Skip to content

Instantly share code, notes, and snippets.

@sgillies

sgillies/Benchmark 4 cores vs 2 cpus

Last active January 3, 2016 06:09
Show Gist options
  • Save sgillies/8420859 to your computer and use it in GitHub Desktop.
Save sgillies/8420859 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
Benchmark 4 cores vs 2 cpus
MapBox-FC:multi sean$ time parallel --progress go run test.go ::: {1..20}
Computers / CPU cores / Max jobs to run
1:local / 4 / 4
Computer:jobs running/jobs completed/%of started jobs/Average seconds to complete
local:0/20/100%/0.1s
real 0m0.733s
user 0m1.559s
sys 0m0.577s
MapBox-FC:multi sean$ time parallel --progress --use-cpus-instead-of-cores go run test.go ::: {1..20}
Computers / CPU cores / Max jobs to run
1:local / 2 / 2
Computer:jobs running/jobs completed/%of started jobs/Average seconds to complete
local:0/20/100%/0.1s
real 0m0.968s
user 0m1.145s
sys 0m0.463s
Raw
GNU parallel + Go test console
MapBox-FC:multi sean$ time parallel --progress -j 1 go run test.go ::: {1..20}
Computers / CPU cores / Max jobs to run
1:local / 4 / 1
Computer:jobs running/jobs completed/%of started jobs/Average seconds to complete
local:0/20/100%/0.1s
real 0m1.662s
user 0m1.094s
sys 0m0.447s
MapBox-FC:multi sean$ time parallel --progress -j 4 go run test.go ::: {1..20}
Computers / CPU cores / Max jobs to run
1:local / 4 / 4
Computer:jobs running/jobs completed/%of started jobs/Average seconds to complete
local:0/20/100%/0.0s
real 0m0.737s
user 0m1.551s
sys 0m0.591s
Raw
GNU parallel + Python test console
MapBox-FC:multi sean$ time parallel --progress -j 1 python test.py ::: {1..20}
Computers / CPU cores / Max jobs to run
1:local / 4 / 1
Computer:jobs running/jobs completed/%of started jobs/Average seconds to complete
local:0/20/100%/0.5s
real 0m9.872s
user 0m8.994s
sys 0m0.232s
MapBox-FC:multi sean$ time parallel --progress -j 4 python test.py ::: {1..20}
Computers / CPU cores / Max jobs to run
1:local / 4 / 4
Computer:jobs running/jobs completed/%of started jobs/Average seconds to complete
local:0/20/100%/0.2s
real 0m4.495s
user 0m16.083s
sys 0m0.297s
Raw
multi.py
from multiprocessing import Pool
def countdown(n):
while n > 0:
n -= 1
COUNT = 10000000 # 10 million
if __name__ == '__main__':
pool = Pool() # multiprocessing.cpu_count() = 4
result = pool.map(countdown, 20*[COUNT])
Raw
Python multiprocessing console
MapBox-FC:multi sean$ time python multi.py
real 0m4.361s
user 0m14.655s
sys 0m0.047s
Raw
test.go
package main
func countdown(n int) {
for n > 0 {
n -= 1
}
}
func main() {
COUNT := 10000000
countdown(COUNT)
}
Raw
test.py
def countdown(n):
while n > 0:
n -= 1
COUNT = 10000000 # 10 million
countdown(COUNT)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment