Skip to content

Instantly share code, notes, and snippets.

@taeguk
Created July 11, 2017 15:44
Show Gist options
  • Save taeguk/36f6fadce38c85096018f4df3a91d552 to your computer and use it in GitHub Desktop.
Save taeguk/36f6fadce38c85096018f4df3a91d552 to your computer and use it in GitHub Desktop.
HPX parallel::unique_copy performance test result with policy.executor() instead of sync in scan_partitioner.
[tkwon@bob1 hpx_build]$ ./bin/benchmark_unique_copy --vector_size 1000000 --hpx:threads 16 --test_count 10 --iterator_tag forward --random_range 1
-------------- Benchmark Config --------------
seed : 1499787616
vector_size : 1000000
random_range : 1
iterator_tag : forward
test_count : 10
os threads : 16
----------------------------------------------
* Preparing Benchmark...
*** Destination iterator distance : 1
* Running Benchmark...
--- run_unique_copy_benchmark_std ---
--- run_unique_copy_benchmark_seq ---
--- run_unique_copy_benchmark_par ---
--- run_unique_copy_benchmark_par_unseq ---
-------------- Benchmark Result --------------
unique_copy (std) : 0.000785199(sec)
unique_copy (seq) : 0.000760707(sec)
unique_copy (par) : 0.00146218(sec)
unique_copy (par_unseq) : 0.000856092(sec)
----------------------------------------------
[tkwon@bob1 hpx_build]$ ./bin/benchmark_unique_copy --vector_size 10000000 --hpx:threads 16 --test_count 10 --iterator_tag forward --random_range 1
-------------- Benchmark Config --------------
seed : 1499787631
vector_size : 10000000
random_range : 1
iterator_tag : forward
test_count : 10
os threads : 16
----------------------------------------------
* Preparing Benchmark...
*** Destination iterator distance : 1
* Running Benchmark...
--- run_unique_copy_benchmark_std ---
--- run_unique_copy_benchmark_seq ---
--- run_unique_copy_benchmark_par ---
--- run_unique_copy_benchmark_par_unseq ---
-------------- Benchmark Result --------------
unique_copy (std) : 0.0084559(sec)
unique_copy (seq) : 0.00795157(sec)
unique_copy (par) : 0.00350438(sec)
unique_copy (par_unseq) : 0.0029189(sec)
----------------------------------------------
[tkwon@bob1 hpx_build]$ ./bin/benchmark_unique_copy --vector_size 100000000 --hpx:threads 16 --test_count 10 --iterator_tag forward --random_range 1
-------------- Benchmark Config --------------
seed : 1499787646
vector_size : 100000000
random_range : 1
iterator_tag : forward
test_count : 10
os threads : 16
----------------------------------------------
* Preparing Benchmark...
*** Destination iterator distance : 1
* Running Benchmark...
--- run_unique_copy_benchmark_std ---
--- run_unique_copy_benchmark_seq ---
--- run_unique_copy_benchmark_par ---
--- run_unique_copy_benchmark_par_unseq ---
-------------- Benchmark Result --------------
unique_copy (std) : 0.0778156(sec)
unique_copy (seq) : 0.0764816(sec)
unique_copy (par) : 0.0241939(sec)
unique_copy (par_unseq) : 0.024247(sec)
----------------------------------------------
[tkwon@bob1 hpx_build]$ ./bin/benchmark_unique_copy --vector_size 100000000 --hpx:threads 16 --test_count 10 --iterator_tag forward --random_range 6
-------------- Benchmark Config --------------
seed : 1499787656
vector_size : 100000000
random_range : 6
iterator_tag : forward
test_count : 10
os threads : 16
----------------------------------------------
* Preparing Benchmark...
*** Destination iterator distance : 83299969
* Running Benchmark...
--- run_unique_copy_benchmark_std ---
--- run_unique_copy_benchmark_seq ---
--- run_unique_copy_benchmark_par ---
--- run_unique_copy_benchmark_par_unseq ---
-------------- Benchmark Result --------------
unique_copy (std) : 0.242744(sec)
unique_copy (seq) : 0.244338(sec)
unique_copy (par) : 0.0515217(sec)
unique_copy (par_unseq) : 0.0518686(sec)
----------------------------------------------
[tkwon@bob1 hpx_build]$ ./bin/benchmark_unique_copy --vector_size 100000000 --hpx:threads 16 --test_count 10 --iterator_tag forward --random_range 10000
-------------- Benchmark Config --------------
seed : 1499787674
vector_size : 100000000
random_range : 10000
iterator_tag : forward
test_count : 10
os threads : 16
----------------------------------------------
* Preparing Benchmark...
*** Destination iterator distance : 99990080
* Running Benchmark...
