In [33]:
import numpy as np
import random



liste = ['Felix', 'Sascha', 'gloria', 'Robin', 'Romy','Joel']
np.random.shuffle(liste)
In [34]:
liste
Out[34]:
['Romy', 'Sascha', 'gloria', 'Robin', 'Felix', 'Joel']
In [9]:
import pandas as pd
import matplotlib.pyplot as plt
In [10]:
mem = pd.read_csv("mem.tsv",sep="\t")
In [11]:
mem.head()
Out[11]:
hostname interval timestamp kbmemfree kbavail kbmemused %memused kbbuffers kbcached kbcommit %commit kbactive kbinact kbdirty
0 gaia2 60 2024-03-01 09:45:27 UTC 137037152 230075116 31003076 11.74 636204 89877820 171802324 41.85 20851076 97457904 184
1 gaia2 60 2024-03-01 09:46:27 UTC 138041912 231080832 30002400 11.36 636328 89874808 161479612 39.33 20894252 96386512 576
2 gaia2 60 2024-03-01 09:47:27 UTC 138006124 231045812 30024476 11.37 636420 89888768 162125444 39.49 20894752 96423336 120
3 gaia2 60 2024-03-01 09:48:27 UTC 137958872 230999448 30071096 11.39 636504 89888952 171039472 41.66 20894832 96508840 3272
4 gaia2 60 2024-03-01 09:49:27 UTC 137909188 230950336 30120168 11.41 636592 89888964 171062608 41.67 20894932 96569740 212
In [12]:
cpu = pd.read_csv("cpu.tsv",sep="\t")
cpu.head()
Out[12]:
hostname interval timestamp CPU %user %nice %system %iowait %steal %idle
0 gaia2 60 2024-03-01 09:45:27 UTC -1 1.05 0.0 2.62 0.09 0.0 96.23
1 gaia2 60 2024-03-01 09:46:27 UTC -1 1.33 0.0 2.92 0.16 0.0 95.60
2 gaia2 60 2024-03-01 09:47:27 UTC -1 1.47 0.0 2.98 0.16 0.0 95.38
3 gaia2 60 2024-03-01 09:48:27 UTC -1 1.65 0.0 3.25 0.09 0.0 95.01
4 gaia2 60 2024-03-01 09:49:27 UTC -1 1.05 0.0 2.42 0.06 0.0 96.47
In [13]:
sorted = cpu.sort_values(["%user", "hostname"], ascending = [False, False])
sorted.head()
Out[13]:
hostname interval timestamp CPU %user %nice %system %iowait %steal %idle
305885 uranus2 60 2024-03-01 09:55:14 UTC -1 76.90 0.0 0.08 0.00 0.0 23.02
309344 uranus2 60 2024-03-03 19:34:54 UTC -1 68.65 0.0 0.09 0.08 0.0 31.18
305886 uranus2 60 2024-03-01 09:56:14 UTC -1 65.68 0.0 0.10 0.00 0.0 34.22
305887 uranus2 60 2024-03-01 09:57:14 UTC -1 56.79 0.0 0.09 0.00 0.0 43.12
305888 uranus2 60 2024-03-01 09:58:14 UTC -1 54.70 0.0 0.10 0.00 0.0 45.21
In [14]:
uranus2 =  cpu[cpu["hostname"]=="uranus2"]
uranus2.tail()
Out[14]:
hostname interval timestamp CPU %user %nice %system %iowait %steal %idle
346225 uranus2 60 2024-03-29 10:16:05 UTC -1 0.03 0.0 0.01 0.0 0.0 99.97
346226 uranus2 60 2024-03-29 10:17:05 UTC -1 0.01 0.0 0.01 0.0 0.0 99.98
346227 uranus2 60 2024-03-29 10:18:05 UTC -1 0.01 0.0 0.01 0.0 0.0 99.98
346228 uranus2 60 2024-03-29 10:19:05 UTC -1 0.01 0.0 0.01 0.0 0.0 99.98
346229 uranus2 60 2024-03-29 10:20:05 UTC -1 0.01 0.0 0.00 0.0 0.0 99.98
In [15]:
mean_usage = uranus2["%user"].mean()
print(mean_usage)

0.44869946709629444
In [16]:
list_hostnames = np.unique(cpu["hostname"])
print(list_hostnames)

['gaia2' 'gaia3' 'gaia4' 'gaia5' 'jupiter' 'jupiter4' 'jupiter5' 'saturn2'
 'uranus2']
In [23]:
cpu[cpu["hostname"]=='gaia2']["%user"].mean()
Out[23]:
1.0927676061688063
In [24]:
usage_list=[]
for hostname in list_hostnames:
    
    usage_list.append(cpu[cpu["hostname"]==hostname]["%user"].mean())
In [25]:
print(usage_list)

[1.0927676061688063, 1.2852238355268124, 0.8303516677155444, 4.063075819362634, 0.2556233854520734, 0.06360919229730251, 0.1664052271374727, 0.12612141192801546, 0.44869946709629444]
In [30]:
plt.scatter(list_hostnames, usage_list)
plt.ylabel("Mean usage [%]")
plt.title("Mean usage percentage per host")
Out[30]:
Text(0.5, 1.0, 'Mean usage percentage per host')
Out[30]:
In [0]: