.200 s, System: 0.984 s]
  Range (min … max):    9.771 s …  9.924 s    2 runs

Summary
  'hyper_threading_main threads: 6' ran
    1.01 ± 0.03 times faster than 'hyper_threading_pr threads: 6'

hyperfine -r 2 -n "hyper_threading_main threads: 8" 'RAYON_NUM_THREADS=8 ./hyper_threading_main' -n "hyper_threading_pr threads: 8" 'RAYON_NUM_THREADS=8 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 8
  Time (mean ± σ):      9.492 s ±  0.129 s    [User: 35.301 s, System: 0.924 s]
  Range (min … max):    9.400 s …  9.583 s    2 runs

Benchmark 2: hyper_threading_pr threads: 8
  Time (mean ± σ):      9.792 s ±  0.216 s    [User: 35.314 s, System: 0.933 s]
  Range (min … max):    9.639 s …  9.944 s    2 runs

Summary
  'hyper_threading_main threads: 8' ran
    1.03 ± 0.03 times faster than 'hyper_threading_pr threads: 8'

hyperfine -r 2 -n "hyper_threading_main threads: 16" 'RAYON_NUM_THREADS=16 ./hyper_threading_main' -n "hyper_threading_pr threads: 16" 'RAYON_NUM_THREADS=16 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 16
  Time (mean ± σ):      9.661 s ±  0.189 s    [User: 35.211 s, System: 1.055 s]
  Range (min … max):    9.527 s …  9.795 s    2 runs

Benchmark 2: hyper_threading_pr threads: 16
  Time (mean ± σ):      9.760 s ±  0.055 s    [User: 35.719 s, System: 1.048 s]
  Range (min … max):    9.721 s …  9.799 s    2 runs

Summary
  'hyper_threading_main threads: 16' ran
    1.01 ± 0.02 times faster than 'hyper_threading_pr threads: 16'
github-actions[bot] commented 2 months ago

Benchmark Results for unmodified programs :rocket:

Command Mean [s] Min [s] Max [s] Relative
base big_factorial 1.921 ± 0.023 1.895 1.957 1.02 ± 0.01
head big_factorial 1.882 ± 0.011 1.868 1.895 1.00
Command Mean [s] Min [s] Max [s] Relative
base big_fibonacci 1.841 ± 0.017 1.820 1.869 1.00
head big_fibonacci 1.843 ± 0.043 1.822 1.963 1.00 ± 0.03
Command Mean [s] Min [s] Max [s] Relative
base blake2s_integration_benchmark 6.391 ± 0.085 6.256 6.545 1.00
head blake2s_integration_benchmark 6.411 ± 0.074 6.332 6.569 1.00 ± 0.02
Command Mean [s] Min [s] Max [s] Relative
base compare_arrays_200000 1.919 ± 0.021 1.895 1.946 1.00 ± 0.02
head compare_arrays_200000 1.918 ± 0.034 1.897 2.006 1.00
Command Mean [s] Min [s] Max [s] Relative
base dict_integration_benchmark 1.284 ± 0.009 1.270 1.295 1.00
head dict_integration_benchmark 1.289 ± 0.024 1.274 1.356 1.00 ± 0.02
Command Mean [s] Min [s] Max [s] Relative
base field_arithmetic_get_square_benchmark 1.158 ± 0.020 1.147 1.209 1.00
head field_arithmetic_get_square_benchmark 1.164 ± 0.019 1.148 1.213 1.01 ± 0.02
Command Mean [s] Min [s] Max [s] Relative
base integration_builtins 6.491 ± 0.084 6.384 6.627 1.00
head integration_builtins 6.502 ± 0.143 6.426 6.894 1.00 ± 0.03
Command Mean [s] Min [s] Max [s] Relative
base keccak_integration_benchmark 6.557 ± 0.061 6.501 6.669 1.00
head keccak_integration_benchmark 6.582 ± 0.033 6.551 6.643 1.00 ± 0.01
Command Mean [s] Min [s] Max [s] Relative
base linear_search 1.953 ± 0.028 1.910 1.996 1.02 ± 0.02
head linear_search 1.908 ± 0.009 1.898 1.919 1.00
Command Mean [s] Min [s] Max [s] Relative
base math_cmp_and_pow_integration_benchmark 1.676 ± 0.042 1.650 1.790 1.01 ± 0.03
head math_cmp_and_pow_integration_benchmark 1.660 ± 0.004 1.654 1.667 1.00
Command Mean [s] Min [s] Max [s] Relative
base math_integration_benchmark 1.459 ± 0.018 1.448 1.507 1.00 ± 0.02
head math_integration_benchmark 1.454 ± 0.021 1.441 1.511 1.00
Command Mean [s] Min [s] Max [s] Relative
base memory_integration_benchmark 1.117 ± 0.004 1.112 1.123 1.00
head memory_integration_benchmark 1.140 ± 0.041 1.118 1.247 1.02 ± 0.04
Command Mean [s] Min [s] Max [s] Relative
base operations_with_data_structures_benchmarks 1.751 ± 0.017 1.727 1.778 1.01 ± 0.01
head operations_with_data_structures_benchmarks 1.740 ± 0.013 1.728 1.769 1.00
Command Mean [ms] Min [ms] Max [ms] Relative
base pedersen 497.7 ± 2.9 494.2 502.6 1.00
head pedersen 502.3 ± 2.3 500.5 507.9 1.01 ± 0.01
Command Mean [ms] Min [ms] Max [ms] Relative
base poseidon_integration_benchmark 936.2 ± 4.2 933.1 946.8 1.01 ± 0.01
head poseidon_integration_benchmark 930.3 ± 6.9 925.9 949.6 1.00
Command Mean [s] Min [s] Max [s] Relative
base secp_integration_benchmark 1.728 ± 0.008 1.721 1.744 1.00
head secp_integration_benchmark 1.739 ± 0.023 1.724 1.804 1.01 ± 0.01
Command Mean [ms] Min [ms] Max [ms] Relative
base set_integration_benchmark 680.8 ± 2.9 677.2 685.1 1.02 ± 0.01
head set_integration_benchmark 669.1 ± 6.2 662.8 680.1 1.00
Command Mean [s] Min [s] Max [s] Relative
base uint256_integration_benchmark 3.767 ± 0.101 3.672 4.039 1.01 ± 0.03
head uint256_integration_benchmark 3.730 ± 0.063 3.688 3.884 1.00
codecov[bot] commented 2 months ago

Codecov Report

Attention: Patch coverage is 97.60479% with 4 lines in your changes are missing coverage. Please review.

Project coverage is 94.89%. Comparing base (1a46c9f) to head (2fc0ba2).

Files Patch % Lines
vm/src/vm/runners/cairo_pie.rs 97.60% 4 Missing :warning:
Additional details and impacted files ```diff @@ Coverage Diff @@ ## main #1729 +/- ## ========================================== + Coverage 94.78% 94.89% +0.10% ========================================== Files 100 100 Lines 38322 38425 +103 ========================================== + Hits 36325 36464 +139 + Misses 1997 1961 -36 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

Oppen commented 2 months ago

There seems to be a significant performance regression.

fmoletta commented 2 months ago

There seems to be a significant performance regression.

This is quite strange, the benchmarks don't produce or consume Cairo PIEs so they shouldn't be affected by this PR

Oppen commented 2 months ago

This is quite strange, the benchmarks don't produce or consume Cairo PIEs so they shouldn't be affected by this PR

Whatever it was it seems to have been reversed by new commits. Maybe it was throttling in the worker.