--- run_unique_copy_benchmark_std ---
--- run_unique_copy_benchmark_seq ---
--- run_unique_copy_benchmark_par ---
--- run_unique_copy_benchmark_par_unseq ---
-------------- Benchmark Result --------------
unique_copy (std) : 0.121086(sec)
unique_copy (seq) : 0.112845(sec)
unique_copy (par) : 0.0545236(sec)
unique_copy (par_unseq) : 0.0550327(sec)
----------------------------------------------
[tkwon@bob1 hpx_build]$ ./bin/benchmark_unique_copy --vector_size 100000000 --hpx:threads 16 --test_count 10 --iterator_tag forward --random_range 100000
-------------- Benchmark Config --------------
seed : 1499787690
vector_size : 100000000
random_range : 100000
iterator_tag : forward
test_count : 10
os threads : 16
----------------------------------------------
* Preparing Benchmark...
*** Destination iterator distance : 99999040
* Running Benchmark...
--- run_unique_copy_benchmark_std ---
--- run_unique_copy_benchmark_seq ---
--- run_unique_copy_benchmark_par ---
--- run_unique_copy_benchmark_par_unseq ---
-------------- Benchmark Result --------------
unique_copy (std) : 0.120402(sec)
unique_copy (seq) : 0.112566(sec)
unique_copy (par) : 0.0546488(sec)
unique_copy (par_unseq) : 0.0549439(sec)
----------------------------------------------
[tkwon@bob1 hpx_build]$ ./bin/benchmark_unique_copy --vector_size 100000000 --hpx:threads 16 --test_count 10 --iterator_tag random --random_range 100000
-------------- Benchmark Config --------------
seed : 1499787708
vector_size : 100000000
random_range : 100000
iterator_tag : random
test_count : 10
os threads : 16
----------------------------------------------
* Preparing Benchmark...
*** Destination iterator distance : 99999360
* Running Benchmark...
--- run_unique_copy_benchmark_std ---
--- run_unique_copy_benchmark_seq ---
--- run_unique_copy_benchmark_par ---
--- run_unique_copy_benchmark_par_unseq ---
-------------- Benchmark Result --------------
unique_copy (std) : 0.121611(sec)
unique_copy (seq) : 0.11046(sec)
unique_copy (par) : 0.0541598(sec)
unique_copy (par_unseq) : 0.0542794(sec)
----------------------------------------------
[tkwon@bob1 hpx_build]$ ./bin/benchmark_unique_copy --vector_size 100000000 --hpx:threads 16 --test_count 10 --iterator_tag random --random_range 10
-------------- Benchmark Config --------------
seed : 1499787722
vector_size : 100000000
random_range : 10
iterator_tag : random
test_count : 10
os threads : 16
----------------------------------------------
* Preparing Benchmark...
*** Destination iterator distance : 90000257
* Running Benchmark...
--- run_unique_copy_benchmark_std ---
--- run_unique_copy_benchmark_seq ---
--- run_unique_copy_benchmark_par ---
--- run_unique_copy_benchmark_par_unseq ---
-------------- Benchmark Result --------------
unique_copy (std) : 0.229246(sec)
unique_copy (seq) : 0.202494(sec)
unique_copy (par) : 0.0525817(sec)
unique_copy (par_unseq) : 0.0523726(sec)
----------------------------------------------
[tkwon@bob1 hpx_build]$ ./bin/benchmark_unique_copy --vector_size 100000000 --hpx:threads 16 --test_count 10 --iterator_tag random --random_range 1
-------------- Benchmark Config --------------
seed : 1499787731
vector_size : 100000000
random_range : 1
iterator_tag : random
test_count : 10
os threads : 16
----------------------------------------------
* Preparing Benchmark...
*** Destination iterator distance : 1
* Running Benchmark...
--- run_unique_copy_benchmark_std ---
--- run_unique_copy_benchmark_seq ---
--- run_unique_copy_benchmark_par ---
--- run_unique_copy_benchmark_par_unseq ---
-------------- Benchmark Result --------------
unique_copy (std) : 0.102806(sec)
unique_copy (seq) : 0.077598(sec)
unique_copy (par) : 0.0245723(sec)
unique_copy (par_unseq) : 0.0241429(sec)
----------------------------------------------
[tkwon@bob1 hpx_build]$ ./bin/benchmark_unique_copy --vector_size 1000000000 --hpx:threads 16 --test_count 10 --iterator_tag random --random_range 1
-------------- Benchmark Config --------------
seed : 1499787749
vector_size : 1000000000
random_range : 1
iterator_tag : random
test_count : 10
os threads : 16
----------------------------------------------
* Preparing Benchmark...
tcmalloc: large alloc 4000006144 bytes == 0x1636000 @
tcmalloc: large alloc 4000006144 bytes == 0xefcea000 @
*** Destination iterator distance : 1
* Running Benchmark...
--- run_unique_copy_benchmark_std ---
--- run_unique_copy_benchmark_seq ---
--- run_unique_copy_benchmark_par ---
--- run_unique_copy_benchmark_par_unseq ---
-------------- Benchmark Result --------------
unique_copy (std) : 0.975272(sec)
unique_copy (seq) : 0.700749(sec)
unique_copy (par) : 0.243507(sec)
unique_copy (par_unseq) : 0.243472(sec)
----------------------------------------------
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment