1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
|
########################################################################
#
# License: MIT
# Created: May 4, 2013
# Author: Valentin Haenel - valentin@haenel.co
# Author: Francesc Alted - faltet@gmail.com
#
########################################################################
"""
Small benchmark that compares a plain NumPy array copy against
compression through different compressors in Blosc.
"""
from __future__ import print_function
import numpy as np
import time
import blosc
N = 1e8
clevel = 5
Nexp = np.log10(N)
blosc.print_versions()
print("Creating a large NumPy array with 10**%d int64 elements:" % Nexp)
in_ = np.arange(N, dtype=np.int64) # the trivial linear distribution
#in_ = np.linspace(0, 100, N) # another linear distribution
#in_ = np.random.random_integers(0, 100, N) # random distribution
print(" ", in_)
tic = time.time()
out_ = np.copy(in_)
toc = time.time()
print(" Time for copying array with np.copy(): %.3f s" % (toc-tic,))
out_ = np.empty_like(in_)
tic = time.time()
np.copyto(out_, in_)
toc = time.time()
print(" Time for copying array with np.copyto and empty_like: %.3f s" % (toc-tic,))
# Unlike numpy.zeros, numpy.zeros_like doens't use calloc, but instead uses
# empty_like and explicitely assigns zeros, which is basically like calling
# full like
# Here we benchmark what happens when we allocate memory using calloc
out_ = np.zeros(in_.shape, dtype=in_.dtype)
tic = time.time()
np.copyto(out_, in_)
toc = time.time()
print(" Time for copying array with np.copyto and zeros : %.3f s" % (toc-tic,))
# Cause a page faults before the benchmark
out_ = np.full_like(in_, fill_value=0)
tic = time.time()
np.copyto(out_, in_)
toc = time.time()
print(" Time for copying array with np.copyto and full_like: %.3f s" % (toc-tic,))
out_ = np.full_like(in_, fill_value=0)
tic = time.time()
tic = time.time()
out_[...] = in_
toc = time.time()
print(" Time for copying array with numpy assignment: %.3f s" % (toc-tic,))
print()
for cname in blosc.compressor_list():
print("Using *** %s *** compressor::" % cname)
ctic = time.time()
c = blosc.pack_array(in_, clevel=clevel, shuffle=True, cname=cname)
ctoc = time.time()
dtic = time.time()
out = blosc.unpack_array(c)
dtoc = time.time()
assert((in_ == out).all())
print(" Time for pack_array/unpack_array: %.3f/%.3f s." % \
(ctoc-ctic, dtoc-dtic), end='')
print("\tCompr ratio: %.2f" % (in_.size*in_.dtype.itemsize*1. / len(c)))
ctic = time.time()
c = blosc.compress_ptr(in_.__array_interface__['data'][0],
in_.size, in_.dtype.itemsize,
clevel=clevel, shuffle=True, cname=cname)
ctoc = time.time()
out = np.full(in_.size, fill_value=0, dtype=in_.dtype)
dtic = time.time()
blosc.decompress_ptr(c, out.__array_interface__['data'][0])
dtoc = time.time()
assert((in_ == out).all())
print(" Time for compress_ptr/decompress_ptr: %.3f/%.3f s." % \
(ctoc-ctic, dtoc-dtic), end='')
print("\tCompr ratio: %.2f" % (in_.size*in_.dtype.itemsize*1. / len(c)))
|
